In article <[log in to unmask]>,
Tracy Johnson <[log in to unmask]> writes
>On Thu, 24 Sep 1998, Gavin Scott wrote:
>
>> Cognos Reports Record Second Quarter Results Earnings Per Share
>> Rise 33 Percent on 22 Percent Revenue Growth
>
>That's mainly because all the holdovers who didn't buy an upgrade before
>(5.09 users, they who they are) finally realized they had to buy into
>version 8 to stay Y2K compliant!
>
Cognos threw a new distinction here, with 7.29C8.

I have a client on 7.29C, with all their QUIZ made Y2k-ready, except
that ADDCENTURY can currently only drive off the hard-coded 19 or 20 in
the PDL Dictionary, in that release.

No problem, there's some sort of pivot-point definition going to be
added to the PDL syntax, so a one-line change and a recompile should get
them fully ready, once they have the Y2k-compliant version of Quiz.

So they ask for it, get a nice big box of stuff sent down. So off I go
to upgrade the client's software.

But, hey - whoah - where do I define the pivot point for 1900/2000?

Ah, says Cognos, you don't get *that* until 8.19, shipping next week.

So what did you just send?, says I. Ah, 7.29C8, they say. That's the
first Y2k-compliant release.

Turns out they mean that it handles the date intrinsics correctly for
Y2k (so it needs TZ and all that good stuff). *If* you have explicitly
coded for it. And maybe the programs get the 19 or 20 for ADDCENTURY at
runtime instead of compile time. But no pivot point.


So now I learn I must make a distinction between Y2k-Compliant 7.29C8
(doesn't actually break in 2000) and Y2k-Friendly 8.19 (supports Y2k
implied centuries with the same ease that, say, VPLUS does. I hope.
Haven't seen it yet).

A new one on me... :-)

--
Roy Brown               Phone : (01684) 291710     Fax : (01684) 291712
Affirm Ltd              Email : [log in to unmask]
The Great Barn, Mill St 'Have nothing on your systems that you do not
TEWKESBURY GL20 5SB (UK) know to be useful, or believe to be beautiful.'