From: Richard Cox (COXRI4@mail.northgrum.com)
Date: Sat May 30 1998 - 03:53:17 EST
The FCS 2.0 standard, to quote "Data File Standards for Flow Cytometry", Cytometry 11:323-332(1990) specifies the DATE field as: $DATE/10-JAN-89/ In practice, BDIS used: \$DATE\01/10/89\ I would not classify this as a bug, but a shortcoming. I think of bugs as problems that cause a program to die whereas this is a design flaw. If the problem in the File Utilities routines did not exist (which is the Y2K problem in the BDIS code) the 2.0 standard would not make a program die. To adopt the FCS 3.0 standard would be more disruptive than agreeing to a reinterpretation of the year field, that is, any value in the "year" field in the range of 00 to 40 (the range to 40 could be any agreed to value; 40 is a straw-man) would be interpreted as 2000 to 2040 (since the Consort 32 system would surely be gone beyond that point). Once the File Utilities bug is corrected and programs like LYSYS, et al. were re-compiled with the corrected routines, this interpretation could be implemented rather simply. I have not seen the 3.0 standard but my guess is that more code would have to be changed to accomodate the newer standard than adopting the DATE format in the 3.0 standard. Correcting the "design flaw" in the routines that handle the date field and sticking with the 2.0 standard is simpler. Remember, the date field is an "optional" field anyway but to adopt a format of \$DATE\01/10/1989\ would make software, that did not get the corrections to the File Utilities, incompatable and not capable of reading files created with that different format. I think this is a viable solution; the question is, will BDIS permit it? Richard Cox At 03:35 PM 5/27/98 +0100, you wrote: > >Hi Mario, > >could you spell this one out please? > >Ray > > >At 10:48 am -0700 26/5/98, Mario Roederer wrote: >>There is a small hitch that even BDIS has apparently forgotten (given the copy >>of their missive posted to the mailing list)... The FCS2.0 standard is >>NOT year >>2000 compliant. Therefore, all software that produces FCS2.0 data, and this >>currently means every manufacturer out there, has the year 2000 bug. >> >>However, the current FCS3.0 standard, which is only a little different from >>FCS2.0, does not have this bug. >> > > Ray Hicks >________________________________________________________________________ >|University of Cambridge |Tel 01223 330149 | >|Department of Medicine |Fax 01223 336846 | >|Level 5, Addenbrookes Hospital |e-mail <rh208@cus.cam.ac.uk> | >|Hills Road Cambridge |Web http://facsmac.med.cam.ac.uk | >|CB2 |ftp server ftp://131.111.80.78 | >|UK | | >|_________________________________|_____________________________________| > > > >
This archive was generated by hypermail 2.1.6 : Thu Jan 01 2004 - 17:35:19 EST
