TMG-L Archives

Archiver > TMG > 2001-01 > 0979243161


From: "Grawrock, David" <>
Subject: RE: [TMG] List of buglets?
Date: Thu, 11 Jan 2001 11:59:21 -0800


Though I don't think he needs it, I'm going to support Bob's side of the
equation.

First a master bug list is a messy thing. It includes all sorts of
information that no one except the developers need. It might have source
code, function names and other data. It might be very, very, very cryptic.
So who takes the time to massage this database? It is a real full time job
believe me. And if it is someone outside of WG how do they mandate that the
list be up to date?

Changes.txt tells you the fixes, but how would correlate your bug with a
GEDCOM import with the EOF read problem listed in a bug list. Only after
fixing the problem does the programmer find out that OH that was that bug
also. And that goes into changes.txt.

One thing to remember is that a bug dressed up is just a feature. And
sometimes that decision is made very late in the release cycle, something
that was a problem turns out to be what people want.

People have been after software companies for years to put their entire bug
database on-line. Nobody does. The closest you get is to be a beta tester
where you see things happening. At some companies beta testers have direct
access to the bug database so that they don't enter duplicate bugs. But that
is rare.

David





Diana and Bob wrote

It doesn't quite feel right to disagree with Bob, but as the person who
originally posted the list of buglets suggestion, I respectfully do so.

Bob Velke wrote:>
>
> The reason that the company is not involved with such an effort is that,
> unless someone is dedicated to keeping it up to date, it gets woefully out
> of date and is more misleading than instructive.

I appreciate that everyone at Wholly Genes is overworked at the moment
(and the timing of this request is lousy), but surely there must be a
master list of known bugs. If someone reports a bug and Wholly Genes
confirms it (not an interpretation issue or a problem that just effects
one user) it must get added to this list. And when it is fixed it must
get taken off the list.

Bob, you said at the end of your message there weren't that many of
them, so why would it be that difficult to let us know about them?
Again, I wasn't even suggesting that TMG post it on their website
(although that would be the ideal), but simply that someone be willing
to confirm for us that a bug was "on the list" so that we could
accurately post the information ourselves. And yes, I understand your
concern that an item be deleted after it has been noted as corrected in
a change text.

> If one wants an historical list of bugs and fixes, changes.txt is a
> much more accurate source.

But that misses the point. I am very familiar with the changes.txt, but
they do not tell me anything at all about bugs that are still in the
system and that might be affecting the way my software is currently
working. They only tell me what has been fixed... which by the very
definition of "change" has to be after the fact.

> As has been discussed here, there is a known problem in v4.0c regarding
> renumbering of embedded citations, some export issues for WordPerfect
> users, a problem with Visual Chartform and custom page sizes on HP
> printers, and a few other issues that are mostly isolated to individual
> users.

I realize the above was not, in any way, intended to be an all inclusive
list but there are two other problems that you do not mention that I
would love to know you are "officially aware of and working on". They
are examples of how much simpler it would be from the standpoint of the
user if we could check an official list.

First, I sent an email to Wholly Genes several months ago and reported
that the exclusion marker does not work on relationship source
citations. I received an email back that this appeared to be a bug and
that it would be reported to you for consideration. Do I assume you
decided it was a bug and that it will eventually be fixed? Or is it
possible you got the message and decided the "show relationship sources"
under report definitions overrode the exclusion marker in this case by
design?

And ... I use split citation fields and unique sources extensively, but
you cannot currently use both of them together. Is this a bug or simply
the way it works? I read all the archived TMG-L messages and discovered
several that mention the problem but do not describe it as a bug - just
say "don't try to use both features together". I called Wholly Genes in
Nov. and was told by the gentleman at tech support that he didn't know
if it was a bug or not. I posted a message to TMG-L in late Nov. and
received a reply from another user that it was indeed a bug but that it
wasn't fixed in 4.0c. Now I have spent many hours entering data under
the assumption the problem will eventually be fixed, but I don't really
know YOU consider it a bug and you are the one who matters here :)

> Like _all_ software, TMG has some bugs which were found after
> release, but I believe that we have a stronger record than most in
> providing regular fixes in the form of free downloadable updates.

There was never meant to be a suggestion otherwise. And TMG support is
a level above any other company I have ever known. I have sent in an
email and had personal phone calls back with a solution while I was
still sitting at my computer. Remarkable.

TMG, wonderful as it is, has a learning curve. People start out with
the basics, then start learning all the other great features. But as
you do this, and make decisions, some of which have long term effects on
the way you design reports and share data, it would be very helpful to
know if there are official bugs that affect the way things are working.

> I do expect there to be a v4.0d soon, despite the fact that most of our
> development effort is now on TMG v5.0.

Cool. And I would wait for 5.0 if need be - this was never a complaint
about the existance of bugs, or the amount of time it takes to resolve
them; I just want to know "for sure" what ones exist and might be
impacting me.

Thanks for taking the time to participate in this discussion.

Diana P.



This thread: