Query value between min-max and return another value

[Marshall Barton]

I don't want to be harsh on anyone. Just this week I broke my own rule
about filing a bug report in 'anger' and recorded a false statement out
of sheer frustration, not from spite but because I didn't revisit my
steps to reproduce and assumed I knew the cause of what I'd seen, etc.

This isn't a matter of 'trust' or 'reputation'. Even the best
scientists and mathematicians make mistakes; that's why a proof is
produced for peer scrutiny which benefits everyone, including the
author. The word 'proof' is not pejorative here; it is not meant to
imply distrust. How can we start looking to solve or ameliorate a
problem if we can't reproduce it, let alone determine causation.

As you point out, we have the mention of a further bug - null collation
- but no proof, no hint of how to reproduce it. So let's use reason.

Not every aspect of the Jet engine is documented, and some which still
could not be described as 'supported' functionality. Null collation,
however, is documented, supported and is to be relied upon.

Don't you think that if there was a known issue with null collation it
would be a more serious problem that a DECIMAL sort order bug that only
affects part of a resultset when the order is descending and includes
negative as well as positive values and/or zeros? I do. I'd certainly
want more details on the status of any null collation bug e.g. what
other functionality is affected e.g. GROUP BY? I'd expect a bug report
from Microsoft. MVPs report bugs to Microsoft, right?


I know what you are doing. Don't get me wrong: I think it's admirable
to show support for each other. The MVP is a good 'brand' because of
it. But I don't think such issues are about having a good track record
(and Allen deservedly has one of the best in the Access community).

But I can't agree your sentiments. I think that being in a position of
trust and mentioning an apparent bug but failing to providing steps to
reproduce is like

MVPs have no control over what bug information Microsoft
makes available to the public.

Since you were responding to my post I thought you were
asking my opinion. If you want more than that, you should
take it up with Allen.

 

[Marshall Barton]

I hope OP got his question answered elsewhere.

It may not be the answer he wanted, but it was answered.
Since he hasn't come back in the last several days, I assume
that part of the thread is either closed or at least
dormant.

 

[Jamie Collins]

[interesting discussion snipped]

I hope OP got his question answered elsewhere.

I post to Usenet via Google Groups, where the strap line is, "Create,
search, and browse groups to discuss and share ideas." Q & A is only
aspect of this group and this discussion hasn't gone too far OT e.g.
should you take a person's reply as 'truth' based on their reputation
or is it important to have code to reproduce the problem, proper
identification of causation, etc?

Jamie.

--

 

[Jamie Collins]

Allen's sample table does not demonstrate Null sorting
randomly and I have not personally seen the issue (because I
have no need to use Decimal type), but I believe there is
such a case. Allen is too careful and digs into this kind
of thing too deeply to make a blatently false statement out
of sheer frustration. I'm just grateful that he's willing
to spend the time to document problems for the rest of us.

Reflecting on this, I think I've written a few original articles on the
DECIMAL data type to be considered a credible witness myself e.g.

Problem with DECIMAL calculations in client side recordsets:
http://groups.google.com/group/microsoft.public.access.modulesdaovba.ado/msg/cafd795b34f175da

Bug with Jet OLEDB provider and DECIMAL subquery?
http://groups.google.com/group/microsoft.public.access/msg/c1759f0dc4499dcf

DECIMAL oddity:
http://groups.google.com/group/microsoft.public.access/msg/2b99730e4badf9b8

Jet 4.0 DECIMAL type: sort order bug:
http://groups.google.com/group/microsoft.public.access/msg/9245b50947e9839f

Jet 4.0 DECIMAL type: VBA Decimal type and take up in Access:
http://groups.google.com/group/microsoft.public.access/msg/076891d5343a0461

Also, I get the impression that Allen avoids the DECIMAL type, whereas
I use almost everyday. I'm prepared to sick my neck out and say I've
had more experience of the DECIMAL data type in Jet that anyone else
who has posted to the Access groups in recent years. If you review the
above articles only briefly I trust you will be left with the
impression that I have investigated at least some of the DECIMAL type's
dark corners (you may also note I always provide SQL code, and often
VBA code, to reproduce problems). Don't you think if there was a
problem with NULL collation that in all my time working with the
DECIMAL type I would have seen it by now?

Jamie.

--

 

[Jamie Collins]

MVPs have no control over what bug information Microsoft
makes available to the public.

I appreciate that. My point is, if Microsoft was aware of a problem
with null collation I'm almost certain it would justify its own bug
report.

Since you were responding to my post I thought you were
asking my opinion. If you want more than that, you should
take it up with Allen.

How would I do that? I've petitioned Allen on the subject often in
these groups, which I believe is the proper conduct, but to date I have
had no reply and the article remains unchanged. I do not want to make
this a personal thing so I think it would be inappropriate to send
email, start new threads with "Attention Allen..." in the subject, etc.
I don't think I am in a position to ask you but if he is a personal
contact of yours perhaps you would be motivated to pass word on to him.

I think the best I can do is challenge those who refer to the article
by asking them if they can defend its content.

Jamie.

--