From fitsbits-request Fri Jul 7 05:56:17 1995 X-VM-Message-Order: (1 2 3 4 6 7 5 8 9 10 11 15 14 36 32 38 18 19 20 47 13 41 27 30 48 16 35 34 28 12 44 46 17 45 33 22 21 25 39 26 23 31 49 29 42 43 51 40 50 37 53 55 57 58 60 59 56 66 67 52 54 65 24 68 71 61 70 69 72 62 64 63) X-VM-Summary-Format: "%n %*%a %-17.17F %-3.3m %2d %4l/%-5c %I\"%s\"\n" X-VM-Labels: nil X-VM-VHeader: ("Resent-" "From:" "Sender:" "To:" "Apparently-To:" "Cc:" "Subject:" "Date:") nil X-VM-Bookmark: 54 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["594" "" "6" "July" "1995" "22:29:22" "GMT" "Andrew J. P. Maclean" "a.maclean at uws.edu.au" "<3tho42$45g at ob1.uws.EDU.AU>" "23" "WIndows FITS Table reader" "^From:" nil nil "7" "1995070622:29:22" "WIndows FITS Table reader" (number " " mark " Andrew J. P. Macl Jul 6 23/594 " thread-indent "\"WIndows FITS Table reader\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA02347; Fri, 7 Jul 1995 05:56:17 -0400 Return-Path: Message-Id: <3tho42$45g at ob1.uws.EDU.AU> Organization: University of Western Sydney Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!maui.cc.odu.edu!kuroshio.ccpo.odu.edu!xanth.cs.odu.edu!lll-winken.llnl.gov!simtel!harbinger.cc.monash.edu.au!news.cs.su.oz.au!metro!ob1.uws.edu.au!news Newsgroups: sci.astro.fits content-length: 592 From: "Andrew J. P. Maclean" Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: WIndows FITS Table reader Date: 6 Jul 1995 22:29:22 GMT Does anyone know if there is a FITS Table reader that runs in Widows 3.1 or Win95 available? -- +----------------------------------+--------------------------- ----+ | Andrew J. P. Maclean | | | University Planner | | | University of Western Sydney | Ph: +61 (02) 678 7848 | | PO Box 1000, St Marys, NSW 2760 | Fax: +61 (02) 678 7804 | | Australia | e-mail: a.maclean at uws.edu.au | +----------------------------------+--------------------------- ----+ From fitsbits-request Fri Jul 14 14:24:33 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["628" "Fri" "14" "July" "1995" "14:24:15" "-0400" "William Pence" "pence at tetra.gsfc.nasa.gov" "<199507141824.OAA26730 at tetra.gsfc.nasa.gov>" "20" "re: FITSIO V4.0" "^From:" nil nil "7" "1995071418:24:15" "FITSIO V4.0" (number " " mark " William Pence Jul 14 20/628 " thread-indent "\"re: FITSIO V4.0\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA14330; Fri, 14 Jul 1995 14:24:33 -0400 Return-Path: Message-Id: <199507141824.OAA26730 at tetra.gsfc.nasa.gov> content-length: 626 From: William Pence Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Cc: pence at tetra.gsfc.nasa.gov Subject: re: FITSIO V4.0 Date: Fri, 14 Jul 1995 14:24:15 -0400 A couple bugs have been reported and fixed in the version 4.0 of the FITSIO software that was announced here on 26 June. One of the problems is potentially serious and could affect programs which have more than 1 FITS file open at one time. Anyone using versions 4.01 thru 4.03 of FITSIO should replace it with the latest V4.04 version. As before, this can be obtained on the WWW at URL: http://heasarc.gsfc.nasa.gov/docs/software/fitsio/fitsio.html or via anonymous ftp from: legacy.gsfc.nasa.gov in the /software/fitsio directory. Sorry for the inconvenience. William Pence pence at tetra.gsfc.nasa.gov. From fitsbits-request Sun Jul 30 15:36:00 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1167" "" "28" "July" "1995" "17:26:25" "GMT" "Mike Fitzgibbon" "fitz at argus.lpl.arizona.edu" "<3vb6k1$icm at news.ccit.arizona.edu>" "38" "Re: FITSVIEW Header Problem" "^From:" nil nil "7" "1995072817:26:25" "FITSVIEW Header Problem" (number " " mark " Mike Fitzgibbon Jul 28 38/1167 " thread-indent "\"Re: FITSVIEW Header Problem\"\n") "<3v9s7a$p3f at crl9.crl.com>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA16968; Sun, 30 Jul 1995 15:36:00 -0400 Return-Path: Message-Id: <3vb6k1$icm at news.ccit.arizona.edu> Organization: University of Arizona, LPL Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!maui.cc.odu.edu!kuroshio.ccpo.odu.edu!xanth.cs.odu.edu!concert!news-server.ncren.net!taco.cc.ncsu.edu!gatech!howland.reston.ans.net!news.cac.psu.edu!news.tc.cornell.edu!newsserver.sdsc.edu!nic-nac.CSU.net!news.Cerritos.edu!news.Arizona.EDU!usenet References: <3v9s7a$p3f at crl9.crl.com> Newsgroups: sci.astro.fits content-length: 1165 From: fitz at argus.lpl.arizona.edu (Mike Fitzgibbon) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: FITSVIEW Header Problem Date: 28 Jul 1995 17:26:25 GMT In article <3v9s7a$p3f at crl9.crl.com>, atherton at crl.com says... ... > > "FITShead::valid_check:Invalid FITS (no SIMPLE = T)" > ... but ... > >SIMPLE = T >BITPIX = 16 >NAXIS = 3 >NAXIS1 = 384 >NAXIS2 = 576 >NAXIS3 = 1 >BZERO = 0.000000E+000 >BSCALE = 1.000000E+000 >ORIGIN = UC Santa Barbara Remote Access Telescope >TELESCOP= Celestron C-14 w/ Photometrics 576x384 CCD >DATE-OBS= 20/02/95 >TIME-OBS= 20:38 >RA = .9616 >DEC = 81.61 >Comment = Exposure := 5sec >COMMENT = Filter Wheel #1: EMPTY >COMMENT = Filter Wheel #2: EMPTY >END ... Well the problem seems to be that "buggy" software is generating these FITS files. According to the format definition, (see Astron. Astrophys. Suppl. Ser. 44, (1981) pp363-370, "FITS: A Flexible Image Transport System" by D.C. Wells, E.W. Greisen, and R.H. Harten), a logical variable (e.g. SIMPLE) has a value of T or F in column 30. This header put the logical value in column 11. Also of note is that integer and real values should be right justified in columns 11-30 and character variables should have a single quote in column 11 with the ending quote no earlier that column 20. From fitsbits-request Mon Jul 31 07:54:52 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1526" "" "29" "July" "1995" "20:18:29" "-0700" "Dave Atherton" "atherton at crl.com" "<3vetm5$h9e at crl13.crl.com>" "29" "Re: FITSVIEW Header Problem" "^From:" nil nil "7" "1995073003:18:29" "FITSVIEW Header Problem" (number " " mark " Dave Atherton Jul 29 29/1526 " thread-indent "\"Re: FITSVIEW Header Problem\"\n") "<3vb6k1$icm at news.ccit.arizona.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA18428; Mon, 31 Jul 1995 07:54:52 -0400 Return-Path: Message-Id: <3vetm5$h9e at crl13.crl.com> Organization: CRL Dialup Internet Access (415) 705-6060 [Login: guest] Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!maui.cc.odu.edu!kuroshio.ccpo.odu.edu!xanth.cs.odu.edu!lll-winken.llnl.gov!uwm.edu!hookup!news.mathworks.com!uunet!in1.uu.net!nntp.crl.com!crl13.crl.com!not-for-mail References: <3v9s7a$p3f at crl9.crl.com> <3vb6k1$icm at news.ccit.arizona.edu> Newsgroups: sci.astro.fits content-length: 1524 From: atherton at crl.com (Dave Atherton) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: FITSVIEW Header Problem Date: 29 Jul 1995 20:18:29 -0700 Mike Fitzgibbon (fitz at argus.lpl.arizona.edu) wrote: : Well the problem seems to be that "buggy" software is generating these FITS : files. According to the format definition, (see Astron. Astrophys. Suppl. : Ser. 44, (1981) pp363-370, "FITS: A Flexible Image Transport System" by D.C. : Wells, E.W. Greisen, and R.H. Harten), a logical variable (e.g. SIMPLE) has a : value of T or F in column 30. This header put the logical value in column 11. : Also of note is that integer and real values should be right justified in : columns 11-30 and character variables should have a single quote in column 11 : with the ending quote no earlier that column 20. Thanks for all the help from those who responded on Usenet and via email. I've succeeded in using DOS DEBUG to touch-up 2 files well enough to get FITSView to read them. Now, with this experience and the NOST FITS manual in hand, I am prepared to touch-up a few more and chat intelligently with the originator of the problematic files. In case anyone is interested, FITSView 0.4.0 required the "T" to be in column 30 but was unconcerned about the character strings not being in quotes. -Dave --------------------------------------------------------------------- | Dave Atherton - Work: cara25 at email.mot.com - Home: atherton at crl.com | | Christian Radio Pipeline Page: http://www.crl.com/~atherton/ | | "There's a fine line between participation and mockery." - S. Adams | --------------------------------------------------------------------- From fitsbits-request Tue Aug 1 08:22:41 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1310" "Sat" "29" "July" "1995" "00:57:59" "GMT" "Stan Voynick" "svoynick at netcom.com" "" "40" "Re: FITSVIEW Header Problem" "^From:" nil nil "7" "1995072900:57:59" "FITSVIEW Header Problem" (number " " mark " Stan Voynick Jul 29 40/1310 " thread-indent "\"Re: FITSVIEW Header Problem\"\n") "<3v9s7a$p3f at crl9.crl.com>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA21133; Tue, 1 Aug 1995 08:22:41 -0400 Return-Path: Message-Id: Organization: NETCOM On-line Communication Services (408 261-4700 guest) Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!maui.cc.odu.edu!xanth.cs.odu.edu!night.primate.wisc.edu!caen!hookup!news.mathworks.com!gatech!howland.reston.ans.net!news.cac.psu.edu!news.tc.cornell.edu!newsserver.sdsc.edu!nic-nac.CSU.net!charnel.ecst.csuchico.edu!csusac!csus.edu!netcom.com!svoynick References: <3v9s7a$p3f at crl9.crl.com> Newsgroups: sci.astro.fits content-length: 1308 From: svoynick at netcom.com (Stan Voynick) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: FITSVIEW Header Problem Date: Sat, 29 Jul 1995 00:57:59 GMT In article <3v9s7a$p3f at crl9.crl.com>, Dave Atherton wrote: >I have used FITSVIEW 0.4.0 to view image files from several different >sources but am having difficulty with the files from a particular >automated telescope. > > .............. > >When I inspected the header, I found the "SIMPLE = T" and found that the >header contained 80-column records with no newlines...just like one >would expect. There's evidently something else about that header that >it doesn't like. > > .................. > >A file that I tested was JATILA01.ZIP. Unzipping produced 2 *.FTS files >which I renamed to *.FIT. Below is a listing of the header and first >line of image data from JATILA01.FIT (with a newline inserted after >every 80 chars for readability). Can somebody suggest what part of this >header might be bothering FITSVIEW? > >SIMPLE = T If this is reproduced accurately, then the "T" is not in the correct column. From the FITS standard: ==================== Start Quote ====================== 5.3.2.2 Logical Variable If the value is a logical constant, it shall appear as a T or F in column 30. ==================== End Quote ======================== Maybe try to manually go in and fix up a file and see if it works... - Stan Voynick - From fitsbits-request Tue Aug 1 08:58:44 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1875" "Tue" "1" "August" "1995" "08:58:32" "EDT" "Bill Cotton" "bcotton at gorilla.cv.nrao.edu" "<9508011258.AA03234 at gorilla.cv.nrao.edu>" "52" "Re: FITSVIEW Header Problem" "^From:" nil nil "8" "1995080112:58:32" "FITSVIEW Header Problem" (number " " mark " Bill Cotton Aug 1 52/1875 " thread-indent "\"Re: FITSVIEW Header Problem\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA21203; Tue, 1 Aug 1995 08:58:44 -0400 Return-Path: Message-Id: <9508011258.AA03234 at gorilla.cv.nrao.edu> In-Reply-To: References: <3v9s7a$p3f at crl9.crl.com> content-length: 1873 From: bcotton at gorilla.cv.nrao.edu (Bill Cotton) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu, svoynick at netcom.com (Stan Voynick) Subject: Re: FITSVIEW Header Problem Date: Tue, 1 Aug 95 08:58:32 EDT Stan Voynick writes: >>In article <3v9s7a$p3f at crl9.crl.com>, Dave Atherton wrote: >>I have used FITSVIEW 0.4.0 to view image files from several different >>sources but am having difficulty with the files from a particular >>automated telescope. >> >> .............. >> >>When I inspected the header, I found the "SIMPLE = T" and found that the >>header contained 80-column records with no newlines...just like one >>would expect. There's evidently something else about that header that >>it doesn't like. >> >> .................. >> >>A file that I tested was JATILA01.ZIP. Unzipping produced 2 *.FTS files >>which I renamed to *.FIT. Below is a listing of the header and first >>line of image data from JATILA01.FIT (with a newline inserted after >>every 80 chars for readability). Can somebody suggest what part of this >>header might be bothering FITSVIEW? >> >>SIMPLE = T > >If this is reproduced accurately, then the "T" is not in the correct >column. From the FITS standard: > >==================== Start Quote ====================== > >5.3.2.2 Logical Variable > >If the value is a logical constant, it shall appear as a T or F >in column 30. > >==================== End Quote ======================== > >Maybe try to manually go in and fix up a file and see if it works... > > - Stan Voynick - FITSview takes a rather strict interpretation of the FITS rules for the SIMPLE= T keyword as it has to use this to distinguish FITS files from others; thus the "T" must be in column 30. The interpretation of other keywords is somewhat more relaxed even though the FITS rules are that the "required" keywords shall be in a fixed format. FITSview ignores keywords that it doesn't need to interprete the image. Editing the file to put the "T" in the correct column should fix the problem. Bill Cotton From fitsbits-request Wed Aug 2 21:49:18 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["834" "" "31" "July" "1995" "06:51:03" "GMT" "Michael Olberg" "olberg at nain.oso.chalmers.se" "<3vhugn$cbn at nyheter.chalmers.se>" "17" "Re: FITSVIEW Header Problem" "^From:" nil nil "7" "1995073106:51:03" "FITSVIEW Header Problem" (number " " mark " Michael Olberg Jul 31 17/834 " thread-indent "\"Re: FITSVIEW Header Problem\"\n") "<3vetm5$h9e at crl13.crl.com>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA25694; Wed, 2 Aug 1995 21:49:18 -0400 Return-Path: Message-Id: <3vhugn$cbn at nyheter.chalmers.se> Organization: Chalmers University of Technology Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!maui.cc.odu.edu!xanth.cs.odu.edu!night.primate.wisc.edu!caen!usenet.cis.ufl.edu!usenet.eel.ufl.edu!news.mathworks.com!gatech!news.sprintlink.net!sunic!sunic.sunet.se!news.chalmers.se!nain.oso.chalmers.se!not-for-mail References: <3v9s7a$p3f at crl9.crl.com> <3vb6k1$icm at news.ccit.arizona.edu> <3vetm5$h9e at crl13.crl.com> Newsgroups: sci.astro.fits content-length: 832 From: olberg at nain.oso.chalmers.se (Michael Olberg) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: FITSVIEW Header Problem Date: 31 Jul 1995 06:51:03 GMT Dave Atherton (atherton at crl.com) wrote: : Thanks for all the help from those who responded on Usenet and via email. : I've succeeded in using DOS DEBUG to touch-up 2 files well enough to get : FITSView to read them. Now, with this experience and the NOST FITS manual Just as a hint: I have at times used emacs (or any of its clones) to edit FITS files, it is one of the few editors who happily work on binary files as well. Might be less awkward to use for some of you than DEBUG. -- --------------------------------------------------------------------------- Michael Olberg | e-mail: olberg at oso.chalmers.se Onsala Space Observatory | tel: +46-31-7725507 S-43992 Onsala, SWEDEN | fax: +46-31-7725590 -------------------- This quote left blank intentionally ------------------ From fitsbits-request Fri Aug 4 05:41:41 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1014" "" "3" "August" "1995" "13:25:35" "GMT" "Ralph P. Pass" "rppass at tasc.com" "<3vqiof$gmu at hazel.Read.TASC.COM>" "30" "FITS 'extensions'" "^From:" nil nil "8" "1995080313:25:35" "FITS 'extensions'" (number " " mark " Ralph P. Pass Aug 3 30/1014 " thread-indent "\"FITS 'extensions'\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA28440; Fri, 4 Aug 1995 05:41:41 -0400 Return-Path: Message-Id: <3vqiof$gmu at hazel.Read.TASC.COM> Organization: TASC Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!sol.ctr.columbia.edu!news.msfc.nasa.gov!newsfeed.internetmci.com!news.sprintlink.net!cs.utexas.edu!uunet!in1.uu.net!newsserver.read.tasc.com!usenet Newsgroups: sci.astro.fits content-length: 1012 From: rppass at tasc.com (Ralph P. Pass) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: FITS 'extensions' Date: 3 Aug 1995 13:25:35 GMT I have looked at some of the FITS files that are generated by various CCD camera interface programs. I notices many non-defined (non-standard??) parameters. While the FITS document allows for this should we not standardize them so that at least we know what the parameters mean and what form they are?? For example, I have seen UT and TIME-OBS for the time. I have seen the values after these labels being in single quotes. I have seen them in the form with colons (:) separating the hours minutes seconds and with slashes (/). It is apparent that the FITS format is flexible but has grown into a non-standard. I, of course, would like to suggest a set of unofficial standards for many of the extensions I have seen (just to try to control the proliferation). Of course, getting the software suppliers to change to these suggestions might prove to be difficult. On the other hand, how about FITS having a set of defined 'unofficial' extensions?? Just some thoughts Ralph Pass rppass at tasc.com a From fitsbits-request Fri Aug 4 10:05:00 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["2065" "Fri" "4" "August" "1995" "10:04:53" "-0400" "Mike Corcoran" "corcoran at barnegat.gsfc.nasa.gov" "<199508041404.KAA02283 at barnegat.gsfc.nasa.gov>" "58" "Re: FITS 'extensions'" "^From:" nil nil "8" "1995080414:04:53" "FITS 'extensions'" (number " " mark " Mike Corcoran Aug 4 58/2065 " thread-indent "\"Re: FITS 'extensions'\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA28651; Fri, 4 Aug 1995 10:05:00 -0400 Return-Path: Message-Id: <199508041404.KAA02283 at barnegat.gsfc.nasa.gov> content-length: 2063 From: Mike Corcoran Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu, rppass at tasc.com Subject: Re: FITS 'extensions' Date: Fri, 4 Aug 1995 10:04:53 -0400 Ralph Pass writes: > > I have looked at some of the FITS files that are generated by various > CCD camera interface programs. I notices many non-defined > (non-standard??) parameters. While the FITS document allows for this > should we not standardize them so that at least we know what the > parameters mean and what form they are?? > > For example, I have seen UT and TIME-OBS for the time. I have seen > the values after these labels being in single quotes. I have seen > them in the form with colons (:) separating the hours minutes seconds > and with slashes (/). > > It is apparent that the FITS format is flexible but has grown into a > non-standard. > > I, of course, would like to suggest a set of unofficial standards for > many of the extensions I have seen (just to try to control the > proliferation). Of course, getting the software suppliers to change to > these suggestions might prove to be difficult. > > On the other hand, how about FITS having a set of defined 'unofficial' > extensions?? > > Just some thoughts > > Ralph Pass > rppass at tasc.com The High Energy Astrophysics Science Archive Research Center (HEASARC) at GSFC came to a similar conclusion. The HEASARC has the responsibility of archiving data from space-based high energy astronomy missions. We've adopted FITS as our archiving format, but we soon realized that non-conventional usages made analyzing the data a real pain. To help remedy this we've tried to promote some conventions. You can get more information about these conventions from the web page http://heasarc.gsfc.nasa.gov/docs/heasarc/ofwg/ofwg_intro.html for more information about the conventions we've recommended. best regards Mike Corcoran ******************************************************************************** Dr. Michael F. Corcoran High Energy Astrophysics Science Archive Research Center Goddard Space Flight Center Greenbelt, MD 20771 corcoran at barnegat.gsfc.nasa.gov LHEAVX::CORCORAN ******************************************************************************** From fitsbits-request Fri Aug 4 19:12:43 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["379" "" "4" "August" "1995" "15:33:05" "GMT" "keith at msmri.med.ubc.ca" "keith at msmri.med.ubc.ca" "<3vtejh$mg at nntp.ucs.ubc.ca>" "16" "Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080415:33:05" "Saving 100GB of data for 15 years cheaply" (number " " mark " keith at msmri.med.u Aug 4 16/379 " thread-indent "\"Saving 100GB of data for 15 years cheaply\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA01201; Fri, 4 Aug 1995 19:12:43 -0400 Return-Path: Message-Id: <3vtejh$mg at nntp.ucs.ubc.ca> Organization: MS/MRI UBC Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!sol.ctr.columbia.edu!news.mindlink.net!vanbc.wimsey.com!unixg.ubc.ca!news Reply-To: keith at msmri.med.ubc.ca (Keith S Cover) Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 377 From: keith at msmri.med.ubc.ca Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Saving 100GB of data for 15 years cheaply Date: 4 Aug 1995 15:33:05 GMT Please post replies to sci.data.formats. I need to save several hundred gigabytes of medical imaging data on a cheap reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes. This is cheap but is only good for about 1 year. Does anyone have any better suggestions? Thanks in advance. Keith S Cover Physics, UBC Vancouver, Canada keith at msmri.med.ubc.ca From fitsbits-request Sat Aug 5 09:45:15 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["603" "" "4" "August" "1995" "19:23:05" "GMT" "Christopher Michael Jones" "cjones at ccserver.uoregon.edu" "<3vts2p$4j0 at pith.uoregon.edu>" "16" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080419:23:05" "Saving 100GB of data for 15 years cheaply" (number " " mark " Christopher Micha Aug 4 16/603 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<3vtejh$mg at nntp.ucs.ubc.ca>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA01872; Sat, 5 Aug 1995 09:45:15 -0400 Return-Path: Message-Id: <3vts2p$4j0 at pith.uoregon.edu> Organization: University of Oregon Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!usenet.cis.ufl.edu!usenet.eel.ufl.edu!