ARev and Y2K (AREV Specific)
At 13 OCT 1998 08:12:59AM C.Miller wrote:
The recent mailing from Revelation Software mentions Y2K compliance for Advanced Revelation versions 2.03 and above, but does not mention versions below 2.03. Are there any differences with how older versions of ARev (1.16, etc.) handle Y2K compliance, as compared to
2.03? The mailing from RevSoft implies it, while the white paper on this web site says nothing about it.
At 13 OCT 1998 02:21PM Gary Gnu wrote:
The original authors of Revelation could not think of everything. Look at RevG. It was written before G-d created colour, that's why most RevG apps are in black & white. Ater colour was introduced, a few people tried to retrofit it in, but it usually resulted in that washed-out sepia look.
Now with the Y2K stuff, no one realised what was going to happen in the future. When 2.02 was written, most programmers were still going on the theory that after 1999 came 199A. They figured by that time, computers would have taken over the world so we'd be doing the years in hex.
Good luck,
Gary Gnu
At 13 OCT 1998 06:49PM Capt'n Kirk wrote:
"The original authors of Revelation could not think of everything. Look at RevG. It was written before G-d created colour, that's why
most RevG apps are in black & white. Ater colour was introduced, a few people tried to retrofit it in, but it usually resulted in that
washed-out sepia look."
Not SO! Most everyone has met someone who has been 'color draped' and realizes that color is a 'mood-enhancing' thing. Real programmers have attitudes, not emotions!
"Now with the Y2K stuff, no one realised what was going to happen in the future."
Not SO (again). Everybody knew what was going to happen but thought that someone else was going to take care of it. That's why we have consultants. To tell management what didn't happen and who should be responsible. And then they advise the exec's to wait to see what a young billionaire will do next.
At 13 OCT 1998 08:22PM Andrew P McAuley wrote:
Unlikely - more likely to be a network products slanted idea. Still - why not upgrade so you can use the NLM/Serice - far mor stable!
World Leaders in all things RevSoft
At 13 OCT 1998 09:24PM Steve Smith wrote:
There are some limitations in the REVG date iconv which were overcome by AREV 1.15. All my assembler date routines were developed and tested under AREV 1.15.
Thinking back, there were few subsequent changes to the date ICONV logic in all versions from AREV 1.15 to AREV 3.12, except in AREV 1.1x, where if you entered 31/09/1997 (invalid date) it would resolve internally to the internal date number equal to 01/10/1997. Some versions of "next month's start date" functions operated using this bug, which was subsequently fixed about 2.01 (from my distant memory).
Similarly, there are no changes in the DATE() function behaviour. In any case, errors in date difference logic, first of month routines, history filenames with just YY instead of CCYY, symbolics with flaws, will still be of concern and greater impact in terms of Y2K.
Jocelyn Amon (who has posted here recently) has a series of excellent routines designed to spot such flaws in your application code. It's worth contacting her for her sample Y2K scan program.
Hope this helps,
Steve