Re: [Fwd: Bug #19187 [Csd->CRp]: ** CRITICAL **: myx_grt_value_as_int: assertion `value != NULL' failed]

On Thu, 27 Apr 2006, Dave Howorth wrote:

> I wrote
>>> I'm not impressed with a bug tracking protocol
>
> Mike Lischke wrote:
>> I'm not sure what your problem is with that.
>
> Mainly the next paragraph in my post (see below).
>
>> And if he cannot reproduce it then it is quite likely the bug is
>> already fixed (although not always, of course).
>
> Not testing in the reported version would be careless. But Alfredo's report 
> did state that he tested the version I reported (good work, IMHO). There's 
> never any chance of it already being fixed in that version.
>
>> Have you actually tried to reproduce the problem more than once? Can
>> you see it 100% of the time?
>
> Yes.
>

Then please give more details. Is reverse-engineering required for the bug 
to appear? Does it appear if you just place new tables? Load a document? 
Do you get any other messages in the console? What exactly was
reverse engineered?

>> tried another machine/another distro?
>
> No. See my last comment.
>
>>> This kind of artificial bug-squashing statistics fiddling doesn't
>>> fall within the bounds of acceptable ethical behaviour IMHO.
>
> This is my problem that you didn't understand.
>

It is not any "artificial bug-squashing", I took the day yesterday to fix 
bugs and actually did spend 15h or so fixing as many bugs as possible 
(many of which, btw, were very helpful reports by John Yodsnukis who even 
attached patches, although the clear and detailed report would have been 
enough). When we stumble upon a bug that's a total mystery there's not 
much we can do, specially when requests for more details are just ignored, 
causing the bug to stay there, in the limbo, without anyone else ever 
seeing it.

You must understand that most of the time, bugs that appear blatantly and 
promptly for you may not appear for someone else. There are hundreds of
variables that can affect how the program behaves (distro, environment 
variables, libraries, mysql server version, schema, data being handled and 
even the locale) and without enough details from the reporter, it is some times 
very difficult to to repeat anything.

Btw, this particular bug has already been fixed, afaik. (but if I'm wrong, 
feel free to correct me, it's not like I'm trying to cover up something)

>> I understand your frustration. You want that beast fixed and we are
>> here to fix it. But please understand also that this bug is one of
>> many and we have a lot of work to do. So if a bug seems to be fixed
>> then we rather spend our time to fix other bugs instead of doing two
>> other rounds of communication just to confirm that the bug is either
>> really fixed or is still present in an older release.
>
> And the fact that it artifically reduces the number of open bugs, 
> artificially reduces time-to-fix and artificially increases the proportion of 
> idiot-user-reported-irreproducible-bugs doesn't worry you or your managers? 
> It's giving a biased view of developer/qa performance and of users.
>

This is insulting, but I'd like to state that I am not worried about 
"artificially inflating" any of these things, especially my own 
performance.

>>> I definitely *won't* be trying 1.0.6beta or any other beta.
>> Please reconsider this.
>
> The main problem I had with the program is its complete lack of 
> documentation. Until that's fixed, I'm just not interested in the program 
> because I can't use it even if it works. Also, in five minutes testing I came 
> across two bugs, either one of which is severe enough to make the product 
> unusable. That's alpha quality, not beta, IMHO.
>
> So I'll wait until the product appears to be stable.
>

Documentation is being worked on. Quality is also being worked on.
We are busy and have tons of things to do (Workbench is not even close to 
being our own project and we're just 4), but the Workbench is our 
"official pet project" and we are personally motivated on making it a great tool.

--
Alfredo



> Sorry, Dave
>
> -- 
> MySQL GUI Tools Mailing List
> For list archives: http://lists.mysql.com/gui-tools
> To unsubscribe:    http://lists.mysql.com/[email protected]
>

-- 
MySQL GUI Tools Mailing List
For list archives: http://lists.mysql.com/gui-tools
To unsubscribe:    http://lists.mysql.com/[email protected]