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!howland.reston.ans.net!vixen.cso.uiuc.edu!news.uoregon.edu!ccserver!cjones References: <3vtejh$mg at nntp.ucs.ubc.ca> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 601 From: cjones at ccserver.uoregon.edu (Christopher Michael Jones) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 4 Aug 1995 19:23:05 GMT keith at msmri.med.ubc.ca wrote: : Please post replies to sci.data.formats. : I need to save several hundred gigabytes of medical imaging data on a cheap : reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes. : This is cheap but is only good for about 1 year. Does anyone have any : better suggestions? : Thanks in advance. I'm not sure hom much getting a CD-ROM made would cost but that could be an option. You could try distributing chunks of it over the internet for people to save for you :-). Or you could wait about a while and put the data on 10 or 7 Gig DVD's. From fitsbits-request Sat Aug 5 11:21:27 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1673" "" "5" "August" "1995" "14:33:34" "GMT" "Tom Potter" "tdp at ix.netcom.com" "<3vvvfu$a1e at ixnews6.ix.netcom.com>" "37" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080514:33:34" "Saving 100GB of data for 15 years cheaply" (number " " mark " Tom Potter Aug 5 37/1673 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA01920; Sat, 5 Aug 1995 11:21:27 -0400 Return-Path: Message-Id: <3vvvfu$a1e at ixnews6.ix.netcom.com> Organization: Netcom Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!uwm.edu!cs.utexas.edu!swrinde!tank.news.pipex.net!pipex!howland.reston.ans.net!ix.netcom.com!netnews Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1671 From: tdp at ix.netcom.com (Tom Potter ) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 5 Aug 1995 14:33:34 GMT In mancini at maxlab.lu.se (Derrick C. Mancini) writes: > >In article <3vts2p$4j0 at pith.uoregon.edu>, cjones at ccserver.uoregon.edu >(Christopher Michael Jones) wrote: >>keith at msmri.med.ubc.ca wrote: >>: I need to save several hundred gigabytes of medical imaging data on a cheap >>: reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes. >>: This is cheap but is only good for about 1 year. Does anyone have any >>: better suggestions? >>I'm not sure hom much getting a CD-ROM made would cost but that could be >>an option. > >CD-ROM blanks will be slow to write, but are cheap (about $20 each) and >will hold about 700K. They also have the advantage of being readable on >a large number of machines, provided you put things into a readily accessible >format. You could also maintain an index CD. Write times are slow compared >to a hard drive, but should be better than the tape. Backup should be good >for 10 years, I think. I would guess that price per gigabyte per year of >storage would be competitive with tapes. The CD-ROM writable drives can >be had for as little as $2000. > >-- >Dr. Derrick C. Mancini >MAXLab, Lund University Advance Photon Source, ANL >Box 118, Ole Romers Vag 1 9700 S. Cass Avenue - APS/431 >S-221 00, Lund, Sweden Argonne, IL, USA 60439 >Tel: +46 46 222 3355 Tel: +1 708 252 0147 >FAX: +46 46 222 4710 FAX: +1 708 252 3222 >Email: mancini at maxlab.lu.se Email: mancini at aps.anl.gov > CD's hold about 700 MEGABYTES of data, uncompressed. They can hold at least two times this if the data is compressed ( Zipped ). From fitsbits-request Sun Aug 6 02:45:54 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["788" "" "4" "August" "1995" "19:53:13" "GMT" "Andy Loughe" "afl at cdc.noaa.gov" "<3vttr9$8r2 at lace.Colorado.EDU>" "26" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080419:53:13" "Saving 100GB of data for 15 years cheaply" (number " " mark " Andy Loughe Aug 4 26/788 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<3vtejh$mg at nntp.ucs.ubc.ca>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA03218; Sun, 6 Aug 1995 02:45:54 -0400 Return-Path: Message-Id: <3vttr9$8r2 at lace.Colorado.EDU> Organization: Climate Diagnostic Center Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!maui.cc.odu.edu!xanth.cs.odu.edu!night.primate.wisc.edu!nntp.msstate.edu!emory!swrinde!howland.reston.ans.net!vixen.cso.uiuc.edu!uchinews!ncar!massive!netsrv1.cmdl.noaa.gov!boulder!cdc.noaa.gov!afl References: <3vtejh$mg at nntp.ucs.ubc.ca> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 786 From: afl at cdc.noaa.gov (Andy Loughe) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 4 Aug 1995 19:53:13 GMT In article <3vtejh$mg at nntp.ucs.ubc.ca>, keith at msmri.med.ubc.ca writes: |> Please post replies to sci.data.formats. |> |> I need to save several hundred gigabytes of medical imaging data on a cheap |> reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes. |> This is cheap but is only good for about 1 year. Does anyone have any better |> suggestions? |> |> Thanks in advance. |> |> Keith S Cover |> Physics, UBC |> Vancouver, Canada |> keith at msmri.med.ubc.ca |> What about a small optical drive with removable disks? -- Andrew F. Loughe email: afl at cdc.noaa.gov University of Colorado, CIRES voice: (303) 492-0707 Campus Box 449 fax: (303) 497-7013 Boulder, CO 80309-0449 USA From fitsbits-request Mon Aug 7 09:37:10 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["3197" "" "7" "August" "1995" "11:24:58" "GMT" "Ralph Martin" "ralf at mail.ast.cam.ac.uk" "<404t6a$fme at lyra.csx.cam.ac.uk>" "49" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080711:24:58" "Saving 100GB of data for 15 years cheaply" (number " " mark " Ralph Martin Aug 7 49/3197 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<404l2e$40v at anarchy.io.com>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA04570; Mon, 7 Aug 1995 09:37:10 -0400 Return-Path: Message-Id: <404t6a$fme at lyra.csx.cam.ac.uk> Organization: Institute of Astronomy, Cambridge Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!spool.mu.edu!agate!sunsite.doc.ic.ac.uk!lyra.csx.cam.ac.uk!ralf References: <3vtejh$mg at nntp.ucs.ubc.ca> <404l2e$40v at anarchy.io.com> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 3195 From: ralf at mail.ast.cam.ac.uk (Ralph Martin) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 7 Aug 1995 11:24:58 GMT The question of how to store digital information for long periods of time is worth a hard look because there are already a number of archives in astronomy which have had to ask the same question. Our archive of telescope data was started in 1984 and originally used 1600 bpi nine track tape. The later computers had 6250 bpi nine track, then Exabyte and now DAT tapes. By the beginning of last year we had some 7000 nine track tapes which used about 60-70 square metres of floor space. More importantly we found out that these kind of nine track tapes, which were mostly held in sealed canisters, were an immense fire risk. There are records of a test done by the fire brigade at the Rutherford Laboratory. A few hundred of these tapes were put into a former bomb store and a lighted rag thrown in. A few minutes later, the fire was going so fiercely that it bent the door lintel which was made of reinforced concrete. The brigade doused the blaze with water but the fire restarted because the canisters provided their own oxygen and the cooled outer shell protected the still hot inner core. It took the fire crew serious amounts of foam to extinguish the blaze. We thus had to choose what medium we were going to transfer all this past data onto and it is here that the various options come into play. How often do you want to reaccess the data? If it rare then Exabyte will do the job but if you want to do it lots of times, the Exabyte drive wear the tape. (This has something to do with the angle that the tape makes with the reading heads). We asked around what was our best option was and we decided to copy onto DAT tapes rather than CD-Rom because a) DAT decks are cheap. b) DAT tapes are cheap c) It didn't need any extra software effort to our derchiving software. I'm prepared to believe that the choice could be made differently but the issue boils down to what is going to be around for the next 10-15 years and I'd say that DAT, Exabyte (maybe) and CD-Rom all have lifetimes of that order. I now have two cabinets full of DATs and a spare room of 65 square metres with students in it rather than tapes. The problem will probably return in 10-15 years by which time I'll have the 'gone fishing' sign over my door. The DAT's are being accessed regularly and so far we haven't had any problems. Since we now dearchive the equivalent of nearly a full telescope of data every year, it's important for it to be right. The DAT's are also considerably quicker than Exabytes when it comes to reading FITS formatted tapes (which are the de facto standard in astronomy) because Exabytes are hopeless at reading tape marks. Tape marks occur after every FITS file on a tape.... The correspondent who wrote about optical storage is also correct as the 10,000 Schmidt plates in our plate library and 150,000 glass plates in our plate archive (some of which are well over a century old) illustrate. -- Ralph Martin Internet: ralf at ast.cam.ac.uk Royal Greenwich Observatory Telephone: UK+ 01223-374000 Cambridge CB3 0EZ ISDN: UK+ 01223-328028 United Kingdom ISDN: UK+ 01223-328087 From fitsbits-request Mon Aug 7 21:56:59 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["654" "" "5" "August" "1995" "16:30:09" "-0700" "Ron Wickersham" "rjw at crl.com" "<400uu1$qqo at crl5.crl.com>" "24" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080523:30:09" "Saving 100GB of data for 15 years cheaply" (number " " mark " Ron Wickersham Aug 5 24/654 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<3vtejh$mg at nntp.ucs.ubc.ca>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA05959; Mon, 7 Aug 1995 21:56:59 -0400 Return-Path: Message-Id: <400uu1$qqo at crl5.crl.com> Organization: CRL Dialup Internet Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.mathworks.com!uhog.mit.edu!bloom-beacon.mit.edu!news.moneng.mei.com!uwm.edu!cs.utexas.edu!usc!howland.reston.ans.net!nntp.crl.com!crl5.crl.com!not-for-mail References: <3vtejh$mg at nntp.ucs.ubc.ca> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 652 From: rjw at crl.com (Ron Wickersham) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 5 Aug 1995 16:30:09 -0700 Please elaborate on the 8mm Exabyte storage lifetime. I thought that tape would be good for the fifteen year requirement stated. Do others report the Exabyte system only reliable for 1 year? -ron wickersham rjw at crl.com keith at msmri.med.ubc.ca wrote: : Please post replies to sci.data.formats. : I need to save several hundred gigabytes of medical imaging data on a cheap : reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes. : This is cheap but is only good for about 1 year. Does anyone have any better : suggestions? : Thanks in advance. : Keith S Cover : Physics, UBC : Vancouver, Canada : keith at msmri.med.ubc.ca From fitsbits-request Tue Aug 8 09:56:33 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["657" "Mon" "7" "August" "1995" "17:13:03" "GMT" "susan cassidy" "susanc at ssd3450.sandiegoca.attgis.com" "" "17" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080717:13:03" "Saving 100GB of data for 15 years cheaply" (number " " mark " susan cassidy Aug 7 17/657 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<404t6a$fme at lyra.csx.cam.ac.uk>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA06640; Tue, 8 Aug 1995 09:56:33 -0400 Return-Path: Message-Id: Organization: NCR LCPD San Diego Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!hookup!news.mathworks.com!zombie.ncsc.mil!paladin.american.edu!howland.reston.ans.net!news-e1a.megaweb.com!newstf01.news.aol.com!uunet!in1.uu.net!ncrgw2.ncr.com!ncrhub2!ncr-sd!lcpd2!ssd3450.SanDiegoCA.ATTGIS.COM!susanc References: <3vtejh$mg at nntp.ucs.ubc.ca> <404l2e$40v at anarchy.io.com> <404t6a$fme at lyra.csx.cam.ac.uk> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 655 From: susanc at ssd3450.SanDiegoCA.ATTGIS.COM (susan cassidy) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: Mon, 7 Aug 1995 17:13:03 GMT I would be reluctant to count on any form of magnetic tape lasting for 15 years. When I visited the U.S. Geological Survey office where they process the data from space probes, they told us they had problems with tapes distintegrating that had data that had not yet been processed. Also, the reliability of tape is very dependent on the environmental conditions. One other factor is that anything like 8mm Exabyte tapes may not be around (the drives/software to read them) 15 years from now. They did not exist 15 years ago. I think the CD/ROM idea is probably a better approach for now. -- Susan Cassidy email: susan.cassidy at sandiegoca.attgis.com From fitsbits-request Tue Aug 8 13:11:43 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1389" "" "8" "August" "1995" "13:34:11" "GMT" "Martin Leese" "mleese at hudson.cs.unb.ca" "<407p4j$ms2 at sol.sun.csd.unb.ca>" "28" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080813:34:11" "Saving 100GB of data for 15 years cheaply" (number " " mark " Martin Leese Aug 8 28/1389 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<406st6$8le at gwis2.circ.gwu.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA07192; Tue, 8 Aug 1995 13:11:43 -0400 Return-Path: Message-Id: <407p4j$ms2 at sol.sun.csd.unb.ca> Organization: Ocean Mapping Group, University of New Brunswick, Canada Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!zip.eecs.umich.edu!newshost.marcam.com!usc!howland.reston.ans.net!torn!news.unb.ca!hudson!mleese References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> Reply-To: MLeese at atlantic.cs.unb.ca Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1387 From: mleese at hudson.CS.unb.ca (Martin Leese - OMG) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 8 Aug 1995 13:34:11 GMT On 8 Aug 1995 01:32:22 -0400 Scott Fearing (sfearing at gwis2.circ.gwu.edu) wrote: >> I should warn you of a potential danger of CD storage: CD 'rot'. >> Apparently, the aluminum can oxidize under the plastic protective covering >> and data can be lost. This is appearing on CDs as well as laserdiscs, >> but I assume it does not happen to the special gold plated discs that come >> out on the audiophile pressings. There are discussions of this from time >> to time on rec.video.laserdisc. Laserdiscs are made in two halves and then glued together. There are problems with the glue eating the aluminum layer - disc rot - but the glue was changed many years ago. It takes years for the Laserdisc to rot, so we are currently reaping the "benefits" of a past mistake. Audio CDs and CD-ROMs are not made in two halves so there is no glue. There was a problem many years ago with acid from the cheap cardboard sleeves of promo CDs attacking the aluminium. There was also a problem many years ago with the ink used to print onto the CD. Both of these problems are long gone. Regards, Martin E-mail: mleese at atlantic.cs.unb.ca WWW: http://www.omg.unb.ca/~mleese/ ________________________________________________________________________ Want to know how Ambisonics can improve the sound of your LPs and CDs? See the Ambisonic Surround Sound FAQ. Details on my WWW Home page. From fitsbits-request Tue Aug 8 16:01:36 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1461" "" "8" "August" "1995" "10:56:29" "-0400" "Ben Weiner" "bweiner at electron.rutgers.edu" "<407tut$21e at electron.rutgers.edu>" "30" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080814:56:29" "Saving 100GB of data for 15 years cheaply" (number " " mark " Ben Weiner Aug 8 30/1461 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<406st6$8le at gwis2.circ.gwu.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA07857; Tue, 8 Aug 1995 16:01:36 -0400 Return-Path: Message-Id: <407tut$21e at electron.rutgers.edu> Organization: Rutgers University Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!hookup!news.mathworks.com!gatech!howland.reston.ans.net!news-e1a.megaweb.com!newstf01.news.aol.com!uunet!in2.uu.net!dziuxsolim.rutgers.edu!electron.rutgers.edu!not-for-mail References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1459 From: bweiner at electron.rutgers.edu (Ben Weiner) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 8 Aug 1995 10:56:29 -0400 sfearing at gwis2.circ.gwu.edu (Scott Fearing) writes: > I should warn you of a potential danger of CD storage: CD 'rot'. >Apparently, the aluminum can oxidize under the plastic protective covering >and data can be lost. This is appearing on CDs as well as laserdiscs, >but I assume it does not happen to the special gold plated discs that come >out on the audiophile pressings. There are discussions of this from time >to time on rec.video.laserdisc. Uh. This sounds like an urban legend. Or one of those audiophile stories. If aluminum oxidizes, it forms, natch, aluminum oxide, which is clear. In fact first-surface aluminized mirrors always form a thin layer of clear aluminum oxide on the top, which is not bad, but good - aluminum oxide is less reactive than raw aluminum, so it prevents tarnish. This is not to in any way endorse claims that CDs are permanent. But I believe that in 15 years your CDs will probably still be good. Also, I believe that in 15 years you will be able to find a CD(ROM) player. I can barely read nine-track tapes now, and in a couple of years I expect it would be a real pain in the ass finding a nine-track drive. Some of the tapes will be less than 10 years old then, yet unreadable. (Fortunately, there's nothing important on them.) I expect the situation to be better with Exabytes and even better with DATs. But still ... -- History / has a stutter / it says / w-w-w-w-watch out! - the mekons From fitsbits-request Wed Aug 9 06:43:07 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["2258" "Tue" "8" "August" "1995" "17:42:05" "GMT" "Mitchell R. Grunes" "grunes at news.nrl.navy.mil" "" "45" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080817:42:05" "Saving 100GB of data for 15 years cheaply" (number " " mark " Mitchell R. Grune Aug 8 45/2258 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<407tut$21e at electron.rutgers.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA09984; Wed, 9 Aug 1995 06:43:07 -0400 Return-Path: Message-Id: Organization: AlliedSignal Technical Services Corp. Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!zip.eecs.umich.edu!panix!news.mathworks.com!newsfeed.internetmci.com!EU.net!howland.reston.ans.net!news.cac.psu.edu!news.math.psu.edu!ra.nrl.navy.mil!news.nrl.navy.mil!grunes References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> <407tut$21e at electron.rutgers.edu> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 2256 From: grunes at news.nrl.navy.mil (Mitchell R Grunes) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: Tue, 8 Aug 1995 17:42:05 GMT 1. Why not leave put the data on those 1 year lifetime (but high capacity) tapes, but re-copy it every 5 or 6 months? Since it would only take a few tapes, and you could copy one overnight, it really wouldn't take much of your time. If you are paranoid, add an extra layer of error detection and/or correction coding, keep the last version's tape around in case an error happens, and I would think you could make it work. This isn't based on extended experience--it just seems reasonable. Re-copying needs to be done even with 9 tracks, though less often. With any media. 2. About your old 9 tracks. How could they burn in sealed containers? Surely there is a simple solution to this problem? Inert gas? 3. Since no one liked my idea of film, I've an even better idea. Print it all out in hex, on acid free paper. When you need it, use a scanner. Let's see: two characters/byte, about 100 characters/line and 60 lines/page...now that's a REAL fire hazard. Compact, too. (-: 4. I've tried to think up a solution to a somewhat similar problem, and wasn't real happy with any of the solutions. My personal solution mostly followed #1 above, because the tape drives were already around, whereas we would have to buy a CD recorder, and it would have taken ALOT of CDs--a lot of time and energy to deal with. 5. For really long term storage, one would have to buy extra tape drives, and computers to put them in, and a few decades of spare parts. I really wish libraries would keep a few old computers around, of every conceivable common design, to make sure people could read old media. Would you believe that most scientific labs would now have trouble reading 9 tracks? It's not very sexy, but the need to read old media (in spite of the errors) is going to be a very real problem. What would you do if someone handed you an old TRS-80 audio tape with data on which the fate of the world (or at least your personal future) depended? ------------------------------------------------------------------------- Opinions are mine alone. I am not a "Real Programmer". I only program in the Real World. Mitchell R Grunes (grunes at nrlvax.nrl.navy.mil) AlliedSignal Technical Services Corp. From fitsbits-request Wed Aug 9 07:01:43 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1944" "" "8" "August" "1995" "18:23:06" "GMT" "Steve Allen" "sla at umbra.ucolick.org" "<408a2a$h50 at darkstar.UCSC.EDU>" "37" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080818:23:06" "Saving 100GB of data for 15 years cheaply" (number " " mark " Steve Allen Aug 8 37/1944 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<40778p$rd9 at noao.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA10017; Wed, 9 Aug 1995 07:01:43 -0400 Return-Path: Message-Id: <408a2a$h50 at darkstar.UCSC.EDU> Organization: UCO/Lick Observatory, Santa Cruz Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!hookup!news.mathworks.com!zombie.ncsc.mil!news.duke.edu!agate!darkstar.UCSC.EDU!umbra.ucolick.org!sla References: <3vtejh$mg at nntp.ucs.ubc.ca> <40778p$rd9 at noao.edu> Newsgroups: sci.astro.fits content-length: 1942 From: sla at umbra.ucolick.org (Steve Allen) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 8 Aug 1995 18:23:06 GMT Note that I have restricted this to sci.astro.fits only, as it is not relevant to any of the other original newsgroups. In article <40778p$rd9 at noao.edu>, Rob Seaman wrote: >Another way to address this problem is to write larger files onto the >tapes - and thus fewer tape marks. FITS image extensions (and other >FITS extensions) allow a good way to do this by concatenating multiple >images into a single large file. But in the instance of non-IMAGE extensions how would it be possible to determine the relationships between the various different kinds of HDUs? I suspect that this would be an excellent application for the FITS Grouping Convention which was proposed last year. >any single image in the archive is accessible in under 10 minutes. Not >exactly what I'd call "online", but quite acceptable for our purposes. >In addition, multiple retrievals are much more likely than single image >retrievals due to the calibration data, multiple filters and exposures, >and the initial seek time is shared between the entire list of images >retrieved. Again, the proposed FITS grouping extension seems to contain the necessary structure to document the relationships between all such images stored together. I'm putting together specs for the Keck DEIMOS data storage, and as a part of the data that comes out of that we think we will be generating FITS tables that describe the slitlet configurations. We're planning on embedding the FITS grouping convention keywords (which are simple to add and unobtrusive to a FITS reader that does not care about them). -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Steve Allen UCO/Lick Observatory Santa Cruz, CA 95064 sla at ucolick.org Voice: +1 408 459 3046 FAX: +1 408 454 9863 Notice: The ucolick.org domain has just changed from being lick.ucsc.edu From fitsbits-request Wed Aug 9 08:57:20 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1409" "Sat" "5" "August" "1995" "14:13:05" "GMT" "Derrick C. Mancini" "mancini at maxlab.lu.se" "" "29" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080514:13:05" "Saving 100GB of data for 15 years cheaply" (number " " mark " Derrick C. Mancin Aug 5 29/1409 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<3vts2p$4j0 at pith.uoregon.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA10236; Wed, 9 Aug 1995 08:57:20 -0400 Return-Path: Message-Id: Organization: MAXLab (Lund University) & APS (Argonne Nat. Lab.) Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!sunic!sunic.sunet.se!news.lth.se!news.lu.se!usx121.maxlab.lu.se!user References: <3vtejh$mg at nntp.ucs.ubc.ca> <3vts2p$4j0 at pith.uoregon.edu> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1407 From: mancini at maxlab.lu.se (Derrick C. Mancini) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: Sat, 5 Aug 1995 14:13:05 GMT In article <3vts2p$4j0 at pith.uoregon.edu>, cjones at ccserver.uoregon.edu (Christopher Michael Jones) wrote: >keith at msmri.med.ubc.ca wrote: >: I need to save several hundred gigabytes of medical imaging data on a cheap >: reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes. >: This is cheap but is only good for about 1 year. Does anyone have any >: better suggestions? >I'm not sure hom much getting a CD-ROM made would cost but that could be >an option. CD-ROM blanks will be slow to write, but are cheap (about $20 each) and will hold about 700K. They also have the advantage of being readable on a large number of machines, provided you put things into a readily accessible format. You could also maintain an index CD. Write times are slow compared to a hard drive, but should be better than the tape. Backup should be good for 10 years, I think. I would guess that price per gigabyte per year of storage would be competitive with tapes. The CD-ROM writable drives can be had for as little as $2000. -- Dr. Derrick C. Mancini MAXLab, Lund University Advance Photon Source, ANL Box 118, Ole Romers Vag 1 9700 S. Cass Avenue - APS/431 S-221 00, Lund, Sweden Argonne, IL, USA 60439 Tel: +46 46 222 3355 Tel: +1 708 252 0147 FAX: +46 46 222 4710 FAX: +1 708 252 3222 Email: mancini at maxlab.lu.se Email: mancini at aps.anl.gov From fitsbits-request Wed Aug 9 13:03:39 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1585" "" "9" "August" "1995" "13:07:45" "GMT" "Mitchell R. Grunes" "grunes at news.nrl.navy.mil" "" "33" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080913:07:45" "Saving 100GB of data for 15 years cheaply" (number " " mark " Mitchell R. Grune Aug 9 33/1585 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<409gdi$2qrq at yuma.ACNS.ColoState.EDU>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA11180; Wed, 9 Aug 1995 13:03:39 -0400 Return-Path: Message-Id: Organization: AlliedSignal Technical Services Corp. Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!uwm.edu!math.ohio-state.edu!howland.reston.ans.net!newsserver.jvnc.net!newsserver2.jvnc.net!news.cac.psu.edu!news.math.psu.edu!ra.nrl.navy.mil!mitch.nrl.navy.mil!grunes References: <3vtejh$mg at nntp.ucs.ubc.ca> <409gdi$2qrq at yuma.ACNS.ColoState.EDU> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1583 From: grunes at news.nrl.navy.mil (Mitchell R Grunes) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 9 Aug 1995 13:07:45 GMT In article <409gdi$2qrq at yuma.ACNS.ColoState.EDU> dean at phobos.cira.colostate.edu writes: > CD Recorders would be my vote. I am currently using a Pinnicle Recorder >to make 650Mb CD-ROMs... > > The CD that I am making with the Pinnicle do have problems. Some CD-ROMs > will not read them. It is not a problem for us, because all the CD we > make work on our CD-ROMs on our Window NT system. However, there have been > some people reporting that their DOS machines with a CD-ROM wil not see the > disk in the drive. Could other people give their experiences on this, with the Pinnicle and with other drives? Does anyone know whether it is an issue of: (1) Format. Windows NT file systems may be unreadable to DOS boxes. In fact, Some new DOS file systems will be unreadable to old DOS's, and any compressed file system (e.g., Stacker) will be unreadable if you aren't using the same compression system. (2) Reliability. I suppose it is possible that some/all CD writers (or some/all brands of blank disks) do a borderline job, producing disks that are just barely readable by some drives, not readable by others. Kind of like writing 360 KB floppies on 1.2 MB drives. If the latter is true, it would make a big difference to those of us who might otherwise consider buying a CD-ROM writer. ------------------------------------------------------------------------- Opinions are mine alone. I am not a "Real Programmer". I only program in the Real World. Mitchell R Grunes (grunes at nrlvax.nrl.navy.mil) AlliedSignal Technical Services Corp. From fitsbits-request Wed Aug 9 14:19:29 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["623" "Wed" "9" "August" "1995" "13:39:37" "GMT" "Doug McDonald" "mcdonald at aries.scs.uiuc.edu" "" "15" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080913:39:37" "Saving 100GB of data for 15 years cheaply" (number " " mark " Doug McDonald Aug 9 15/623 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA11473; Wed, 9 Aug 1995 14:19:29 -0400 Return-Path: Message-Id: Organization: UIUC Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!uwm.edu!vixen.cso.uiuc.edu!crux.scs.uiuc.edu!mcdonald References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 621 From: mcdonald at aries.scs.uiuc.edu (Doug McDonald) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: Wed, 9 Aug 1995 13:39:37 GMT In article mancini at maxlab.lu.se (Derrick C. Mancini) writes: >> I should warn you of a potential danger of CD storage: CD 'rot'. >>Apparently, the aluminum can oxidize under the plastic protective covering >>and data can be lost. Actually, the data can be recovered from a CD anyway. The overcoat can be stripped and the metal completely removed with acid. Then recoat. Or the uncoated disk can be played like a record, mechanically, with a simple low-resolution AFM of the "beetle" type. In fact, uncoated CDs are used as tests for these gizmos. Doug McDonald From fitsbits-request Wed Aug 9 15:42:12 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["644" "" "9" "August" "1995" "14:13:18" "GMT" "Mitchell R. Grunes" "grunes at news.nrl.navy.mil" "" "15" "QIC Tape Lifetimes? was: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080914:13:18" "QIC Tape Lifetimes? was: Saving 100GB of data for 15 years cheaply" (number " " mark " Mitchell R. Grune Aug 9 15/644 " thread-indent "\"QIC Tape Lifetimes? was: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA12153; Wed, 9 Aug 1995 15:42:12 -0400 Return-Path: Message-Id: Organization: AlliedSignal Technical Services Corp. Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!uwm.edu!vixen.cso.uiuc.edu!howland.reston.ans.net!news.cac.psu.edu!news.math.psu.edu!ra.nrl.navy.mil!mitch.nrl.navy.mil!grunes References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> Newsgroups: sci.data.formats,sci.image.processing,sci.astro.fits,comp.sys.ibm.pc.hardware.storage content-length: 642 From: grunes at news.nrl.navy.mil (Mitchell R Grunes) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: QIC Tape Lifetimes? was: Saving 100GB of data for 15 years cheaply Date: 9 Aug 1995 14:13:18 GMT As long as we are on the subject of tape lifetimes, what is the lifetime of QIC family tapes, such as the ones rated at (compressed) capacities of 250 MB, 350 MB, and 700 MB? Also, does one extend the # of uses (usually rated at 3-5 back-ups, due to the large number of passes required to handle multiple tracks with one head) of these tapes if they are commercially pre-formatted? ------------------------------------------------------------------------- Opinions are mine alone. I am not a "Real Programmer". I only program in the Real World. Mitchell R Grunes (grunes at nrlvax.nrl.navy.mil) AlliedSignal Technical Services Corp. From fitsbits-request Wed Aug 9 15:50:38 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["624" "" "9" "August" "1995" "14:20:14" "GMT" "Dennis Ward" "dward at comet.ucar.edu" "<40ag6u$b9q at ncar.ucar.edu>" "27" "FITS to TIFF conversion?" "^From:" nil nil "8" "1995080914:20:14" "FITS to TIFF conversion?" (number " " mark " Dennis Ward Aug 9 27/624 " thread-indent "\"FITS to TIFF conversion?\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA12178; Wed, 9 Aug 1995 15:50:38 -0400 Return-Path: Message-Id: <40ag6u$b9q at ncar.ucar.edu> Organization: UCAR / Comet Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!uwm.edu!lll-winken.llnl.gov!simtel!news.kei.com!bloom-beacon.mit.edu!newsfeed.internetmci.com!tank.news.pipex.net!pipex!howland.reston.ans.net!swrinde!elroy.jpl.nasa.gov!ncar!usenet Newsgroups: sci.astro.fits content-length: 622 From: Dennis Ward Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: FITS to TIFF conversion? Date: 9 Aug 1995 14:20:14 GMT I am looking for software that will allow me to convert FITS image files to TIFF, TARGA or any other 24-bit graphics format. The conversion software MUST run on a PC (either DOS or Windows). Thanks in advance. Clear Skies! Dennis L. Ward Internet: dward at comet.ucar.edu Multimedia Author/Programmer UCAR/COMET PO BOX 3000 Phone: (303) 497-8345 Boulder, Co. 80307-3000 FAX: (303) 497-8491 WWW HomePage: http://adder.colorado.edu/~ward/DLW/dlw.html We are all in the gutter, but some of us are looking at the stars. -- Oscar Wilde From fitsbits-request Wed Aug 9 16:32:17 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1041" "Mon" "7" "August" "1995" "06:12:23" "GMT" "Jack Hudler" "jack at itsnet.com" "<404au3$jlh at itchy.itsnet.com>" "26" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080706:12:23" "Saving 100GB of data for 15 years cheaply" (number " " mark " Jack Hudler Aug 7 26/1041 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<3vvvfu$a1e at ixnews6.ix.netcom.com>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA12274; Wed, 9 Aug 1995 16:32:17 -0400 Return-Path: Message-Id: <404au3$jlh at itchy.itsnet.com> Organization: Internet Technology Systems, Provo UT USA Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.mathworks.com!news.kei.com!nntp.et.byu.edu!news.itsnet.com!usenet References: <3vvvfu$a1e at ixnews6.ix.netcom.com> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1039 From: jack at itsnet.com (Jack Hudler) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: Mon, 07 Aug 1995 06:12:23 GMT tdp at ix.netcom.com (Tom Potter ) wrote: >In mancini at maxlab.lu.se (Derrick C. Mancini) writes: >> >>In article <3vts2p$4j0 at pith.uoregon.edu>, cjones at ccserver.uoregon.edu >>(Christopher Michael Jones) wrote: >>>keith at msmri.med.ubc.ca wrote: >>>: I need to save several hundred gigabytes of medical imaging data on a cheap >>>: reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes. >>>: This is cheap but is only good for about 1 year. Does anyone have any >>>: better suggestions? >CD's hold about 700 MEGABYTES of data, uncompressed. >They can hold at least two times this if the data is compressed ( Zipped ). If you find the right software (Young Minds comes to mind :-) ), and if your not worried about losses in the image, you could write the data in raw CD data... in this case a CD can hold 1.2 to 1.6 gig. The final amount would depend on the amount of correction information you supply with the image. A CD has a huge overhead for the correction logic. From fitsbits-request Wed Aug 9 22:25:29 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1435" "" "7" "August" "1995" "09:06:22" "GMT" "Edwin Strickland" "EdStrick at io.com" "<404l2e$40v at anarchy.io.com>" "29" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080709:06:22" "Saving 100GB of data for 15 years cheaply" (number " " mark " Edwin Strickland Aug 7 29/1435 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<3vtejh$mg at nntp.ucs.ubc.ca>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA14058; Wed, 9 Aug 1995 22:25:29 -0400 Return-Path: Message-Id: <404l2e$40v at anarchy.io.com> Organization: ... someday I'll get organized ........... Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.mathworks.com!hookup!hudson.lm.com!news.math.psu.edu!news.cac.psu.edu!howland.reston.ans.net!news-e1a.megaweb.com!newstf01.news.aol.com!uunet!in2.uu.net!news.io.com!usenet References: <3vtejh$mg at nntp.ucs.ubc.ca> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1433 From: EdStrick at io.com (Edwin Strickland) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 7 Aug 1995 09:06:22 GMT In article <3vtejh$mg at nntp.ucs.ubc.ca>, keith at msmri.med.ubc.ca says... > >Please post replies to sci.data.formats. > >I need to save several hundred gigabytes of medical imaging data on a cheap >reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes >This is cheap but is only good for about 1 year. Does anyone have any better suggestions? CD-ROM recorders mail order for $1,599 US dollars, and I see prices of "less than" $8. for recordable CDs (when ordered in large quantity -- hundreds?). Recordable CD's are supposed to be as stable as factor- pressed ones, or so I think I've been told. This is "decades" of shelf life. I doubt you can do better, especially since it is a medium that will be supported for a couple decades, at least. When they have high- density recordable CD's (several gig, instead of 640 meg) switch to that. --------------------------------------------------------------------- Edwin L. Strickland III Image Processor/Analyst Austin, Tx. Planetary, Geologist (esp. Mars) Ed Strick at IO .com SF, Hard Fantasy, Classical | Music, Wild-Cat Species Fan | | IO.com is Illuminati Online, of Steve Jackson Games << Shameless Plug --------------------------------------------------------------------- From fitsbits-request Wed Aug 9 23:25:26 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["729" "" "6" "August" "1995" "17:59:19" "+0200" "Paul Schlyter" "pausch at electra.saaf.se" "<402osn$q30 at electra.saaf.se>" "19" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080615:59:19" "Saving 100GB of data for 15 years cheaply" (number " " mark " Paul Schlyter Aug 6 19/729 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<3vtejh$mg at nntp.ucs.ubc.ca>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA14274; Wed, 9 Aug 1995 23:25:26 -0400 Return-Path: Message-Id: <402osn$q30 at electra.saaf.se> Organization: Svensk Amat|rAstronomisk F|rening Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.mathworks.com!news.ultranet.com!news.sprintlink.net!sunic!sunic.sunet.se!nntp-trd.UNINETT.no!Norway.EU.net!nac.no!news.kth.se!admin.kth.se!merope.saaf.se!electra.saaf.se!not-for-mail References: <3vtejh$mg at nntp.ucs.ubc.ca> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 727 From: pausch at electra.saaf.se (Paul Schlyter) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 6 Aug 1995 17:59:19 +0200 In article <3vtejh$mg at nntp.ucs.ubc.ca>, wrote: >Please post replies to sci.data.formats. > >I need to save several hundred gigabytes of medical imaging data on a cheap >reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes. >This is cheap but is only good for about 1 year. Does anyone have any better >suggestions? Since the data consists of images, why not try photographic plates or film? That medium is known to be long-lasting and reliable. -- ---------------------------------------------------------------- Paul Schlyter, Swedish Amateur Astronomer's Society (SAAF) Nybrogatan 75 A, S-114 40 Stockholm, SWEDEN e-mail: pausch at saaf.se paul.schlyter at ausys.se From fitsbits-request Thu Aug 10 00:48:34 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1063" "" "9" "August" "1995" "15:56:48" "GMT" "Anthony J. Ferro" "tferro at merlin.la.asu.edu" "<40als0$f6k at news.ccit.arizona.edu>" "25" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080915:56:48" "Saving 100GB of data for 15 years cheaply" (number " " mark " Anthony J. Ferro Aug 9 25/1063 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA14452; Thu, 10 Aug 1995 00:48:34 -0400 Return-Path: Message-Id: <40als0$f6k at news.ccit.arizona.edu> Organization: The University of Arizona Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!zip.eecs.umich.edu!newshost.marcam.com!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!howland.reston.ans.net!swrinde!elroy.jpl.nasa.gov!ncar!noao!CS.Arizona.EDU!news.Arizona.EDU!merlin!tferro References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1061 From: tferro at merlin.la.asu.edu (Anthony J. Ferro) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 9 Aug 1995 15:56:48 GMT Just to pitch in a couple of items to this discussion: CD-R (recordable CD-ROM) is different than normal, manufactured CDs. The process is more photographic (the laser changes the reflectivity of the media), and isn't magnetic (MO). With both CD and CD-R, the lifetimes aren't well established. I think the claims are that CD-R might be around the 15-year timescale. A well made normal CD should be much longer, 50-100 years. The problems someone was having with Pinnacle CD-Rs is mostly to do with differences in how people interpret the ISO 9660 standard, and which level they support. If you stick to the level 1 (the wonderful 8.3 file names, etc), I've been able read disks on almost any computer. When you go above that, you can have problems. Some of these problems can be solved using other software, besides that shipped by Pinnacle. --------- Anthony J. Ferro aferro at as.arizona.edu Steward Observatory, NICMOS Project Ph: (520) 621-8683 University of Arizona FAX: (520) 621-1891 Tucson, AZ 85721 --------- From fitsbits-request Thu Aug 10 09:21:03 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1262" "" "7" "August" "1995" "10:47:53" "-0700" "Stephen Walton" "swalton at galileo.csun.edu" "<405jk9$3vq at galileo.csun.edu>" "30" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080717:47:53" "Saving 100GB of data for 15 years cheaply" (number " " mark " Stephen Walton Aug 7 30/1262 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<3vts2p$4j0 at pith.uoregon.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA15163; Thu, 10 Aug 1995 09:21:03 -0400 Return-Path: Message-Id: <405jk9$3vq at galileo.csun.edu> Organization: Cal State Northridge Dept. of Physics & Astronomy Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.mathworks.com!news.ultranet.com!news.sprintlink.net!newsfeed.internetmci.com!tank.news.pipex.net!pipex!howland.reston.ans.net!usc!nic-nac.CSU.net!csun.edu!galileo.csun.edu!not-for-mail References: <3vtejh$mg at nntp.ucs.ubc.ca> <3vts2p$4j0 at pith.uoregon.edu> Newsgroups: sci.data.formats,sci.image.processing,sci.astro.fits content-length: 1260 From: swalton at galileo.csun.edu (Stephen Walton) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 7 Aug 1995 10:47:53 -0700 In article <3vts2p$4j0 at pith.uoregon.edu>, Christopher Michael Jones wrote: >keith at msmri.med.ubc.ca wrote: >: Please post replies to sci.data.formats. OK. >: I need to save several hundred gigabytes of medical imaging data on a cheap >: reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes. >: This is cheap but is only good for about 1 year. Wow, I had no idea Exabytes were good for such a short time. >I'm not sure hom much getting a CD-ROM made would cost but that could be >an option. CD-ROM recorders are down to about $2000, and the recordable media are about $7.50 per disk. These are write-once, and (of course) have the advantage that any $1000 'multimedia' PC can read them. This didn't become a good option until recently, when 'multisession' recorders made it possible to record data in multiple pieces. Before that, at least as far as I understand, you had to have a complete image of all 660 MB on a hard disk and record the CD in one pass. We just bought a Yamaha 4X CD-ROM recorder with the 'EasyPro for Windows' software but haven't tried it out yet. -- Stephen Walton, California State University, Northridge "Be careful what you wish for; you might get it." swalton at csun.edu From fitsbits-request Thu Aug 10 10:05:14 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["828" "Mon" "7" "August" "1995" "23:31:55" "GMT" "Mitchell R. Grunes" "grunes at news.nrl.navy.mil" "" "22" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080723:31:55" "Saving 100GB of data for 15 years cheaply" (number " " mark " Mitchell R. Grune Aug 7 22/828 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<3vtejh$mg at nntp.ucs.ubc.ca>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA15305; Thu, 10 Aug 1995 10:05:14 -0400 Return-Path: Message-Id: Organization: AlliedSignal Technical Services Corp. Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!cs.utexas.edu!swrinde!tank.news.pipex.net!pipex!howland.reston.ans.net!news.cac.psu.edu!news.math.psu.edu!ra.nrl.navy.mil!mitch.nrl.navy.mil!grunes References: <3vtejh$mg at nntp.ucs.ubc.ca> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 826 From: grunes at news.nrl.navy.mil (Mitchell R Grunes) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: Mon, 7 Aug 1995 23:31:55 GMT In article <3vtejh$mg at nntp.ucs.ubc.ca> keith at msmri.med.ubc.ca writes: >I need to save several hundred gigabytes of medical imaging data on a cheap >reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes. >This is cheap but is only good for about 1 year. Does anyone have any better >suggestions? I guess saving imagery on film sounds ridiculous, doesn't it? Surely no one would do such a thing? (-: (Some films have fairly long lives, it's cheap, and it doesn't require specialized and soon-outdated hardware to read. It's even obvious.) ------------------------------------------------------------------------- Opinions are mine alone. I am not a "Real Programmer". I only program in the Real World. Mitchell R Grunes (grunes at nrlvax.nrl.navy.mil) AlliedSignal Technical Services Corp. From fitsbits-request Thu Aug 10 18:47:46 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["3085" "" "8" "August" "1995" "08:29:13" "GMT" "Rob Seaman" "seaman at noao.edu" "<40778p$rd9 at noao.edu>" "61" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080808:29:13" "Saving 100GB of data for 15 years cheaply" (number " " mark " Rob Seaman Aug 8 61/3085 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA17621; Thu, 10 Aug 1995 18:47:46 -0400 Return-Path: Message-Id: <40778p$rd9 at noao.edu> Organization: National Optical Astronomy Observatories Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!tank.news.pipex.net!pipex!howland.reston.ans.net!vixen.cso.uiuc.edu!uchinews!ncar!noao!caliche.tuc.noao.edu!seaman References: <3vtejh$mg at nntp.ucs.ubc.ca> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 3083 From: Rob Seaman Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 8 Aug 1995 08:29:13 GMT In article <404t6a$fme at lyra.csx.cam.ac.uk> ralf at mail.ast.cam.ac.uk, Ralph Martin writes: > The DAT's are also considerably quicker than Exabytes when it comes > to reading FITS formatted tapes (which are the de facto standard in > astronomy) because Exabytes are hopeless at reading tape marks. Tape > marks occur after every FITS file on a tape.... Another way to address this problem is to write larger files onto the tapes - and thus fewer tape marks. FITS image extensions (and other FITS extensions) allow a good way to do this by concatenating multiple images into a single large file. The NOAO archive (http://iraf.noao.edu/~seaman/archive/archive.html) multiplexes images from a half dozen or so of our telescopes on Kitt Peak into FITS image extension files about 50 Mbytes in size. These are taped onto duplicate exabyte 8505s (non-compressed) over the network onto a central archive machine. The software would just as easily support DATs or other formats, but exabytes remain cheaper per Gbyte. With a capacity of 4.12 Gbytes (a very conservative choice amounting to 1.5M FITS records), this file size results in about 80 files per tape. Combined with the much smaller tape marks on the 8505 drives compared with the original 8200 drives, this small number of files means that any single image in the archive is accessible in under 10 minutes. Not exactly what I'd call "online", but quite acceptable for our purposes. In addition, multiple retrievals are much more likely than single image retrievals due to the calibration data, multiple filters and exposures, and the initial seek time is shared between the entire list of images retrieved. After a pair of tapes is full, active taping is shifted to a second pair of drives and the tapes are FITS checksum verified. The checksums also provide a way to verify the readability of the tapes at a later date (see ftp://iraf.noao.edu/misc/checksum/checksum.ps). One copy of each pair is shipped downtown, the other stays on the mountain providing some disaster protection. The tapes are stored in climate controlled computer labs in Tucson and on Kitt Peak. Note that climate control is important for blank media, too. Our tape handling procedures allow several days or weeks of preconditioning in the mountain computer room before the blank tapes are mounted in the drives - this allows the drives and media to equilibrate thermally. We also store the header catalog and the header/tape cross-index as separate data products from the data tapes themselves. This has obvious advantages, but from the standpoint of simple survival of the data, the biggest advantage is that the data can be easily recast onto new media, perhaps reordering or regrouping the data into new tape or disk FITS files, without actually having to read through and interpret the data. This isn't a minor point with Tbyte size archives. Recasting the data will be required periodically no matter what the current medium - it's only a question of how often. Rob Seaman seaman at noao.edu National Optical Astronomy Observatories, Tucson From fitsbits-request Thu Aug 10 19:20:42 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["545" "" "8" "August" "1995" "01:32:22" "-0400" "Scott Fearing" "sfearing at gwis2.circ.gwu.edu" "<406st6$8le at gwis2.circ.gwu.edu>" "18" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080805:32:22" "Saving 100GB of data for 15 years cheaply" (number " " mark " Scott Fearing Aug 8 18/545 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<404au3$jlh at itchy.itsnet.com>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA17676; Thu, 10 Aug 1995 19:20:42 -0400 Return-Path: Message-Id: <406st6$8le at gwis2.circ.gwu.edu> Organization: The George Washington University Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!uunet!in1.uu.net!gwu.edu!not-for-mail References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 543 From: sfearing at gwis2.circ.gwu.edu (Scott Fearing) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 8 Aug 1995 01:32:22 -0400 I should warn you of a potential danger of CD storage: CD 'rot'. Apparently, the aluminum can oxidize under the plastic protective covering and data can be lost. This is appearing on CDs as well as laserdiscs, but I assume it does not happen to the special gold plated discs that come out on the audiophile pressings. There are discussions of this from time to time on rec.video.laserdisc. -- Scott Scott Fearing National Symphony Orchestra and George Washington University: Washington,DC From fitsbits-request Thu Aug 10 22:05:48 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1351" "Tue" "8" "August" "1995" "00:54:37" "GMT" "Bradford Holden" "holden at oddjob.uchicago.edu" "" "36" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080800:54:37" "Saving 100GB of data for 15 years cheaply" (number " " mark " Bradford Holden Aug 8 36/1351 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA17829; Thu, 10 Aug 1995 22:05:48 -0400 Return-Path: Message-Id: Organization: University of Chicago, Astronomy and Astrophysics Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!dispatch.news.demon.net!demon!tank.news.pipex.net!pipex!howland.reston.ans.net!vixen.cso.uiuc.edu!uchinews!oddjob.uchicago.edu!holden References: <3vtejh$mg at nntp.ucs.ubc.ca> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1349 From: holden at oddjob.uchicago.edu (Bradford Holden) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: Tue, 8 Aug 1995 00:54:37 GMT In article grunes at news.nrl.navy.mil (Mitchell R Grunes) writes: >In article <3vtejh$mg at nntp.ucs.ubc.ca> keith at msmri.med.ubc.ca writes: >>I need to save several hundred gigabytes of medical imaging data on a cheap >>reliable media. We are currently saving the data to 10GB 8mm Exabyte tapes. >>This is cheap but is only good for about 1 year. Does anyone have any better >>suggestions? > >I guess saving imagery on film sounds ridiculous, doesn't it? Surely >no one would do such a thing? Oddly enough large numbers of astronomers do this. Film however, has some disadvantages. It doesn't have a great dynamic range (though fairly good for some of them) its not machine readable (unless you build a machine to do it), it has a nonlinear response, and its hard to make good quality copies cheaply. > >(-: > >(Some films have fairly long lives, it's cheap, and it doesn't require >specialized and soon-outdated hardware to read. It's even obvious.) > >------------------------------------------------------------------------- > Opinions are mine alone. > I am not a "Real Programmer". I only program in the Real World. > Mitchell R Grunes (grunes at nrlvax.nrl.navy.mil) > AlliedSignal Technical Services Corp. > -- Bradford "PROS is evil" Holden gradual student at the University of Chicago From fitsbits-request Fri Aug 11 02:43:20 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["967" "" "8" "August" "1995" "20:23:05" "GMT" "Martin Leese" "mleese at hudson.cs.unb.ca" "<408h39$m8p at sol.sun.csd.unb.ca>" "23" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080820:23:05" "Saving 100GB of data for 15 years cheaply" (number " " mark " Martin Leese Aug 8 23/967 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<400uu1$qqo at crl5.crl.com>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA18026; Fri, 11 Aug 1995 02:43:20 -0400 Return-Path: Message-Id: <408h39$m8p at sol.sun.csd.unb.ca> Organization: Ocean Mapping Group, University of New Brunswick, Canada Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!tank.news.pipex.net!pipex!howland.reston.ans.net!spool.mu.edu!torn!news.unb.ca!hudson!mleese References: <3vtejh$mg at nntp.ucs.ubc.ca> <400uu1$qqo at crl5.crl.com> Reply-To: MLeese at atlantic.cs.unb.ca Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 965 From: mleese at hudson.CS.unb.ca (Martin Leese - OMG) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 8 Aug 1995 20:23:05 GMT On 5 Aug 1995 16:30:09 -0700 Ron Wickersham (rjw at crl.com) wrote: >> Please elaborate on the 8mm Exabyte storage lifetime. I thought that tape >> would be good for the fifteen year requirement stated. Do others report >> the Exabyte system only reliable for 1 year? I used to help run an archive of seismic data using 300 GByte WORM jukeboxes. The data arrived on Exabytes and our experience was that if an Exabyte is less that 1 year old you should be surprised if you cannot read it. If an Exabyte is more than 1 year old you should not be surprised if you cannot read. Having said this, most Exabytes will read for many years, just don't rely on it. Regards, Martin E-mail: mleese at atlantic.cs.unb.ca WWW: http://www.omg.unb.ca/~mleese/ ________________________________________________________________________ Want to know how Ambisonics can improve the sound of your LPs and CDs? See the Ambisonic Surround Sound FAQ. Details on my WWW Home page. From fitsbits-request Fri Aug 11 04:25:29 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["2048" "Tue" "8" "August" "1995" "22:40:19" "GMT" "Paul Carpenter" "paul at pcserv.demon.co.uk" "<807921619snz at pcserv.demon.co.uk>" "51" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080822:40:19" "Saving 100GB of data for 15 years cheaply" (number " " mark " Paul Carpenter Aug 8 51/2048 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA18096; Fri, 11 Aug 1995 04:25:29 -0400 Return-Path: Message-Id: <807921619snz at pcserv.demon.co.uk> Organization: PC Services Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!tank.news.pipex.net!pipex!dispatch.news.demon.net!demon!mail2news.demon.co.uk!pcserv.demon.co.uk!paul References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> <407tut$21e at electron.rutgers.edu> Reply-To: paul at pcserv.demon.co.uk Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 2046 From: Paul Carpenter Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: Tue, 08 Aug 95 22:40:19 GMT grunes at news.nrl.navy.mil "Mitchell R Grunes" writes: -1. Why not leave put the data on those 1 year lifetime (but high -capacity) tapes, but re-copy it every 5 or 6 months? Since it would only -take a few tapes, and you could copy one overnight, it really wouldn't -take much of your time..... -This isn't based on extended experience--it just seems reasonable. AFAIR major banks, used to do this practise with their archives, to avoid tape ageing and stretch.. This was with old 9 track tapes and the like. How many people create archives/backups of important data without a true verification pass, let alone a compatability test. Too often I hear of backups being no use because:- 1/ Data was not checked after being written 2/ Drive being used for storage was out of calibration so backup media only worked on that drive. 3/ No non-compressed etc. copy of software to read backup was kept in an even safer place(s), that was verified to work!. -Re-copying needs to be done even with 9 tracks, though less often. With -any media. -2. About your old 9 tracks. How could they burn in sealed containers? -Surely there is a simple solution to this problem? Inert gas? Tried filling an individual tape canister with inert gas, and ensuring that oxygen was not released from the plastics used :-) -3.... In short whatever the archive/backup method used ensure it works and test that archive/backup can actually be used at some point in the future. No system is infallible, assess the archive/backups means/media/procedures/software/host to suit your needs now and in N years time. Keeping a system with spares for large archives may seem silly but in fact is very practical as long as people are around who still know how to fix them or run them. =------------------------------------------------- Paul Carpenter | paul at pcserv.demon.co.uk PC Services | tel (44) 1734-463634 PO Box 2030, Reading | fax (44) 1734-463423 RG4 8YW, UK | Consulting for DSP, Image processing, Data Acquisition, System tuning & more.. From fitsbits-request Fri Aug 11 06:28:52 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["851" "" "8" "August" "1995" "20:16:59" "GMT" "Martin Leese" "mleese at hudson.cs.unb.ca" "<408gnr$m8p at sol.sun.csd.unb.ca>" "20" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080820:16:59" "Saving 100GB of data for 15 years cheaply" (number " " mark " Martin Leese Aug 8 20/851 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA18193; Fri, 11 Aug 1995 06:28:52 -0400 Return-Path: Message-Id: <408gnr$m8p at sol.sun.csd.unb.ca> Organization: Ocean Mapping Group, University of New Brunswick, Canada Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!howland.reston.ans.net!torn!news.unb.ca!hudson!mleese References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> <407tut$21e at electron.rutgers.edu> Reply-To: MLeese at atlantic.cs.unb.ca Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 849 From: mleese at hudson.CS.unb.ca (Martin Leese - OMG) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 8 Aug 1995 20:16:59 GMT On Tue, 8 Aug 1995 17:42:05 GMT Mitchell R Grunes (grunes at news.nrl.navy.mil) wrote: ... >> 2. About your old 9 tracks. How could they burn in sealed containers? >> Surely there is a simple solution to this problem? Inert gas? Presumably the sealed container heated up, heating up the tape inside until it ignited. (Incidentially, this is how fire spreads on a ship, even through airtight steel bulkheads.) I cannot see how inert gas would help. Once the fire was going, you would have sealed canisters popping open all over the place. Regards, Martin E-mail: mleese at atlantic.cs.unb.ca WWW: http://www.omg.unb.ca/~mleese/ ________________________________________________________________________ Want to know how Ambisonics can improve the sound of your LPs and CDs? See the Ambisonic Surround Sound FAQ. Details on my WWW Home page. From fitsbits-request Fri Aug 11 12:06:31 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["617" "" "9" "August" "1995" "02:12:51" "GMT" "Adrian Whichello" "adrianw at cassius" "<4095j3$lfl at metro.ucc.su.OZ.AU>" "18" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080902:12:51" "Saving 100GB of data for 15 years cheaply" (number " " mark " Adrian Whichello Aug 9 18/617 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<406st6$8le at gwis2.circ.gwu.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA18843; Fri, 11 Aug 1995 12:06:31 -0400 Return-Path: Message-Id: <4095j3$lfl at metro.ucc.su.OZ.AU> Organization: Electrical Engineering, University of Sydney Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!zombie.ncsc.mil!simtel!harbinger.cc.monash.edu.au!news.cs.su.oz.au!metro!cassius!adrianw References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 615 From: adrianw at cassius (Adrian Whichello) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 9 Aug 1995 02:12:51 GMT In article <406st6$8le at gwis2.circ.gwu.edu> sfearing at gwis2.circ.gwu.edu (Scott Fearing) writes: > I should warn you of a potential danger of CD storage: CD 'rot'. >Apparently, the aluminum can oxidize under the plastic protective covering >and data can be lost. This is appearing on CDs as well as laserdiscs, >but I assume it does not happen to the special gold plated discs that come >out on the audiophile pressings. There are discussions of this from time >to time on rec.video.laserdisc. > > >-- > Scott yeah, but can't you paint all around the edge of the CD with a special green felt pen... Adrian. From fitsbits-request Fri Aug 11 12:41:32 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["658" "" "9" "August" "1995" "05:17:38" "GMT" "dean at phobos.cira.colostate.edu" "dean at phobos.cira.colostate.edu" "<409gdi$2qrq at yuma.ACNS.ColoState.EDU>" "17" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080905:17:38" "Saving 100GB of data for 15 years cheaply" (number " " mark " dean at phobos.cira. Aug 9 17/658 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<3vtejh$mg at nntp.ucs.ubc.ca>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA18983; Fri, 11 Aug 1995 12:41:32 -0400 Return-Path: Message-Id: <409gdi$2qrq at yuma.ACNS.ColoState.EDU> Organization: Colorado State Univ, CIRA Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!zombie.ncsc.mil!cs.umd.edu!haven.umd.edu!purdue!yuma!DEAN at phobos.cira.colostate.edu References: <3vtejh$mg at nntp.ucs.ubc.ca> Reply-To: dean at sol.cira.colostate.edu Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 656 From: dean at phobos.cira.colostate.edu Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 9 Aug 1995 05:17:38 GMT CD Recorders would be my vote. I am currently using a Pinnicle Recorder to make 650Mb CD-ROMs. It take about 30 min to make, but some time is involved with preparing the directory tree. We use 3M blank disk which are about $11. However, I hear TDK blank disk are better and they cost about $7.50 a piece. The CD that I am making with the Pinnicle do have problems. Some CD-ROMs will not read them. It is not a problem for us, because all the CD we make work on our CD-ROMs on our Window NT system. However, there have been some people reporting that their DOS machines with a CD-ROM wil not see the disk in the drive. Kelly Dean CSU/CIRA From fitsbits-request Fri Aug 11 12:47:55 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1828" "Wed" "9" "August" "1995" "09:29:21" "GMT" "Derrick C. Mancini" "mancini at maxlab.lu.se" "" "34" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080909:29:21" "Saving 100GB of data for 15 years cheaply" (number " " mark " Derrick C. Mancin Aug 9 34/1828 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<406st6$8le at gwis2.circ.gwu.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA19005; Fri, 11 Aug 1995 12:47:55 -0400 Return-Path: Message-Id: Organization: MAXLab (Lund University) & APS (Argonne Nat. Lab.) Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!sunic!sunic.sunet.se!news.lth.se!news.lu.se!usx121.maxlab.lu.se!user References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1826 From: mancini at maxlab.lu.se (Derrick C. Mancini) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: Wed, 9 Aug 1995 09:29:21 GMT In article <406st6$8le at gwis2.circ.gwu.edu>, sfearing at gwis2.circ.gwu.edu (Scott Fearing) wrote: > I should warn you of a potential danger of CD storage: CD 'rot'. >Apparently, the aluminum can oxidize under the plastic protective covering >and data can be lost. This is appearing on CDs as well as laserdiscs, >but I assume it does not happen to the special gold plated discs that come >out on the audiophile pressings. There are discussions of this from time >to time on rec.video.laserdisc. I believe, Scott, that your comments may be only applicable to pressed CDs. The writable CDs are a MO technology, I think. I am not sure what they use for reflective coat. In properly manufactured CDs, the aluminum should last for 15 years. It is the presence of water and salts that cause the damage. I know I have never seen this effect on anything in my CD collection, which started when CD's started, about 12 years (and a few hundred CDs) ago. By the same token, poor manufacture and improper storage of tape is far more likely to fail than aluminum in CDs. In either case, you would be advised to store in a cool, dry place. And make duplicates. Another advantage of CDs is that they are a random access storage device, which most tapes are not, and should have better ability to recover good files if a particular sector or track were to go bad. Many tape systems will lose the whole tail of a tape from the point a defect is found. -- Dr. Derrick C. Mancini MAXLab, Lund University Advance Photon Source, ANL Box 118, Ole Romers Vag 1 9700 S. Cass Avenue - APS/431 S-221 00, Lund, Sweden Argonne, IL, USA 60439 Tel: +46 46 222 3355 Tel: +1 708 252 0147 FAX: +46 46 222 4710 FAX: +1 708 252 3222 Email: mancini at maxlab.lu.se Email: mancini at aps.anl.gov From fitsbits-request Fri Aug 11 19:42:53 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1397" "Wed" "9" "August" "1995" "06:59:46" "GMT" "James Petts" "pettsj at visigoth.demon.co.uk" "" "31" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080906:59:46" "Saving 100GB of data for 15 years cheaply" (number " " mark " James Petts Aug 9 31/1397 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA20383; Fri, 11 Aug 1995 19:42:53 -0400 Return-Path: Message-Id: Organization: Huh? Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!dispatch.news.demon.net!demon!visigoth.demon.co.uk!user References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> <407tut$21e at electron.rutgers.edu> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1395 From: pettsj at visigoth.demon.co.uk (James Petts) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: Wed, 09 Aug 1995 06:59:46 GMT In article , grunes at news.nrl.navy.mil (Mitchell R Grunes) wrote: > 3. Since no one liked my idea of film, I've an even better idea. > Print it all out in hex, on acid free paper. When you need it, use a > scanner. Let's see: two characters/byte, about 100 characters/line > and 60 lines/page...now that's a REAL fire hazard. Compact, too. I came into this discussion late (i.e. this is the first message on the subject I've seen), but what was the main objection to film. If it is the stability of the film base, that can almost certainly be addressed by storing the film in vacuum containers. If it is image permanence that can be addressed by using metallic replacement in the developed film, substituting gold for silver. Both techniques have been used at Kodak Limited in England: when Domesday Book was taken apart and rebound in 1985-1986 it was photographed and the film treated as mentioned above. Accelerated ageing test show that the film is expected to retain its image in essentially he same form for at least 100 years. I can speak with the person who did it, and get more information if required. -- === James Petts === In the world of CD-ROM, a horse will always be an herbivorous quadruped, and there'll be endless examples to prove just that. But that world will never tell you how it feels to ride a horse. Or fall off it. From fitsbits-request Fri Aug 11 19:43:36 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["728" "" "9" "August" "1995" "20:45:57" "GMT" "Mark Overton" "mark at sdd.hp.com" "<40b6q5$59f at news.sdd.hp.com>" "20" "Re: QIC Tape Lifetimes? was: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995080920:45:57" "QIC Tape Lifetimes? was: Saving 100GB of data for 15 years cheaply" (number " " mark " Mark Overton Aug 9 20/728 " thread-indent "\"Re: QIC Tape Lifetimes? was: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA20392; Fri, 11 Aug 1995 19:43:36 -0400 Return-Path: Message-Id: <40b6q5$59f at news.sdd.hp.com> Organization: Hewlett-Packard, San Diego Division Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!howland.reston.ans.net!swrinde!sdd.hp.com!nobody References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> Newsgroups: sci.data.formats,sci.image.processing,sci.astro.fits,comp.sys.ibm.pc.hardware.storage content-length: 726 From: mark at sdd.hp.com (Mark Overton) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: QIC Tape Lifetimes? was: Saving 100GB of data for 15 years cheaply Date: 9 Aug 1995 20:45:57 GMT grunes at news.nrl.navy.mil (Mitchell R Grunes) writes: > Also, does one extend the # of uses (usually rated at 3-5 back-ups, due > to the large number of passes required to handle multiple tracks with > one head) of these tapes if they are commercially pre-formatted? Only 3-5 backups?! I asked a Colorado tech support person about this, who said that a tape will last a year if used (weekly or daily or in rotation, I forget). Our tapes are lasting a year, being used twice a week (ie, around 100 backups). - Mark ------------------------------------------------------------------------------ Mark Overton, Hewlett-Packard (San Diego Division), mark at sdd.hp.com "A hundred thousand lemmings can't be wrong." ... Unknown From fitsbits-request Sat Aug 12 01:42:08 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["899" "" "9" "August" "1995" "21:28:33" "-0400" "Mark Edward Schill" "gt7602b at prism.gatech.edu" "<40bnc1$fia at acmex.gatech.edu>" "20" "looking for fits viewer" "^From:" nil nil "8" "1995081001:28:33" "looking for fits viewer" (number " " mark " Mark Edward Schil Aug 9 20/899 " thread-indent "\"looking for fits viewer\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA20785; Sat, 12 Aug 1995 01:42:08 -0400 Return-Path: Message-Id: <40bnc1$fia at acmex.gatech.edu> Organization: Georgia Institute of Technology Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!tank.news.pipex.net!pipex!howland.reston.ans.net!swrinde!gatech!gt-news!prism!prism!not-for-mail Newsgroups: sci.astro.fits content-length: 897 From: gt7602b at prism.gatech.edu (Mark Edward Schill) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: looking for fits viewer Date: 9 Aug 1995 21:28:33 -0400 I am looking for a program that will allow me to view fits images. I would like a program that would allow me to view the image and the header silmutaneously. I would also like to have the option of saving the image in either gif or jpg format so that the images can be readily displayed. A program that would allow me to convert or create fits images with interactive prompts or a template to facilitate the header information would be ideal. I would like a program that will run on Solaris Openwindows, but either Ibm or Macintosh programs will be gladly accepted. I appreciate any help that is given. -- | Mark E. Schill |" Imagination is more | | Email: gt7602b at prism.gatech.edu | important than knowledge | | |_______________- A.Einstein | | WWW: http://kelvin.physics.gatech.edu/sps/mschill |______________| From fitsbits-request Sat Aug 12 14:29:41 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1190" "" "10" "August" "1995" "10:42:24" "GMT" "Ralph Martin" "ralf at mail.ast.cam.ac.uk" "<40cnqg$p7m at lyra.csx.cam.ac.uk>" "32" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995081010:42:24" "Saving 100GB of data for 15 years cheaply" (number " " mark " Ralph Martin Aug 10 32/1190 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<40als0$f6k at news.ccit.arizona.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA21164; Sat, 12 Aug 1995 14:29:41 -0400 Return-Path: Message-Id: <40cnqg$p7m at lyra.csx.cam.ac.uk> Organization: Institute of Astronomy, Cambridge Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!tank.news.pipex.net!pipex!lyra.csx.cam.ac.uk!ralf References: <40als0$f6k at news.ccit.arizona.edu> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1188 From: ralf at mail.ast.cam.ac.uk (Ralph Martin) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 10 Aug 1995 10:42:24 GMT I checked back in my records for systems that store terabytes of data. (1 terabyte =1000 Gigabytes = 1,000,000 Megabytes). Optical Memory News, February 1991, Published Monthly by Rothchild Consultants, ran an article about a device marketed by ICI Imagedata which wrote to optical tape. In it they described a device made by a firm called CREO which had the following capabilities. 1 Terabyte capacity per 12" reel. Cost ' a few dollars per gigabyte' Transfer Rate up to 3 Megs/sec sustained. Media Life 30 years in office environment SCSI-2 interface Search speed 68 sec max for full reel. The last contact I had with them was some years ago but the address of the suppliers was ICI Imagedata PO BOX 6 Shire Park Bessemer Road Welwyn Garden City Hertfordshire AL7 1HD Telephone 01707 323400 Fax 01707 337896. I have no connection either business or private with the afore mentioned firm. -- Ralph Martin Internet: ralf at ast.cam.ac.uk Royal Greenwich Observatory Telephone: UK+ 01223-374000 Cambridge CB3 0EZ ISDN: UK+ 01223-328028 United Kingdom ISDN: UK+ 01223-328087 From fitsbits-request Sat Aug 12 18:28:55 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1518" "" "10" "August" "1995" "04:37:53" "GMT" "Rob Seaman" "seaman at noao.edu" "<40c2f1$6rp at noao.edu>" "33" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995081004:37:53" "Saving 100GB of data for 15 years cheaply" (number " " mark " Rob Seaman Aug 10 33/1518 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA21337; Sat, 12 Aug 1995 18:28:55 -0400 Return-Path: Message-Id: <40c2f1$6rp at noao.edu> Organization: National Optical Astronomy Observatories Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!howland.reston.ans.net!vixen.cso.uiuc.edu!uchinews!ncar!noao!caliche.tuc.noao.edu!seaman References: <3vtejh$mg at nntp.ucs.ubc.ca> <408h39$m8p at sol.sun.csd.unb.ca> <409gdi$2qrq at yuma.ACNS.ColoState.EDU> Newsgroups: sci.astro.fits content-length: 1516 From: Rob Seaman Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 10 Aug 1995 04:37:53 GMT In article <408a2a$h50 at darkstar.UCSC.EDU> sla at umbra.ucolick.org, Steve Allen writes: > But in the instance of non-IMAGE extensions how would it be possible > to determine the relationships between the various different kinds of > HDUs? I suspect that this would be an excellent application for > the > FITS Grouping Convention which was proposed last year. Well, this applies to IMAGE extensions as well as other extensions. One difficulty with archiving raw data is that not all members of a given group are defined - typically not even acquired yet - when an image is archived. Archiving the group interrelationships as a separate FITS object at some later time is attractive - although this will complement, not replace, a catalog database. The grouping convention relies on URLs (or URNs) to refer to group members that are external to the current file. It isn't obvious (to me, anyway) how best to construct a URL that references a tape sitting on a shelf. Examples, anybody? Note also that since the physical media will be replaced periodically, and since a new medium will likely have different constraints on size, the arrangement of the data into files and HDUs in those files is likely to change. It may be tricky to define groupings whose URLs (or whatever) will remain valid through a recasting of the data. This would be preferable to discarding and then regenerating the grouping extensions, however. Rob Seaman seaman at noao.edu From fitsbits-request Sun Aug 13 05:43:47 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["766" "" "10" "August" "1995" "16:39:33" "-0400" "Albert Hybl" "hybl at umbc.edu" "<40dqq5$20j at umbc7.umbc.edu>" "19" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995081020:39:33" "Saving 100GB of data for 15 years cheaply" (number " " mark " Albert Hybl Aug 10 19/766 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA21880; Sun, 13 Aug 1995 05:43:47 -0400 Return-Path: Message-Id: <40dqq5$20j at umbc7.umbc.edu> Organization: University of Maryland, Baltimore County Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!newsfeed.internetmci.com!news.msfc.nasa.gov!news.larc.nasa.gov!lerc.nasa.gov!purdue!haven.umd.edu!news.umbc.edu!not-for-mail Newsgroups: sci.astro.fits content-length: 764 From: hybl at umbc.edu (Dr. Albert Hybl) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 10 Aug 1995 16:39:33 -0400 Recently, a TV station reported that Dr. Uzasny Blazen, CEO of The Black Hole Storage Devices Inc., announced that they are about to start shipping a new device that promises to provide a near infinite storage capacity. A reporter and a technician have been dispatched with a mobile satellite dish to obtain an interview with him. Their report had not arrived in time for the 6 pm news. Hopefully, it will be available for the 11 pm news. In related news, an airline pilot reported seeing a building collapse into a huge subterranean chasm. The cavern is located exactly where Dr. Blazen's factory should be. The last report broadcast from the station's traffic helicopter estimated that the crater is expanding at about five feet per hour. :-) Albert From fitsbits-request Sun Aug 13 07:19:52 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["6823" "" "11" "August" "1995" "08:50" "EDT" "Barry M. Schlesinger" "bschlesinger at nssdca.gsfc.nasa.gov" "<11AUG199508503853 at nssdca.gsfc.nasa.gov>" "163" "Sources of FITS Information" "^From:" nil nil "8" "1995081112:50:00" "Sources of FITS Information" (number " " mark " Barry M. Schlesin Aug 11 163/6823 " thread-indent "\"Sources of FITS Information\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA21917; Sun, 13 Aug 1995 07:19:52 -0400 Return-Path: Message-Id: <11AUG199508503853 at nssdca.gsfc.nasa.gov> Organization: NASA Goddard Space Flight Center -- Greenbelt, Maryland USA Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!newsfeed.internetmci.com!news.dacom.co.kr!news.kreonet.re.kr!usenet.!ames!newsfeed.gsfc.nasa.gov!nssdca.gsfc.nasa.gov!bschlesinger Reply-To: fits at nssdca.gsfc.nasa.gov Newsgroups: sci.astro.fits,sci.data.formats content-length: 6821 From: bschlesinger at nssdca.gsfc.nasa.gov (Barry M. Schlesinger) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Sources of FITS Information Date: 11 Aug 1995 08:50 EDT Sources of FITS Information Preface This material on sources of Flexible Image Transport System (FITS) information is posted and updated periodically by the FITS Support Office at the NASA Goddard Space Flight Center (GSFC). It discusses where general FITS information, including some answers to frequently asked questions, can be found, and provides sources for detailed information on FITS software and documentation. Discussions are under way for addition of this material to the *.answers hierarchy; further developments are expected in about a month. It is posted here in the meantime. Followups are to sci.astro.fits. FITS Support Office The FITS Support Office maintains a library of FITS information accessible from http://www.gsfc.nasa.gov/astro.fits/fits_home.html or ftp://nssdc.gsfc.nasa.gov/pub/fits/. The material available includes o "Definition of FITS," a codification of FITS for the NASA/Science Office of Standards and Technology (NOST), available in LaTeX, PostScript, and ASCII text o "A User's Guide to FITS", published by the FITS Support Office, in LaTex, and compressed and uncompressed PostScript o Proposed revisions to the "Definition of FITS"covering the specification of units (text) o A current list of the extension type (structure) names registered with the International Astronomical Union FITS Working Group (IAUFWG) (text) o Rules for physical blocking on various media adopted by the IAUFWG, (text) In the same directory, but accessible directly via http://www.gsfc.nasa.gov/astro/fits/basics_info.html is the FITS Basics and Information that used to be regularly posted to sci.astro.fits and sci.data.formats under the heading of FITS Basics and Information. It continues to be revised to reflect current FITS developments. It contains the following material: o An overview of FITS o A list of FITS documents o A list of software packages that support FITS, including FITS-image converters for various platforms o A list of on-line FITS resources o A description of the FITS Support Office The hypertext version provides links to many of the documents, software, and network locations listed. The text version provides information on how to obtain some of this material. Links from the Web page and subdirectories of the ftp directory contain o Software developed by the FITS Support Office. o Error test files, primary HDUs useful for testing the ability of software designed to read FITS files to cope with files that have errors or are non-standard. Printed copies of the material in the FITS directory can be obtained from the Coordinated Request and User Support Office (CRUSO): (Postal) Coordinated Request and User Support Office Code 633 National Space Science Data Center NASA Goddard Space Flight Center Greenbelt MD 20771 USA (Electronic mail) request at nssdca.gsfc.nasa.gov (Telephone) +1-301-286-6695 8:00 A. M. - 4:30 P.M. U. S. Eastern Time (-0500 from the last Sunday in October through the first Saturday in April; -0400 the remainder of the year) When no one is available, messages can be left on voice mail. (FAX) +1-301-286-1635 National Radio Astronomy Observatory (NRAO) A FITS Archive can be found at URL http://fits.cv.nrao.edu/ or at ftp://fits.cv.nrao.edu/fits, located at NRAO. This machine supports a WAIS server named nrao-fits which has an index of all of the FITS-related text files in the archive; the file nrao-fits.src is available at think.com and at ftp://fits.cv.nrao.edu/fits/wais-sources/nrao-fits.src. The documents subdirectory of the fits directory contains a number of subdirectories. The BINTABLE draft for Astronomy and Astrophysics is, in various formats, in files bintable_aa.*. A proposals subdirectory is reserved for detailed proposals currently being considered by the FITS committees. A drafts subdirectory contains drafts of designs not yet submitted. The wcs subdirectory contains a draft of the current proposal for world coordinate system conventions now under community review and earlier documents and presentations on world coordinates. Other subdirectories include sample FITS files -- both actual data files and files specially constructed to test the ability of software to read all kinds of FITS structures, some code for particular environments, pointers to other code, and an archive of Usenet postings related to FITS. This machine supports a WAIS server named nrao-fits (URL wais://fits.cv.nrao.edu/nrao-fits; also see ftp://fits.cv.nrao.edu/fits/wais-sources/nrao-fits.src) which has an index of all of the FITS-related text files in the archive. Some of the more noteworthy materials in this archive are o Drafts of proposed additions to the FITS standard and other drafts that may in the future be formally proposed o Conventions specific to particular projects or disciplines o Some code for various environments and Usenet postings about code o Sample data and special test files designed to measure the ability of a FITS reader to handle a wide variety of FITS files o Archives of traffic on FITS-related newsgroups and exploders HEASARC The NASA/Goddard High Energy Astrophysics Science Archive Research Center (HEASARC) Web server at http://heasarc.gsfc.nasa.gov/docs/heasarc/fits.html and the anonymous ftp access through ftp://heasarc.gsfc.nasa.gov/fits_info/ provide FITS material. HEASARC has developed a the FITSIO package of software routines for easily reading and writing FITS files, in FORTRAN with a C interface available, portable to a wide variety of machines. There is also the FTOOLS collection of software tools and the VERIFITS FITS conformance verifier. HEASARC software is available directly through http://heasarc.gsfc.nasa.gov/docs/heasarc/tech_res_software.html or ftp://heasarc.gsfc.nasa.gov/fits_info/software/ . The HEASARC server also provides information from the OGIP/HEASARC FITS Working Group, (HFWG) the internal legislative body on FITS-related matters within the Office of Guest Investigator Programs (OGIP) at NASA/GSFC, at http://heasarc.gsfc.nasa.gov/heasarc/ofwg/ofwg_intro.html or ftp://heasarc.gsfc.nasa.gov/fits_info/ . The HFWG has developed a number of FITS conventions that are more specific than the requirements of the FITS standards. Proposed conventions are publicized to the FITS community as a whole, with the goal of collaborative development of a set of conventions that will be accepted throughout the community as well as within OGIP/HEASARC. Direct questions about this material to Barry M. Schlesinger Coordinator, FITS Support Office Electronic mail: fits at nssdca.gsfc.nasa.gov Telephone: +1-301-441-4205 The FITS Support Office is operated under the guidance of the NASA/GSFC Astrophysics Data Facility. From fitsbits-request Sun Aug 13 10:03:53 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["3293" "" "10" "August" "1995" "13:53:08" "GMT" "Art Munson" "artmuns at tape.com" "<40d304$hd2 at excalibur.edge.net>" "86" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995081013:53:08" "Saving 100GB of data for 15 years cheaply" (number " " mark " Art Munson Aug 10 86/3293 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<409gdi$2qrq at yuma.ACNS.ColoState.EDU>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA22028; Sun, 13 Aug 1995 10:03:53 -0400 Return-Path: Message-Id: <40d304$hd2 at excalibur.edge.net> Organization: The Edge Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.mathworks.com!news.ultranet.com!news.sprintlink.net!news.edge.net!news References: <3vtejh$mg at nntp.ucs.ubc.ca> <409gdi$2qrq at yuma.ACNS.ColoState.EDU> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 3291 From: Art Munson Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 10 Aug 1995 13:53:08 GMT If anyone would be interested in our latest prices on TDK CDR blanks, just let me know and I'll e-mail them to you. I'm also posting a recent fax from TDK regarding longevity. Thanks Art Cassette House http://www.tape.com/ch dean at phobos.cira.colostate.edu wrote: > We use 3M blank disk which are > about $11. However, I hear TDK blank disk are better and they cost > about > $7.50 a piece. > ------------------------------------------------------------------------- T-TEC TDK-TECHNICAL EXCELLENCE CORNER Topics: Mix Magazine Article Compiled by Tom Izumi & Henry Dobashi Number 4 - July, 1995 In MIX magazine (July 1995), an article purportedly comparing cyanine to phthalocyanine-based CD-R by Steven St. Croix contained injurious comments about cyanine. Here is a summery of our reactions. What Mr. St. Croix says: He speaks ill of many things about cyanine-based CD-R, but most of his comments center on the liefspan of cyanine-based CD-Rs. He leads readers to believe that cyanine CD-Rs are veritable time bombs, designed to self-destruct in as little as five years, one year or even over the weekend. The most puzzling aspect of the article is that his conclusions are based on secret data -- research that he refuses to make public. TDK'S Reaction: We are afraid that the readers may have serious doubts about the longevity of our media. It is our responsibility to rebut. You will see a two-page open letter in MIX, October issue, showing our assertions as follows. Discussions in our open letter: We believe the writer may have based many of his conclusions on data describing the performance of raw cyanine dye which TDK does not use. We will publish data that demonstrates a big improvement in light resistivity, when metal- stabilized cyanine dye is used. We will share our aging test data, too. The results demonstrate our CD-R lasts at least 70 years when stored at 30 degrees C (86 degrees F). We will also present our test results comparing the light fastness of cyanine and phthalocyanine discs. Even after 40 Mlux*h exposure under Xenon test lamp, every parameter stays within spec for both disc types. This test indicates that both types of discs will last 10 years at least when left beside the windows with their dye side up. According to the test results mentioned above, phthalocyanine is slightly more resistant to light than our metal-stabilized cyanine formulation. That's why phthalocyanine discs are specified at 100 years; ours at 70 years. So why use cyanine? The answer is compatibility with CD recorders and long-term writability. A graph will be published showing block error rate (BLER) vs. write power, which compares our CD-R with a typical phthalocyanine disc. Our disc maintains its low BLER over a broader range of laser power than phthalocyanine media. This is significant because factory-fresh CD recorders may generate higher or lower power output than the nominal manufacturer spec. Also, the laser ages over time and operation temperature may fluctuate. These factors may cause recording power to fluctuate. Compatibility and long-term writability are the strongest and most rational arguments for use of cyanine dye. These arguments are made very strongly in our open letter to MIX. From fitsbits-request Sun Aug 13 23:20:57 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1935" "" "10" "August" "1995" "17:25:09" "GMT" "Glen Kilpatrick" "szkilpat at rocky.ucdavis.edu" "<40dfdl$67l at mark.ucdavis.edu>" "34" "Re: looking for fits viewer" "^From:" nil nil "8" "1995081017:25:09" "looking for fits viewer" (number " " mark " Glen Kilpatrick Aug 10 34/1935 " thread-indent "\"Re: looking for fits viewer\"\n") "<40bnc1$fia at acmex.gatech.