Re: y2k

I am sorry to bother you with another compliance question (I am sure
you get many), but could you direct
me to were I can obtain a Y2K compliance statement regarding
Postgres95 v6.3.2 for a Sparc/Solaris platform?
I need to provide this document to my systems administration team
before I can install the dbms. Any help would be greatly appreciated.

Hmm. I know what I _think_ about y2k on Postgres (no problem), but I'm
not sure what your admin team is looking for here (besides an excuse to
avoid work -- *slap* ooh, that wasn't nice was it?).

I'm not a big fan of holding open-source software to a legally binding
statement which could confer liability ("jeesh, you have the source
dudes, fix it yourself! And contribute back the solution!" :)

Anyway, if you have an example of what they are looking for, I'd be
happy to look at it and write something up. I can imagine something
which says:

I have added a mention to the FAQ on the web site, saying the we are Y2K
compliant.

-- 
  Bruce Momjian                        |  http://www.op.net/~candle
  maillist@candle.pha.pa.us            |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

Thus spake Bruce Momjian

I have added a mention to the FAQ on the web site, saying the we are Y2K
compliant.

Er, is that such a good idea? I might stick my neck out if I am being paid
for it but I don't know that I would want lawyers arguing over exactly
what constitutes "Y2K compliant." Sure, we are but that won't stop
ambulance ch^H^H^H^H^H^H^H^H^H^H^Hlawyers from causing us grief.

-- 
D'Arcy J.M. Cain <darcy@{druid|vex}.net>   |  Democracy is three wolves
http://www.druid.net/darcy/                |  and a sheep voting on
+1 416 424 2871     (DoD#0082)    (eNTP)   |  what's for dinner.

Thus spake Bruce Momjian

I have added a mention to the FAQ on the web site, saying the we are Y2K
compliant.

Er, is that such a good idea? I might stick my neck out if I am being paid
for it but I don't know that I would want lawyers arguing over exactly
what constitutes "Y2K compliant." Sure, we are but that won't stop
ambulance ch^H^H^H^H^H^H^H^H^H^H^Hlawyers from causing us grief.

Our software is supplied as-is. Period.

-- 
  Bruce Momjian                        |  http://www.op.net/~candle
  maillist@candle.pha.pa.us            |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

On 22-Oct-98 D'Arcy J.M. Cain wrote:

Thus spake Bruce Momjian

I have added a mention to the FAQ on the web site, saying the we are Y2K
compliant.

Er, is that such a good idea? I might stick my neck out if I am being paid
for it but I don't know that I would want lawyers arguing over exactly
what constitutes "Y2K compliant." Sure, we are but that won't stop
ambulance ch^H^H^H^H^H^H^H^H^H^H^Hlawyers from causing us grief.

But doesn't that depend entirely on what the meaning of "is" is? :)

Vince.
--
==========================================================================
Vince Vielhaber -- KA8CSH email: vev@michvhf.com flame-mail: /dev/null
# include <std/disclaimers.h> TEAM-OS2
Online Searchable Campground Listings http://www.camping-usa.com
"There is no outfit less entitled to lecture me about bloat
than the federal government" -- Tony Snow
==========================================================================

Why should we state any differently then most Open Software products:

http://www.sendmail.org/faq/section4.html#4.11
http://www.gnu.org/software/year2000.html
http://samba.gorski.net/samba/sambay2k.html

How do we store our dates? Same as everyone else...seconds since epoch?
If so, then its just our 'external representations' that would risk being
off, no? And, as someone else pointed out, we display ours as 4-digit.

On Thu, 22 Oct 1998, Thomas G. Lockhart wrote:

I am sorry to bother you with another compliance question (I am sure
you get many), but could you direct
me to were I can obtain a Y2K compliance statement regarding
Postgres95 v6.3.2 for a Sparc/Solaris platform?
I need to provide this document to my systems administration team
before I can install the dbms. Any help would be greatly appreciated.

Hmm. I know what I _think_ about y2k on Postgres (no problem), but I'm
not sure what your admin team is looking for here (besides an excuse to
avoid work -- *slap* ooh, that wasn't nice was it?).

I'm not a big fan of holding open-source software to a legally binding
statement which could confer liability ("jeesh, you have the source
dudes, fix it yourself! And contribute back the solution!" :)

Anyway, if you have an example of what they are looking for, I'd be
happy to look at it and write something up. I can imagine something
which says:

1) at the time of writing, Thomas Lockhart, a member of the loose
confederation of the Postgres support team, is not aware of having
received any reports of any problems in the Postgres code base related
to time transitions around Jan 1, 2000.

2) at the time of writing, in limited testing, which is documented in
the included regression tests, there have been no reports of problems
related to time transitions around Jan 1, 2000.

Aren't 1 & 2 saying the same thing?

3) at the time of writing, to the best of Thomas' knowledge, the
assumptions Postgres makes about dates specified with a two-digit year
are documented in the current User's Guide. (the significant transition
year is 1970, not 2000.)

URL reference to this section?

3) any y2k problems in the underlying OS related to obtaining "the
current time" may propagate into apparent y2k problems in Postgres.

This is basically what I read on most of the y2k statements...

What do the rest of the developers think about this?

The only thing I'd mention/provide is a URL to the section of the
User's Guide so that ppl dont' have to go searching for it...other then
that, I'd say it sounds both accurate to what we know at this time, while
not leaving any of us open to "but, hey, you said there wouldn't be any
problems"...

Might it not be wise to add in a comment dealing with the
version(s) of PostgreSQL that this pertains to? Something like
"referencing the currently released as well as development source trees"?
The only concern would be someone popping up and mentioning 1.01, cause,
well, they are still running that?

Marc G. Fournier
Systems Administrator @ hub.org
primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org

http://www.gnu.org/software/year2000.html

1) at the time of writing, Thomas Lockhart, a member of the loose
confederation of the Postgres support team, is not aware of having
received any reports of any problems in the Postgres code base
related to time transitions around Jan 1, 2000.
2) at the time of writing, in limited testing, which is documented
in the included regression tests, there have been no reports of
problems related to time transitions around Jan 1, 2000.

Aren't 1 & 2 saying the same thing?

They probably are, but I didn't mean them to. I wanted (1) to refer to
my knowledge of the code base, and it's behavior on my machine. I wanted
(2) to refer to the results of the regression testing on a wider mix of
machines, with better docs on what was actually tested and with (a lack
of) reports of problems from other users.

It seemed reasonable to (try to) say what I knew for sure, and put my
name on it, rather than have the group make a statement. But I like the
Gnu statement, and we should refer to that.

Probably better to make a short statement and then refer to the Gnu site
for related info. I'll put it in the docs, after running it by the
hackers group again.

3) any y2k problems in the underlying OS related to obtaining "the
current time" may propagate into apparent y2k problems in Postgres.

This is basically what I read on most of the y2k statements...
The only thing I'd mention/provide is a URL to the section of the
User's Guide so that ppl dont' have to go searching for it...other
than that, I'd say it sounds both accurate to what we know at this
time, while not leaving any of us open to "but, hey, you said there
wouldn't be any problems"...

Might it not be wise to add in a comment dealing with the
version(s) of PostgreSQL that this pertains to? Something like
"referencing the currently released as well as development source
trees"? The only concern would be someone popping up and mentioning
1.01, cause, well, they are still running that?

OK, all sounds good. We'll have something in the v6.4 docs.

- Tom

On Fri, 23 Oct 1998, Thomas G. Lockhart wrote:

Probably better to make a short statement and then refer to the Gnu site
for related info. I'll put it in the docs, after running it by the
hackers group again.

Actually, there is a really nice document that I found concerning
the whole Y2K issue that we also might want to refer to. It basically
refers to the "myths and lies" concerning Y2K, and I found it to be both
accurate and eye-opening :)

http://language.perl.com/news/y2k.html

Marc G. Fournier
Systems Administrator @ hub.org
primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org

I'll put it in the docs, after running it by the
hackers group again.

Y2K Statement

Author: Written by Thomas Lockhart on 1998-10-22.

The PostgreSQL Global Development Team provides the Postgres software
code tree as a public service, without warranty and without liability
for it's behavior or performance.

However, at the time of writing:

o The author, a volunteer on the Postgres support team since November,
1996, is not aware of any problems in the Postgres code base related to
time transitions around Jan 1, 2000 (Y2K).

o The author is not aware of any reports of Y2K problems uncovered in
regression testing or in other field use of recent or current versions
of Postgres. We might have expected to hear about problems if they
existed, given the installed base and the active participation of users
on the support mailing lists.

o To the best of the author' knowledge, the assumptions Postgres makes
about dates specified with a two-digit year are documented in the
current User's Guide in the chapter on data types. For two-digit years,
the significant transition year is 1970, not 2000; i.e. "70-01-01" is
interpreted as "1970-01-01", whereas "69-01-01" is interpreted as
"2069-01-01".

o Any Y2K problems in the underlying OS related to obtaining "the
current time" may propagate into apparent Y2K problems in Postgres.

Refer to The Gnu Project and The Perl Institute for further discussion
of Y2K issues, particularly as it relates to open source, no fee
software.

Note that in the html docs both references above contain URLs. Comments?
Suggestions?

- Tom