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA22492; Sun, 13 Aug 1995 23:20:57 -0400 Return-Path: Message-Id: <40dfdl$67l at mark.ucdavis.edu> Organization: University of California, Davis Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.mathworks.com!tank.news.pipex.net!pipex!swrinde!sgigate.sgi.com!genmagic!bug.rahul.net!a2i!rahul.net!a2i!ccnet.com!news.cerf.net!newsserver.sdsc.edu!nic-nac.CSU.net!charnel.ecst.csuchico.edu!csusac!csus.edu!news.ucdavis.edu!rocky!szkilpat References: <40bnc1$fia at acmex.gatech.edu> Newsgroups: sci.astro.fits content-length: 1933 From: szkilpat at rocky.ucdavis.edu (Glen Kilpatrick) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: looking for fits viewer Date: 10 Aug 1995 17:25:09 GMT Mark Edward Schill (gt7602b at prism.gatech.edu) wrote: : I am looking for a program that will allow me to view fits images. : I would like a program that would allow me to view the image and the : header silmutaneously. I would also like to have the option of saving the : image in either gif or jpg format so that the images can be readily : displayed. A program that would allow me to convert or create fits images : with interactive prompts or a template to facilitate the header : information would be ideal. I would like a program that will run on : Solaris Openwindows, but either Ibm or Macintosh programs will be gladly : accepted. : I appreciate any help that is given. Try ftp://fits.cv.nrao.edu/fits/os-support/ms-windows/fitsview/ for a WINdoze program. It doesn't do everything you want (but then -what- does everything -I- want? :), but it's a start. Be warned that, while I have no problems using the program, I sweated plenty when the installation program -disabled- WIN 3.1 for me. I suspect the problem was a matter of too many program groups, but -whatever-, the solution to re-enabling Windows was to find progman.ini, delete it, and rename progman.bak (.bac?) to the former progman.ini, and then WIN worked. The second attempt at installation (with fewer program groups :) worked fine, and I'm now a Happy Camper. -- --------------------------------------------------------------------- ___ | [A-Z] is for { Amendment Black Clipper / \ | Decryption Encryption Factor Gore Hacker | ____ | ITAR Judicial Kafkaesque Lincoln Munitions \___/ | / Glen Kilpatrick | Numerical Ossifrage Privacy Q-Clearance RSA | \ | White XOR Yankee Zimmermann }, {:-(.... PGP Fingerprint Real Soon Now! --------------------------------------- From fitsbits-request Tue Aug 15 00:25:48 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["532" "" "14" "August" "1995" "16:48:03" "GMT" "Richard Ottolini" "stgprao at sugarland.unocal.com" "<40nuo3$1pa at unogate.unocal.com>" "9" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995081416:48:03" "Saving 100GB of data for 15 years cheaply" (number " " mark " Richard Ottolini Aug 14 9/532 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") "<404t6a$fme at lyra.csx.cam.ac.uk>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA25155; Tue, 15 Aug 1995 00:25:48 -0400 Return-Path: Message-Id: <40nuo3$1pa at unogate.unocal.com> Organization: Unocal Corporation Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!howland.reston.ans.net!usc!news.service.uci.edu!news.intelenet.com!unogate!stgprao References: <3vtejh$mg at nntp.ucs.ubc.ca> <404l2e$40v at anarchy.io.com> <404t6a$fme at lyra.csx.cam.ac.uk> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 530 From: stgprao at sugarland.unocal.COM (Richard Ottolini) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 14 Aug 1995 16:48:03 GMT There was a nice review of the issue of digital archives in Scientific American about a year ago. Cost is not so much the issue as is recoverable formats in a rapidly change technology. If things aren't copied into the new formats every few years, they could be lost. Much like if a classical work wasn't recopied by a medieval monk or Arab classicist, most books distintigrated or were lost. Much of early NASA planetary data is "lost" in the old formats. And this doesn't even address the issue of physical media lifetimes. From fitsbits-request Tue Aug 15 02:11:01 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["621" "" "14" "August" "1995" "18:35:08" "GMT" "John Atwood" "atwoodj at ada.cs.orst.edu" "<40o50s$mej at engr.orst.edu>" "14" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995081418:35:08" "Saving 100GB of data for 15 years cheaply" (number " " mark " John Atwood Aug 14 14/621 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA25232; Tue, 15 Aug 1995 02:11:01 -0400 Return-Path: Message-Id: <40o50s$mej at engr.orst.edu> Organization: Computer Science Department, Oregon State University Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.mathworks.com!usenet.eel.ufl.edu!col.hp.com!sdd.hp.com!hp-pcd!hp-cv!reuter.cse.ogi.edu!engr.orst.edu!ada.CS.ORST.EDU!atwoodj References: Newsgroups: sci.astro.fits,sci.data.formats,sci.astro,sci.physics.accelerators,sci.image.processing,sci.image.processing content-length: 619 From: atwoodj at ada.CS.ORST.EDU (John Atwood) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 14 Aug 1995 18:35:08 GMT I'm surprised no one has mentioned Optical Tape. The August issue of IEEE Spectrum shows it as a medium with good characteristics. The article says Sony, Philips, Storage Technonlogy, E-Systems EMAS, and Kodak are "actively investigating optical tape". One graph shows the capacity of optical tape to be, in 1993, 27GB, in 1995, 50GB, in 1999, 100GB. John -- _________________________________________________________________ Office phone: 503-737-5583 (Batcheller 349);home: 503-757-8772 Office mail: 303 Dearborn Hall, OSU, Corvallis, OR 97331 _________________________________________________________________ From fitsbits-request Tue Aug 15 20:48:04 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["383" "Tue" "15" "August" "1995" "07:50:33" "GMT" "Tim Atkinson" "Tim at npagroup.demon.co.uk" "<808473033snz at npagroup.demon.co.uk>" "10" "Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995081507:50:33" "Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply" (number " " mark " Tim Atkinson Aug 15 10/383 " thread-indent "\"Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA26357; Tue, 15 Aug 1995 20:48:04 -0400 Return-Path: Message-Id: <808473033snz at npagroup.demon.co.uk> Organization: NPA Group Ltd. Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.sprintlink.net!in2.uu.net!tank.news.pipex.net!pipex!dispatch.news.demon.net!demon!mail2news.demon.co.uk!npagroup.demon.co.uk!Tim References: Reply-To: Tim at npagroup.demon.co.uk Newsgroups: sci.astro.fits,sci.data.formats,sci.astro,sci.physics.accelerators,sci.image.processing,sci.image.processing content-length: 381 From: Tim Atkinson Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply Date: Tue, 15 Aug 95 07:50:33 GMT I upgraded to a Yamaha 4 speed writer from a Phillips CDD521 (original) and found that hardly anything would read the disks the Yamaha produced. Reducing the Yamaha write speed to double speed from quad speed solved this problem mostly, for some really old pre-multisession players, you still have to write disks at single speed. Don't know why but it works. -- Tim Atkinson From fitsbits-request Wed Aug 16 04:15:34 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1671" "Mon" "14" "August" "1995" "16:19:47" "" "A. P. R. Cooper" "aprc at pcmail.nerc-bas.ac.uk" "" "39" "Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data" "^From:" nil nil "8" "1995081416:19:47" "Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data" (number " " mark " A. P. R. Cooper Aug 14 39/1671 " thread-indent "\"Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA26785; Wed, 16 Aug 1995 04:15:34 -0400 Return-Path: Message-Id: Organization: British Antarctic Survey Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!newsfeed.internetmci.com!news.sprintlink.net!dispatch.news.demon.net!demon!uknet!newsfeed.ed.ac.uk!nerc-keyworth!aprc.nerc-bas.ac.uk!aprc References: Newsgroups: sci.astro.fits,sci.data.formats,sci.astro,sci.physics.accelerators,sci.image.processing,sci.image.processing content-length: 1669 From: aprc at pcmail.nerc-bas.ac.uk (A P R Cooper) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data Date: Mon, 14 Aug 1995 16:19:47 In article Jeff Bloch writes: >Subject: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data >From: Jeff Bloch >Date: Sun, 13 Aug 1995 12:46:03 -0600 >We store 650MB of satellite data on a CD ROM every 5 days using a setup from >Young Minds on a Sparcstation. We have over 100 CD volumes in the archive. >All has been well for the last two years until Sun started shipping new 4x CD >ROM readers from Toshiba with new workstations. Suddenly, 50% of our archive >is unreadable by the new drives. Old drives (mostly Sony) continue to read >the archive just fine. This has nothing to do with the ISSO standards. Young >Minds has not quite fessed up to the problem yet, but there may be a hardware >problem with the particular model of the Phillips writer that we have. Surely we're not talking about CD but CD-R here? They are different technologies, even though they are compatible. >-------------------------------------------------------------------------- >Jeffrey Bloch Office: (505) 665-2568 >Astrophysics and Radiation Measurements Group ALEXIS Soc: (505) 665-5975 >Los Alamos National Laboratory FAX: (505) 665-4414 >Group NIS-2, Mail Stop D436 e-mail: jbloch at lanl.gov >Los Alamos, NM 87545 >-------------------------------------------------------------------------- A. Paul R. Cooper British Antarctic Survey aprc at pcmail.nerc-bas.ac.uk ***** Any opinions given here are my own, ***** ***** and not necessarily those of British Antarctic Survey ***** From fitsbits-request Wed Aug 16 12:12:24 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["2212" "" "16" "August" "1995" "08:32:48" "+0800" "David Brooks" "daveb at perth.dialix.oz.au" "<40rebg$di2$1 at perth.DIALix.oz.au>" "49" "Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data" "^From:" nil nil "8" "1995081600:32:48" "Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data" (number " " mark " David Brooks Aug 16 49/2212 " thread-indent "\"Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA27273; Wed, 16 Aug 1995 12:12:24 -0400 Return-Path: Message-Id: <40rebg$di2$1 at perth.DIALix.oz.au> Organization: DIALix Services, Perth, Australia. Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!sol.ctr.columbia.edu!news.uoregon.edu!newsfeed.internetmci.com!news.sprintlink.net!simtel!harbinger.cc.monash.edu.au!news.uwa.edu.au!DIALix!not-for-mail References: Newsgroups: sci.astro.fits,sci.data.formats,sci.astro,sci.physics.accelerators,sci.image.processing,sci.image.processing content-length: 2210 From: daveb at perth.DIALix.oz.au (David Brooks) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data Date: 16 Aug 1995 08:32:48 +0800 aprc at pcmail.nerc-bas.ac.uk (A P R Cooper) writes: >In article Jeff Bloch writes: >>Subject: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data >>From: Jeff Bloch >>Date: Sun, 13 Aug 1995 12:46:03 -0600 >>We store 650MB of satellite data on a CD ROM every 5 days using a setup from >>Young Minds on a Sparcstation. We have over 100 CD volumes in the archive. >>All has been well for the last two years until Sun started shipping new 4x CD >>ROM readers from Toshiba with new workstations. Suddenly, 50% of our archive >>is unreadable by the new drives. Old drives (mostly Sony) continue to read >>the archive just fine. This has nothing to do with the ISSO standards. Young >>Minds has not quite fessed up to the problem yet, but there may be a hardware >>problem with the particular model of the Phillips writer that we have. >Surely we're not talking about CD but CD-R here? They are different >technologies, even though they are compatible. >>-------------------------------------------------------------------------- >>Jeffrey Bloch Office: (505) 665-2568 >>Astrophysics and Radiation Measurements Group ALEXIS Soc: (505) 665-5975 >>Los Alamos National Laboratory FAX: (505) 665-4414 >>Group NIS-2, Mail Stop D436 e-mail: jbloch at lanl.gov >>Los Alamos, NM 87545 >>-------------------------------------------------------------------------- >A. Paul R. Cooper >British Antarctic Survey >aprc at pcmail.nerc-bas.ac.uk >***** Any opinions given here are my own, ***** >***** and not necessarily those of British Antarctic Survey ***** I have encountered this problem also. A friend recently loaned me an archive on a Kodak CD-R. I found I could read it on my old 486 (2x drive), but not on the Pentium (4x drive). The old drive is a Panasonic CR-562-B, the newer one came with a Sound Balster Discovery kit. -- David R. Brooks Tel/fax. +61 9 434 4280 PGP public key via finger or keyservers Public key fingerprint: 20 8F 95 22 96 D6 1C 0B 3D 4D C3 D4 50 A1 C4 34 From fitsbits-request Thu Aug 17 03:08:34 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["7158" "" "15" "August" "1995" "16:36:26" "GMT" "Perry Greenfield" "perry at eclipse.stsci.edu" "<40qiea$bl2 at marvel.stsci.edu>" "141" "FITS Image Extension support in IRAF" "^From:" nil nil "8" "1995081516:36:26" "FITS Image Extension support in IRAF" (number " " mark " Perry Greenfield Aug 15 141/7158 " thread-indent "\"FITS Image Extension support in IRAF\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA28311; Thu, 17 Aug 1995 03:08:34 -0400 Return-Path: Message-Id: <40qiea$bl2 at marvel.stsci.edu> Organization: Space Telescope Science Institute Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!ncar!noao!stsci!usenet Newsgroups: sci.astro.fits content-length: 7156 From: perry at eclipse.stsci.edu (Perry Greenfield) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: FITS Image Extension support in IRAF Date: 15 Aug 1995 16:36:26 GMT The Science Software Group (i.e., the STSDAS group) at the Space Telescope Science Institute has been involved in a cooperative effort with NOAO to modify the IRAF imio kernel to support the FITS Image Extensions within IRAF as a native disk format. We have tentatively settled on a specification for how we see the interface working and have been implementing a kernel to this specification. The implementation we come up with may be modified as a result of comments from NOAO or may later be enhanced or modified by NOAO before distribution as part of IRAF v2.11. This posting is intended to serve the purpose of notifying the FITS community about our intended use and interpretation of the FITS format as well as solicit comments about its compliance with the standard. Before discussing the details, some brief background. The calibration pipelines being developed for the two new instruments being installed during the 97 servicing mission, STIS and NICMOS, will produce significantly more datasets than are now being produced for the existing instruments. This is a result of generating associated data arrays for all science images that will contain data quality and variance information. Furthermore, both instruments have modes which produce multiple images for each observation (possibly more than 50 for NICMOS). FITS Image Extensions have been chosen as a means of reducing the number of file generated by combining all the associated arrays for an observation into one file. The following issues have arisen in the design of the data files and the ease in which a user may access an image contained in such a file. I) "INHERITANCE" There are two aspects of the interface we are developing that relate to issues with the FITS standard. The first deals with relation of the keywords in the primary header to the extension images that follow in the same file. This particular aspect engendered much discussion and consideration. There is an intuitive expectation among many that the Primary Header keywords should apply to all extension headers, but in fact it seems that most FITS software makes no such assumptions, i.e., all extension headers are treated as independent of the Primary Header. The FITS standard makes no explicit reference to this issue. After much discussion it was decided to include a mechanism for Extension Header Units (here- after EHU) to "inherit" Primary Header Unit (PHU) keywords. The scheme being adopted is that the EHU will inherit the PHU keywords if it contains a line specifying "INHERIT=T". In the absence of such a line (or if "INHERIT=F" exists) no inheritance of PHU keywords will take place. In cases where the PHU and EHU both contain the same keyword, the EHU value takes precedence. If the EHU is propagated by the applications program to an output FITS file, the inherited keywords will then appear in the new EHU (or PHU if it is written as the Primary Unit). An obvious problem with this scheme is in the effective duplication of all PHU keywords in the output EHUs after one step of processing or data reduction effectively defeating the usefulness of having "global" keywords in the PHU so they wouldn't be repeated everywhere. To avoid this needless replication, we have incorporated a mechanism to override the inheritance when opening an Extension Unit. This is accomplished by specifying an option as part of the filename/ extension selection specification. This is intended to be used primarily by the applications programs that expect files in FITS Image Extension format (e.g., the future STIS and NICMOS calibration pipeline programs) and expect the "global" PHU keywords. In this situation, we expect that the application program will disable inheritance, read the PHU and EHU as separate entities, and propagate them separately. The inheritance of keywords is accomplished by the use of routines that search the EHU and PHU (in that order) for a given keyword. A possibility being considered (but certainly more difficult to implement) is to use the output PHU to determine if the keyword in the combined input PHU/EHU exists in the output PHU and has the same value. If so then the keyword would not be written to the output EHU. This scheme would also prevent the automatic duplication of PHU keywords in output EHUs. Since most FITS readers will have no knowledge about our local convention regarding keyword inheritance, it will be important for us (and any others for that matter) to ensure that all keywords that would be reasonably expected to be useful for analysis be present in the EHU even if they are the same for all Extensions. (This still leaves a healthy number of keywords for the global header as far as HST data products are concerned!) II) EXTENSION NAMES The second issue related to the FITS standard is the selection of Image Extensions. The obvious and straightforward way to specify an extension is to refer to it by number. The convention we have adopted is to refer to the Primary Unit as extension 0, and the first Extension Unit as 1. We have also adopted a naming scheme so that Image Extensions can be referred to by name. For the time being the name is defined by the combination of the values of the EXTNAME and EXTVER keywords, if they exist (future enhancements by NOAO may include more keywords as part of the naming scheme). This naming convention is essentially identical to that proposed in "A Hierarchical Grouping Convention for FITS" by Jennings et al. The planned data products of the future instruments will incorporate one or more "groups" of images into one FITS file. Such a group might consist of a science image, data quality image, and error image. There are occasions where multiple groups will be in the same file. Our plan is to use EXTNAME to specify the image type and EXTVER to specify the number in the sequence, e.g., SCIENCE,3 would specify the first extension with EXTNAME=SCIENCE and EXTVER=3. There are many more details involved in specifying the IRAF imio interface for FITS image extensions, but they don't directly relate to FITS issues. Anyone who is interested in those details can email me for the specifics. NULL IMAGES: There is one last issue that does not involve the IRAF imio interface directly but is related to planned HST data products and the FITS standard. The data products plan to use "NULL" Data Units for two distinct cases. For the first case, and intended to be used for all STIS and NICMOS data, the Primary Data Unit will be a NULL array. This is so that all the data products can be referred to by name, and a use of either EXTNAME or EXTVER in the Primary Header Unit would clearly be at odds with their definition. The second use of NULL arrays is when certain data arrays, for example data quality arrays or error images, will be filled with a constant value. In these cases it makes sense to use a NULL array with a keyword in the EHU specifying the constant value. The IRAF imio kernel will *not* automatically convert these Extensions into constant arrays. It will be up to specific application programs to interpret the meaning of these NULL images. Perry Greenfield From fitsbits-request Thu Aug 17 06:47:25 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["741" "Mon" "14" "August" "1995" "03:56:46" "GMT" "Mark Calabretta" "mcalabre at atnf.csiro.au" "" "23" "Reference point for conic projections" "^From:" nil nil "8" "1995081403:56:46" "Reference point for conic projections" (number " " mark " Mark Calabretta Aug 14 23/741 " thread-indent "\"Reference point for conic projections\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA28458; Thu, 17 Aug 1995 06:47:25 -0400 Return-Path: Message-Id: Organization: Australia Telescope National Facility, CSIRO Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!hookup!hudson.lm.com!news.math.psu.edu!news.cac.psu.edu!newsserver.jvnc.net!newsserver2.jvnc.net!howland.reston.ans.net!tank.news.pipex.net!pipex!news.mathworks.com!usenet.eel.ufl.edu!news.gmi.edu!msunews!harbinger.cc.monash.edu.au!news.cs.su.oz.au!metro!news.ci.com.au!wabbit.cc.uow.edu.au!news.nsw.CSIRO.AU!crux.rp.CSIRO.AU!usenet Reply-To: mcalabre at atnf.csiro.au Newsgroups: sci.astro.fits content-length: 739 From: mcalabre at atnf.csiro.au (Mark Calabretta) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Reference point for conic projections Date: Mon, 14 Aug 1995 03:56:46 GMT I would like to propose a change to the World Coordinate System (WCS) proposal. I believe it would be more useful if the reference point for the conic projections was midway between the two standard parallels rather than at the native pole. This "midpoint" convention would allow conic projections to be constructed with the reference point at the centre of the image. Is anyone currently using any of the conic projections defined in the WCS draft? Would anyone be inconvenienced if the reference point was changed to be at the midpoint of the two standard parallels? Are there any other objections to the change? Please respond by email to mcalabre at atnf.csiro.au. --- Dr. Mark Calabretta Australia Telescope National Facility From fitsbits-request Thu Aug 17 07:29:38 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["659" "" "16" "August" "1995" "09:40:37" "GMT" "William Sommerwerck" "williams at nwlink.com" "<40seel$hkh at alaska.nwlink.com>" "14" "Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995081609:40:37" "Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply" (number " " mark " William Sommerwer Aug 16 14/659 " thread-indent "\"Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply\"\n") "<808473033snz at npagroup.demon.co.uk>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA28520; Thu, 17 Aug 1995 07:29:38 -0400 Return-Path: Message-Id: <40seel$hkh at alaska.nwlink.com> Organization: Northwest Link Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!news.sprintlink.net!in1.uu.net!nwlink.com!usenet References: <808473033snz at npagroup.demon.co.uk> Newsgroups: sci.astro.fits,sci.data.formats,sci.astro,sci.physics.accelerators,sci.image.processing,sci.image.processing content-length: 657 From: William Sommerwerck Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply Date: 16 Aug 1995 09:40:37 GMT Tim Atkinson wrote: > I upgraded to a Yamaha 4X writer from a Phillips CDD521 (original) > and found that hardly anything would read the disks the Yamaha produced. > Reducing the Yamaha write speed to 2X speed solved this problem > mostly, for some really old pre-multisession players, you still have > to write disks at single speed. Don't know why but it works. The faster the CD ROM turns, the more power the recorder's laser has to put out to "burn" the disk. If it doesn't put out enough power, the disk will be "underexposed" (that's not right, but I can't think of a better analogy) and players won't be able to read it. From dwells Fri Aug 18 14:04:31 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["2797" "Fri" "18" "August" "1995" "15:33:30" "+0200" "Francois Ochsenbein" "francois at simbad.u-strasbg.fr" "<9508181804.AA00637 at fits.cv.nrao.edu>" "64" "Question on FITS keywords" "^Resent-Message-Id:" nil nil "8" "1995081813:33:30" "Question on FITS keywords" (number " " mark " Francois Ochsenbe Aug 18 64/2797 " thread-indent "\"Question on FITS keywords\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA00643; Fri, 18 Aug 1995 14:04:31 -0400 Return-Path: Message-Id: <9508181804.AA00637 at fits.cv.nrao.edu> content-length: 2795 Resent-Message-Id: <9508181804.AA00643 at fits.cv.nrao.edu> Resent-Date: Fri, 18 Aug 95 15:33:30 +0200 Resent-From: dwells (Don Wells) Resent-To: fitsbits From: francois at SIMBAD.u-strasbg.fr (Francois Ochsenbein) Sender: fitsbits-request at fits.cv.nrao.edu Apparently-To: fitsbits Subject: Question on FITS keywords Date: Fri, 18 Aug 95 15:33:30 +0200 I have just a questions raised by a problem in the order of the FITS keywords in in TABLE extension (XTENSION= 'TABLE '): the header starts with XTENSION= 'TABLE ' / Ascii Table Extension BITPIX = 8 / Character data NAXIS = 2 / Simple 2-D matrix NAXIS1 = 242 / Number of bytes per record NAXIS2 = 137 / Number of records PCOUNT = 0 / Get rid of random parameters GCOUNT = 1 / Only one group (isn't it obvious?) EXTNAME = 'table1.dat' / Catalog of Bright Galaxies in Hydra I EQUINOX = 1950 / Equinox of coordinates ==================================================================== TFIELDS = 48 / Number of data fields (columns) TBCOL1 = (etc )... and the FITS Users Guide specifies in section 3.4.1 that the order must be 1. XTENSION= 'TABLE ' 2. BITPIX 3. NAXIS 4. NAXIS1 5. NAXIS2 6. PCOUNT 7. GCOUNT 8. TFIELDS Does the Users Guide mean that exactly the 8 keywords must be consecutive (without intermediate other keywords) or not ? In other words, does the example conform or not to the standard, or should it be rewritten XTENSION= 'TABLE ' / Ascii Table Extension BITPIX = 8 / Character data NAXIS = 2 / Simple 2-D matrix NAXIS1 = 242 / Number of bytes per record NAXIS2 = 137 / Number of records PCOUNT = 0 / Get rid of random parameters GCOUNT = 1 / Only one group (isn't it obvious?) TFIELDS = 48 / Number of data fields (columns) EXTNAME = 'table1.dat' / Catalog of Bright Galaxies in Hydra I EQUINOX = 1950 / Equinox of coordinates ==================================================================== TBCOL1 = (etc )... Thank you for the expert's comments ... Francois Ochsenbein / CDS Strasbourg ================================================================================ _/_/_/_/_/_/ . . . Centre de Donnees astronomiques de _/ Strasbourg _/ http://cdsweb.u-strasbg.fr/CDS.html _/ _/_/ _/_/_/ _/ _/ _/ _/ Francois Ochsenbein _/_/_/_/ _/ _/_/_/_/ Internet: francois at astro.u-strasbg.fr _/ _/ _/ _/ Phone: +33 88 35 84 11 _/_/ _/_/_/ _/ Fax: +33 88 37 14 08 _/ Post: Observatoire Astronomique _/ 11, rue de l'Universite . . . _/_/_/_/_/_/_/ 67000 STRASBOURG, France From fitsbits-request Fri Aug 18 17:35:27 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1078" "Fri" "18" "August" "1995" "17:35:20" "-0400" "William Pence" "pence at tetra.gsfc.nasa.gov" "<199508182135.RAA22316 at tetra.gsfc.nasa.gov>" "28" "Re: Question on FITS keywords" "^From:" nil nil "8" "1995081821:35:20" "Question on FITS keywords" (number " " mark " William Pence Aug 18 28/1078 " thread-indent "\"Re: Question on FITS keywords\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA01040; Fri, 18 Aug 1995 17:35:27 -0400 Return-Path: Message-Id: <199508182135.RAA22316 at tetra.gsfc.nasa.gov> content-length: 1076 From: William Pence Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Cc: francois at SIMBAD.u-strasbg.fr, pence at tetra.gsfc.nasa.gov Subject: Re: Question on FITS keywords Date: Fri, 18 Aug 1995 17:35:20 -0400 Regarding Francois Ochsenbein's question about the order of the FITS keywords in a TABLE extension, the A&A Suppl. Series paper (73, 365), which defines the format of the TABLE extension, says that the 8 required keywords must be in order, *and* must be the first 8 keywords in the file: "The first eight keywords (XTENSION through TFIELDS) *must* appear, and in the order shown here." Also, the BINTABLE definition paper, (in the March 22, 1995 issue of A&A) says: "The mandatory keywords XTENSION, BITPIX, NAXIS, NAXIS1, NAXIS2, PCOUNT, GCOUNT, and TFIELDS must be in order; other keywords follow these in an arbitrary order." While this does not unambiguously state that these 8 required keywords must also be the first 8 keywords in the file, the statement that "other keywords follow these" seems to strongly imply that the 8 keywords must appear before any other keywords. Thus, I think the example header with the EXTNAME and EQUINOX keywords appearing before the TFIELDS keyword would not conform to the FITS standard. -Bill Pence HEASARC/NASA From fitsbits-request Fri Aug 18 22:17:24 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1024" "Sun" "13" "August" "1995" "12:46:03" "-0600" "Jeff Bloch" "jbloch at alexis" "" "26" "Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995081318:46:03" "Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply" (number " " mark " Jeff Bloch Aug 13 26/1024 " thread-indent "\"Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA01750; Fri, 18 Aug 1995 22:17:24 -0400 Return-Path: Message-Id: Organization: Los Alamos National Laboratory Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!tank.news.pipex.net!pipex!howland.reston.ans.net!vixen.cso.uiuc.edu!uchinews!ncar!newshost.lanl.gov!alexis!jbloch Newsgroups: sci.astro.fits,sci.data.formats,sci.astro,sci.physics.accelerators,sci.image.processing,sci.image.processing content-length: 1022 From: Jeff Bloch Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply Date: Sun, 13 Aug 1995 12:46:03 -0600 We store 650MB of satellite data on a CD ROM every 5 days using a setup from Young Minds on a Sparcstation. We have over 100 CD volumes in the archive. All has been well for the last two years until Sun started shipping new 4x CD ROM readers from Toshiba with new workstations. Suddenly, 50% of our archive is unreadable by the new drives. Old drives (mostly Sony) continue to read the archive just fine. This has nothing to do with the ISSO standards. Young Minds has not quite fessed up to the problem yet, but there may be a hardware problem with the particular model of the Phillips writer that we have. -------------------------------------------------------------------------- Jeffrey Bloch Office: (505) 665-2568 Astrophysics and Radiation Measurements Group ALEXIS Soc: (505) 665-5975 Los Alamos National Laboratory FAX: (505) 665-4414 Group NIS-2, Mail Stop D436 e-mail: jbloch at lanl.gov Los Alamos, NM 87545 -------------------------------------------------------------------------- From fitsbits-request Sun Aug 20 11:58:47 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1581" "Mon" "14" "August" "1995" "09:47:07" "-0600" "Jeff Bloch" "jbloch at alexis" "" "31" "Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995081415:47:07" "Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply" (number " " mark " Jeff Bloch Aug 14 31/1581 " thread-indent "\"Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA04437; Sun, 20 Aug 1995 11:58:47 -0400 Return-Path: Message-Id: Organization: Los Alamos National Laboratory Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!news.mathworks.com!newsfeed.internetmci.com!ncar!newshost.lanl.gov!alexis!jbloch References: Newsgroups: sci.astro.fits,sci.data.formats,sci.astro,sci.physics.accelerators,sci.image.processing,sci.image.processing content-length: 1579 From: Jeff Bloch Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Beware about CD ROM Data Longevity - Was Re: Saving 100GB of data for 15 years cheaply Date: Mon, 14 Aug 1995 09:47:07 -0600 The folks here who did the CD ROM study pointed out to me that I made an error in my message below. Toshiba 2X readers provided in new Sparcstations exhibited the problem, and actually a Toshiba 4X reader did not exhibit the difficulty. The actual failure rate was 20% to 50% depending on the reader. We have both a Phillips "CDD 521 Upgraded" writer and an older "CDD 521" model. Disks written with the older model also show no difficulties with any drive, just ones written on the newer writer. -------------------------------------------------------------------------- Jeffrey Bloch Office: (505) 665-2568 Astrophysics and Radiation Measurements Group ALEXIS Soc: (505) 665-5975 Los Alamos National Laboratory FAX: (505) 665-4414 Group NIS-2, Mail Stop D436 e-mail: jbloch at lanl.gov Los Alamos, NM 87545 -------------------------------------------------------------------------- On Sun, 13 Aug 1995, Jeff Bloch wrote: > > We store 650MB of satellite data on a CD ROM every 5 days using a setup from > Young Minds on a Sparcstation. We have over 100 CD volumes in the archive. > All has been well for the last two years until Sun started shipping new 4x CD > ROM readers from Toshiba with new workstations. Suddenly, 50% of our archive > is unreadable by the new drives. Old drives (mostly Sony) continue to read > the archive just fine. This has nothing to do with the ISSO standards. Young > Minds has not quite fessed up to the problem yet, but there may be a hardware > problem with the particular model of the Phillips writer that we have. > From fitsbits-request Sun Aug 20 14:54:38 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["2760" "" "17" "August" "1995" "19:31:28" "GMT" "Francois Ochsenbein" "francois at cdsxb6.nosubdomain.nodomain" "<4105eg$rri at apopi.u-strasbg.fr>" "62" "Order of FITS keywords in ASCII TABLE extension" "^From:" nil nil "8" "1995081719:31:28" "Order of FITS keywords in ASCII TABLE extension" (number " " mark " Francois Ochsenbe Aug 17 62/2760 " thread-indent "\"Order of FITS keywords in ASCII TABLE extension\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA04588; Sun, 20 Aug 1995 14:54:38 -0400 Return-Path: Message-Id: <4105eg$rri at apopi.u-strasbg.fr> Organization: CRC - Universite Louis Pasteur - Strasbourg France Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!zip.eecs.umich.edu!newshost.marcam.com!news.mathworks.com!tank.news.pipex.net!pipex!oleane!jussieu.fr!univ-lille1.fr!u-strasbg.fr!cdsxb6!francois Newsgroups: sci.astro.fits content-length: 2758 From: francois at cdsxb6.NoSubdomain.NoDomain (Francois Ochsenbein) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Order of FITS keywords in ASCII TABLE extension Date: 17 Aug 1995 19:31:28 GMT I have just a questions raised by a problem in the order of the FITS keywords in in TABLE extension (XTENSION= 'TABLE '): the header starts with XTENSION= 'TABLE ' / Ascii Table Extension BITPIX = 8 / Character data NAXIS = 2 / Simple 2-D matrix NAXIS1 = 242 / Number of bytes per record NAXIS2 = 137 / Number of records PCOUNT = 0 / Get rid of random parameters GCOUNT = 1 / Only one group (isn't it obvious?) EXTNAME = 'table1.dat' / Catalog of Bright Galaxies in Hydra I EQUINOX = 1950 / Equinox of coordinates ==================================================================== TFIELDS = 48 / Number of data fields (columns) TBCOL1 = (etc )... and the FITS Users Guide specifies in section 3.4.1 that the order must be 1. XTENSION= 'TABLE ' 2. BITPIX 3. NAXIS 4. NAXIS1 5. NAXIS2 6. PCOUNT 7. GCOUNT 8. TFIELDS Does the Users Guide mean that exactly the 8 keywords must be consecutive (without intermediate other keywords) or not ? In other words, does the example conform or not to the standard, or should it be rewritten XTENSION= 'TABLE ' / Ascii Table Extension BITPIX = 8 / Character data NAXIS = 2 / Simple 2-D matrix NAXIS1 = 242 / Number of bytes per record NAXIS2 = 137 / Number of records PCOUNT = 0 / Get rid of random parameters GCOUNT = 1 / Only one group (isn't it obvious?) TFIELDS = 48 / Number of data fields (columns) EXTNAME = 'table1.dat' / Catalog of Bright Galaxies in Hydra I EQUINOX = 1950 / Equinox of coordinates ==================================================================== TBCOL1 = (etc )... Thank you for the expert's comments ... ================================================================================ _/_/_/_/_/_/ . . . Centre de Donnees astronomiques de _/ Strasbourg _/ http://cdsweb.u-strasbg.fr/CDS.html _/ _/_/ _/_/_/ _/ _/ _/ _/ Francois Ochsenbein _/_/_/_/ _/ _/_/_/_/ Internet: francois at astro.u-strasbg.fr _/ _/ _/ _/ Phone: +33 88 35 84 11 _/_/ _/_/_/ _/ Fax: +33 88 37 14 08 _/ Post: Observatoire Astronomique _/ 11, rue de l'Universite . . . _/_/_/_/_/_/_/ 67000 STRASBOURG, France From fitsbits-request Mon Aug 21 09:43:33 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1013" "" "18" "August" "1995" "15:07:56" "GMT" "Guy Rixon" "gtr at ast.cam.ac.uk" "<412acc$tq at lyra.csx.cam.ac.uk>" "21" "Re: Saving 100GB of data for 15 years cheaply" "^From:" nil nil "8" "1995081815:07:56" "Saving 100GB of data for 15 years cheaply" (number " " mark " Guy Rixon Aug 18 21/1013 " thread-indent "\"Re: Saving 100GB of data for 15 years cheaply\"\n") ""] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA05858; Mon, 21 Aug 1995 09:43:33 -0400 Return-Path: Message-Id: <412acc$tq at lyra.csx.cam.ac.uk> Organization: Royal Greenwich Observatory Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!howland.reston.ans.net!tank.news.pipex.net!pipex!lyra.csx.cam.ac.uk!news References: <3vvvfu$a1e at ixnews6.ix.netcom.com> <404au3$jlh at itchy.itsnet.com> <406st6$8le at gwis2.circ.gwu.edu> <407tut$21e at electron.rutgers.edu> Newsgroups: sci.data.formats,sci.astro,sci.image.processing,sci.astro.fits,sci.physics.accelerators content-length: 1011 From: Guy Rixon Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: Saving 100GB of data for 15 years cheaply Date: 18 Aug 1995 15:07:56 GMT pettsj at visigoth.demon.co.uk (James Petts) wrote: >In article , >grunes at news.nrl.navy.mil (Mitchell R Grunes) wrote: > >> 3. Since no one liked my idea of film, I've an even better idea. >> Print it all out in hex, on acid free paper. When you need it, use a >> scanner. Let's see: two characters/byte, about 100 characters/line >> and 60 lines/page...now that's a REAL fire hazard. Compact, too. > >I came into this discussion late (i.e. this is the first message on >the subject I've seen), but what was the main objection to film. [Suggestions for safe storage of film deleted] The point about film is that you can't do all the analysis of the images you want to without re-digitizing the data and that's an inherently lossy process. We're talking about science data here, of course. If the pictures were held for their aesthetic value, or if they were, say, medical images that could be analyzed by looking at them on a light-box, then film would be ideal. From fitsbits-request Mon Aug 21 13:33:24 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["667" "" "18" "August" "1995" "22:09:13" "GMT" "Joe Heafner" "heafnerj at mercury.interpath.net" "<413329$p2s at redstone.interpath.net>" "16" "IMDISP under OS/2????" "^From:" nil nil "8" "1995081822:09:13" "IMDISP under OS/2????" (number " " mark " Joe Heafner Aug 18 16/667 " thread-indent "\"IMDISP under OS/2????\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA06270; Mon, 21 Aug 1995 13:33:24 -0400 Return-Path: Message-Id: <413329$p2s at redstone.interpath.net> Organization: Interpath -- Providing Internet access to North Carolina Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!concert!gatech!psuvax1!uwm.edu!reuter.cse.ogi.edu!engr.orst.edu!news.PEAK.ORG!news.orst.edu!wsrcc.com!wetware!hoho.quake.net!hilbert.dnai.com!redstone.interpath.net!mercury!heafnerj Newsgroups: sci.astro.fits content-length: 665 From: heafnerj at mercury.interpath.net (Joe Heafner - Astronomer) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: IMDISP under OS/2???? Date: 18 Aug 1995 22:09:13 GMT Can anyone give me any tips in using IMDISP under OS/2 Warp? I've tried it and while it doesn't generate any error messages, I can't seem to actually SEE the image! What am I doing wrong? Joe -- ************************************************************* * Paul J. Heafner, Astronomy and Physics Instructor * * Work:(704)327-7000, x.246 Home:(704)328-9425 * * HEAFNERJ at MERCURY.INTERPATH.NET HEAFNER_PJ at ALICE.LRC.EDU * * HEAFNERJ at ECSVAX.UNCECS.EDU * * Proud user of OS/2 Warp! * ************************************************************* From fitsbits-request Mon Aug 21 20:55:44 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["991" "" "21" "August" "1995" "09:57:29" "GMT" "Hartmut Frommert" "phfrom at nyx.uni-konstanz.de" "" "16" "FITS viewer for OS/2 Warp ?" "^From:" nil nil "8" "1995082109:57:29" "FITS viewer for OS/2 Warp ?" (number " " mark " Hartmut Frommert Aug 21 16/991 " thread-indent "\"FITS viewer for OS/2 Warp ?\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA07340; Mon, 21 Aug 1995 20:55:44 -0400 Return-Path: Message-Id: Organization: University of Constance, Dept. of Physics Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!uwm.edu!lll-winken.llnl.gov!simtel!swidir.switch.ch!scsing.switch.ch!news.belwue.de!news.uni-konstanz.de!maus.physik.uni-konstanz.de!phfrom Newsgroups: sci.astro.fits content-length: 989 From: phfrom at nyx.uni-konstanz.de (Hartmut Frommert) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: FITS viewer for OS/2 Warp ? Date: 21 Aug 1995 09:57:29 GMT Is there a fits viewer for native OS/2 Warp ? I have ViewFits which is fine under OS/2 2.1 but crashes under Warp, and I know that IMDISP runs in a Dos session of Warp, but certainly would like a native 32bit PM app more .. | OS/2 vs. NT: Better Half than Not Testified Hartmut Frommert | Warp 3.0 vs. 95: Cheaper, Smaller, Faster | and finally, it exists.. ----------- Get astronomical and space gifs via anon ftp from: ------------- ftp.seds.org; explorer.arc.nasa.gov: /pub/SPACE/GIF; jplinfo.jpl.nasa.gov | garbo.uwasa.fi:/pc/gif*; ftp.fce.vutbr.cz; ftp.univ-rennes1.fr; ftp.bdl.fr | HST: marvel.stsci.edu:/pubinfo/gif; ftp.cnam.fr; crux.astr.ua.edu| Updates | Clementine: clementine.s1.gov; images.jsc.nasa.gov (no more ftp) | welcome | - More in SEDS' Astro FTP List: ftp://ftp.seds.org/faq/astroftp.txt The SL9 crash was just a heavenly celebration of Apollo 11 25th anniversary! From fitsbits-request Tue Aug 22 12:48:42 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["5218" "Tue" "22" "August" "1995" "12:22:00" "-0400" "William Pence" "pence at tetra.gsfc.nasa.gov" "<199508221622.MAA23921 at tetra.gsfc.nasa.gov>" "106" "Re: FITS Image Extension support in IRAF" "^From:" nil nil "8" "1995082216:22:00" "FITS Image Extension support in IRAF" (number " " mark " William Pence Aug 22 106/5218 " thread-indent "\"Re: FITS Image Extension support in IRAF\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA00401; Tue, 22 Aug 1995 12:48:42 -0400 Return-Path: Message-Id: <199508221622.MAA23921 at tetra.gsfc.nasa.gov> content-length: 5216 From: William Pence Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Cc: perry at eclipse.stsci.edu, pence at tetra.gsfc.nasa.gov Subject: Re: FITS Image Extension support in IRAF Date: Tue, 22 Aug 1995 12:22:00 -0400 I have 4 suggested changes or additions to the proposed STScI local FITS convention for the INHERIT keyword (posted by Perry Greenfield on 17 Aug). As it stands, I think the proposal is too general or vague, and could cause inappropriate keywords to be inherited in some cases which could confuse users and possibly lead to incorrect results from the analysis of FITS files using this convention. Here are the suggested changes to the proposal: 1) The required keywords in the PHU shall never be inherited. 2) The PHU must have NAXIS = 0 for any keywords to be inherited. 3) If software modifies the value of an inherited keyword, that new keyword is simply added to the EHU, and the value of the keyword in the PHU is not changed. 4) Any COMMENT, HISTORY, or blank keywords in the PHU shall not be inherited. The following sections describe why I think the above restrictions are needed. 1. Inheritance of the required PHU keywords: There could be problems if the required FITS keywords are inherited from the primary header unit (PHU). Clearly the SIMPLE keyword must never be inherited. If the primary array has NAXIS = 3 and the extension has NAXIS = 2, then the extension must not inherit the NAXIS3 keyword from the PHU. Also, it should be forbidden to have an extension header unit (EHU) with just the following keywords: XTENSION= 'IMAGE ' INHERIT = T END with the expectation that the other required keywords would be inherited from the PHU. (This would violate the requirements for a FITS conforming extension in any case). All of these potential problems can be avoided simply by restricting the inheritance proposal to state that the following required or reserved PHU keywords shall never be inherited: SIMPLE, NAXIS, NAXISn, EXTEND, PCOUNT, GCOUNT, and END. 2. Inheritance of inappropriate PHU keywords by an Image extension: If the primary array is not null (NAXIS != 0) then there may be keywords in the PHU that should not be inherited. For instance, the BSCALE, BZERO, DATAMIN, and DATAMAX keywords might well only apply to the primary array. One solution to this problem would be to ensure that these keywords are always repeated in each EHU, so they will never be inherited, but due to negligence or ignorance users may forget to do this, potentially leading to disasterous results when analyzing the FITS file. There is a subtle pitfall here, too, in that often the absence of a keyword is used to imply a default value. BSCALE and BZERO for example should be assumed to have values of 1.0 and 0.0 if not present in the EHU, and one should presumably not inherit the value from the PHU in this case. This also applies to the CROTA2 WCS keyword which should be assumed to be 0.0 if not present in the EHU. I think the cleanest solution to this class of problems would be to restrict the inherit convention to only apply in cases where the PHU has NAXIS = 0. If NAXIS != 0, then the PHU keywords would not be inherited even if the EHU has the INHERIT= T keyword. If the PHU does have NAXIS = 0, then it is safer to assume that any keywords may be inherited. 3. Inheritance of inappropriate PHU keywords by a Table extension: The same problems described above in (2) also apply to TABLE or BINTABLE extensions, only more so. There are whole classes of PHU keywords which would be illegal to inherit into tables such as the WCS keywords, BSCALE, BZERO, BLANK, etc. Any implementation of this scheme would need to filter out these keywords before inheriting them to a table extension. However, this would not catch any user-defined keywords which only apply to a primary array and are meaningless, or confusing in the context of a table extension. One could of course restrict this proposal to only apply to IMAGE extensions, but this would be inconsistant since the arguments in favor of this proposal apply just as well to tables as they do to images. As in case (2), I think the simplest solution to these sorts of problems is to restrict the inheritance convention to only apply if the PHU has NAXIS = 0. By doing this, then it is less likely that there would be keywords in the PHU that would be forbidden in a table. 4. Modification of inherited keywords: If a program opens a particular EHU and inherits some keywords from the PHU, what should it do if it needs to modify the values of some of the inherited keywords? I think it would be too dangerous to directly modify the value in the PHU because of the unknown effect this would have on any other EHU which may inherit this same keyword. So I think the software should simply add this modified keyword value to the EHU, thus overriding the value in the PHU. If the software really needs to modify (or delete) the PHU keyword value then it should be required to physically open the PHU and make the modification directly. 5. Are HISTORY, COMMENT and blank keywords inherited? This is not a major problem, but the proposal should state how these keywords should be treated. I would think these keywords should not be inherited, but there may be cases where it would be desireable. Bill Pence HEASARC/NASA/GSFC pence at tetra.gsfc.nasa.gov From fitsbits-request Wed Aug 23 08:57:12 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["619" "" "22" "August" "1995" "08:54" "EDT" "Barry M. Schlesinger" "bschlesinger at nssdca.gsfc.nasa.gov" "<22AUG199508543192 at nssdca.gsfc.nasa.gov>" "20" "Re: FITS viewer for DECWindows?" "^From:" nil nil "8" "1995082212:54:00" "FITS viewer for DECWindows?" (number " " mark " Barry M. Schlesin Aug 22 20/619 " thread-indent "\"Re: FITS viewer for DECWindows?\"\n") "<4190f9$626 at gap.cco.caltech.edu>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA03773; Wed, 23 Aug 1995 08:57:12 -0400 Return-Path: Message-Id: <22AUG199508543192 at nssdca.gsfc.nasa.gov> Organization: NASA Goddard Space Flight Center -- Greenbelt, Maryland USA Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!caen!uwm.edu!lll-winken.llnl.gov!ames!newsfeed.gsfc.nasa.gov!nssdca.gsfc.nasa.gov!bschlesinger References: <4190f9$626 at gap.cco.caltech.edu> Newsgroups: sci.astro.fits content-length: 617 From: bschlesinger at nssdca.gsfc.nasa.gov (Barry M. Schlesinger) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: FITS viewer for DECWindows? Date: 22 Aug 1995 08:54 EDT In article <4190f9$626 at gap.cco.caltech.edu>, carl at SOL1.GPS.CALTECH.EDU writes... >Anybody out there got a FITS viewer that runs under X-Windows? The Sources of FITS Information posting, which on our system has not expired, points out a number of sites where information about software to read FITS files may be obtained. For locations where it has expired already, more general information about FITS and network references can be found at http://www.gsfc.nasa.gov/astro/fits/basics_info.html or ftp://nssdc.gsfc.nasa.gov/pub/fits/basics_info.txt Barry Schlesinger FITS Support Office GSFC/ADF From fitsbits-request Thu Aug 24 18:21:32 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1016" "" "21" "August" "1995" "04:01:45" "GMT" "Carl J. Lydick" "carl at sol1.gps.caltech.edu" "<4190f9$626 at gap.cco.caltech.edu>" "19" "FITS viewer for DECWindows?" "^From:" nil nil "8" "1995082104:01:45" "FITS viewer for DECWindows?" (number " " mark " Carl J. Lydick Aug 21 19/1016 " thread-indent "\"FITS viewer for DECWindows?\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA11201; Thu, 24 Aug 1995 18:21:32 -0400 Return-Path: Message-Id: <4190f9$626 at gap.cco.caltech.edu> Organization: HST Wide Field/Planetary Camera Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!portal.gmu.edu!europa.chnt.gtegsc.com!howland.reston.ans.net!swrinde!elroy.jpl.nasa.gov!lll-winken.llnl.gov!fnnews.fnal.gov!nntp-server.caltech.edu!SOL1.GPS.CALTECH.EDU!CARL Reply-To: carl at SOL1.GPS.CALTECH.EDU Newsgroups: sci.astro.fits content-length: 1014 From: carl at SOL1.GPS.CALTECH.EDU (Carl J Lydick) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: FITS viewer for DECWindows? Date: 21 Aug 1995 04:01:45 GMT Anybody out there got a FITS viewer that runs under X-Windows? Source code for such a beast would be very helpful to me in writing some software here. I only need something that will handle a FITS file with: SIMPLE = T / Conformance to FITS BITPIX = 16 / Bits per pixel in data file NAXIS = 2 / Number of dimensions in data file Thanks in advance. -------------------------------------------------------------------------------- Carl J Lydick | INTERnet: CARL at SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My understanding of astronomy is purely at the amateur level (or below). So unless what I'm saying is directly related to VAX/VMS, don't hold me or my organization responsible for it. If it IS related to VAX/VMS, you can try to hold me responsible for it, but my organization had nothing to do with it. From fitsbits-request Sat Aug 26 10:15:10 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["596" "" "24" "August" "1995" "22:57:25" "GMT" "Rob Seaman" "seaman at noao.edu" "<41j04l$66r at noao.edu>" "19" "FITS checksum proposal" "^From:" nil nil "8" "1995082422:57:25" "FITS checksum proposal" (number " " mark " Rob Seaman Aug 24 19/596 " thread-indent "\"FITS checksum proposal\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA17957; Sat, 26 Aug 1995 10:15:10 -0400 Return-Path: Message-Id: <41j04l$66r at noao.edu> Organization: National Optical Astronomy Observatories, Tucson, AZ, USA Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!maui.cc.odu.edu!news.larc.nasa.gov!lerc.nasa.gov!magnus.acs.ohio-state.edu!math.ohio-state.edu!howland.reston.ans.net!swrinde!gatech!ncar!noao!seaman Newsgroups: sci.astro.fits content-length: 594 From: seaman at noao.edu (Rob Seaman) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: FITS checksum proposal Date: 24 Aug 1995 22:57:25 GMT A new draft of the FITS checksum proposal is available from: ftp://iraf.noao.edu/misc/checksum/checksum.ps A README file and example data and code may be retrieved from the same directory. The major change from earlier drafts is that the unsigned integer value of the DATASUM keyword is no longer ASCII encoded. Various suggestions from Barry Schlesinger and Arnold Rots have improved the document. Rob Seaman, Bill Pence -- seaman at noao.edu, http://iraf.noao.edu/~seaman NOAO, 950 N Cherry Ave, Tucson AZ 85719, 520-318-8248 PGP: 98 8D 8B 49 74 9A 41 88 3A 43 87 54 51 BF 30 4B From fitsbits-request Tue Aug 29 00:00:42 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["2459" "" "25" "August" "1995" "15:51" "EDT" "Barry M. Schlesinger" "bschlesinger at nssdca.gsfc.nasa.gov" "<25AUG199515513160 at nssdca.gsfc.nasa.gov>" "51" "Re: FITS Image Extension support in IRAF" "^From:" nil nil "8" "1995082519:51:00" "FITS Image Extension support in IRAF" (number " " mark " Barry M. Schlesin Aug 25 51/2459 " thread-indent "\"Re: FITS Image Extension support in IRAF\"\n") "<199508221622.MAA23921 at tetra.gsfc.nasa.gov>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA28760; Tue, 29 Aug 1995 00:00:42 -0400 Return-Path: Message-Id: <25AUG199515513160 at nssdca.gsfc.nasa.gov> Organization: NASA Goddard Space Flight Center -- Greenbelt, Maryland USA Path: solitaire.cv.nrao.edu!hearst.acc.Virginia.EDU!news-server.ncren.net!concert!gatech!news.mathworks.com!newsfeed.internetmci.com!news.msfc.nasa.gov!pendragon.jsc.nasa.gov!ames!newsfeed.gsfc.nasa.gov!nssdca.gsfc.nasa.gov!bschlesinger References: <199508221622.MAA23921 at tetra.gsfc.nasa.gov> Newsgroups: sci.astro.fits content-length: 2457 From: bschlesinger at nssdca.gsfc.nasa.gov (Barry M. Schlesinger) Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: FITS Image Extension support in IRAF Date: 25 Aug 1995 15:51 EDT In article <199508221622.MAA23921 at tetra.gsfc.nasa.gov>, William Pence writes... >I have 4 suggested changes or additions to the proposed STScI local >FITS convention for the INHERIT keyword (posted by Perry Greenfield on >17 Aug). As it stands, I think the proposal is too general or vague, >and could cause inappropriate keywords to be inherited in some cases >which could confuse users and possibly lead to incorrect results from >the analysis of FITS files using this convention. > The general issue raised here, that of which header keywords are inherited, is an important one, not just the question of the particular solution suggested. The proposal needs to define more precisely which keywords in the header are inherited and which are not. More generally, it is important in the context of this discussion to remember that individual FITS HDUs may be copied from one file to another. That possibility is one reason why most FITS software treats extensions independently. Although the rule stated in the Generalized Extensions Paper states only that keywords in the primary header are not permitted to announce the presence of a particular kind of extension, in general, it is not advisable to have keywords in the primary header that mandate the structure or content of the extensions in the file, since there is no requirement to copy the extensions along with the primary header. EXTEND = T, for example means only that extensions *may* be present, not necessarily that they are. The discussion in the proposal about maintaining information in the extensions, >> Since most FITS readers will have no knowledge about our local >> convention regarding keyword inheritance, it will be important for us >> (and any others for that matter) to ensure that all keywords that >> would be reasonably expected to be useful for analysis be present >> in the EHU even if they are the same for all Extensions. (This still >> leaves a healthy number of keywords for the global header as far >> as HST data products are concerned!) also applies to the case where the extension is moved from one file to another by software that does not recognize the inheritance convention. Perhaps the best use of inheritance would be to express information that is relevant to a set of HDUs considered together but not of particular interest when only one is read. Barry Schlesinger FITS Support Office GSFC/ADF From fitsbits-request Tue Aug 29 11:54:09 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["2105" "Tue" "29" "August" "1995" "11:54:02" "-0400" "RANDY THOMPSON" "RTHOMPSON at iuedac.gsfc.nasa.gov" "<950829115402.2900012b at iuedac.gsfc.nasa.gov>" "35" "RE: FITS Image Extension support in IRAF" "^From:" nil nil "8" "1995082915:54:02" "FITS Image Extension support in IRAF" (number " " mark " RANDY THOMPSON Aug 29 35/2105 " thread-indent "\"RE: FITS Image Extension support in IRAF\"\n") nil] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA00846; Tue, 29 Aug 1995 11:54:09 -0400 Return-Path: Message-Id: <950829115402.2900012b at iuedac.gsfc.nasa.gov> content-length: 2103 From: RANDY THOMPSON (301) 286-8800 Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Cc: RTHOMPSON at iuedac.gsfc.nasa.gov Subject: RE: FITS Image Extension support in IRAF Date: Tue, 29 Aug 1995 11:54:02 -0400 (EDT) I think the issue of inheritance is important, and the question of how to keep data sets connected with the data descriptions has become more involved with the addition of FITS extensions and FITS readers which separate extensions into separate files. The adoption of the Image extension may have made separating FITS extensions into separate FITS files more feasible since the Image extension header format is so similiar to that of a primary header. It needs to be emphasized however that the issue of inheritance is not specific to the Image extension but applies equally to all the existing FITS extensions. For this reason, I think it's important that new conventions do not change the interpretation of existing FITS files. Obviously LOCAL conventions should not impact other projects, but IRAF users will undoubtedly want to read FITS files from other institutions. Our project adopted the policy of storing all but the required extension header keywords in the primary header. In the most complicated file format, we use a primary HDU containing a 2-dimensional image, followed by a Binary table extension, followed by two Image extensions. All the data in the file pertains to one observation. It did not seem necessary to us to repeat all the observational and processing history information (roughly 400 keywords) in each of the extension headers. The FILENAME keyword included in the extension headers provides (some) information about the individual data sets and does afford a way to keep data sets connected if they are written to separate files. In retrospect, particularly in view of the discussions on inheritance, perhaps we should have considered a different format for the IUE archive in which more information is stored in the individual extension headers. Unfortunately, I think its too late for us to change, and because FITS files like ours exist, it may not be valid to make assumptions about how primary header information should or should not be inherited by the extensions when newly-defined keywords (such as INHERIT = T) are not present. From fitsbits-request Wed Aug 30 02:57:44 1995 X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["2275" "Wed" "30" "August" "1995" "09:01:45" "+0200" "Lucio Chiappetti" "lucio at ifctr.mi.cnr.it" "" "47" "Re: FITS Image Extension support in IRAF" "^From:" nil nil "8" "1995083007:01:45" "FITS Image Extension support in IRAF" (number " " mark " Lucio Chiappetti Aug 30 47/2275 " thread-indent "\"Re: FITS Image Extension support in IRAF\"\n") "<25AUG199515513160 at nssdca.gsfc.nasa.gov>"] nil) Received: by fits.cv.nrao.edu (5.x/S2.3/NRAO-CV/2.3) id AA02274; Wed, 30 Aug 1995 02:57:44 -0400 Return-Path: Organization: Istituto di Fisica Cosmica e Tecnologie Relative In-Reply-To: <25AUG199515513160 at nssdca.gsfc.nasa.gov> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII content-length: 2273 From: Lucio Chiappetti Sender: fitsbits-request at fits.cv.nrao.edu To: fitsbits at fits.cv.nrao.edu Subject: Re: FITS Image Extension support in IRAF Date: Wed, 30 Aug 1995 09:01:45 +0200 (MET DST) On 25 Aug 1995, Barry M. Schlesinger wrote: > The general issue raised here, that of which header keywords are > inherited, is an important one, .... [...] > > More generally, it is important in the context of this discussion to > remember that individual FITS HDUs may be copied from one file to > another. That possibility is one reason why most FITS software treats > extensions independently. Although the rule stated in the Generalized > Extensions Paper states ... Uh, my apologies to everybody, but I remain constantly puzzled by the emphasis on the ACTUAL USAGE of multi-extension files. It is clear to me why the "general extension" mechanism was invented that way (to allow the basic "image" files to coexist with new extensions), but it appears it did not have a widespread usage until binary tables came out. And even then I cannot understand why the majority of the cases (95 % ? or more ?) cannot be handled either : - as independent "image" files (i.e. primary HDU only) - as independent "single table" files (i.e. a primary HDU without data array, and ONLY another ONE HDU with a BINTABLE) I do not understand what are the reasons for having more tables in the same file (and not separate files). There is no significant gain in space (an argument which might perhaps be relevant for a DATA EXCHANGE format), and there is complication in the reading/analysis software. This is particularly true if the tables collated together are of miscellaneous nature and format. I could PERHAPS find some sense for IMAGE extensions (the 5% or less missing above) where all extensions are of the same and simple nature and layout, and only in the case the extensions are of "ancillary" nature w.r.t. the primary data array (like the case of the "epsilons" for IUE). ---------------------------------------------------------------------------- Lucio Chiappetti - IFCTR/CNR | Ma te' vugl' da' quost avis a ti' Orsign via Bassini 15 - I-20133 Milano | Buttet rabios intant te se' pisnign Internet: LUCIO at IFCTR.MI.CNR.IT | Decnet: IFCTR::LUCIO | (Rabisch, II 46, 119-120) ----------------------------------------------------------------------------