quoting, tasks, semantics

[Christopher Hicks]

On Mon, 13 Sep 2004, Gervase Markham wrote:
> I'd argue if you have a single bug for an entire web application, something's 
> wrong already :-) Regardless, I would close that bug when finished rather 
> than attempt to re-use it for post-release feedback.

I did close it and subsequently a few bugs have been opened and closed and 
one bug has been there for little spelling tweaks and such.  The main 
motivation was that we had reached over 100 comments, some of which went 
on for a few screens in and of themselves.

I understand that it may seem odd to have one bug for an "entire" web 
application, but in the minds of the people paying the money (who get 
bills that show time applied to specific bugs) and to me this small web 
application was "one thing to do", so logically it should mostly be in one 
bug.

> If it's so small that it's not worth filing a bug about, then it can be 
> implemented without discussion after you've seen the comment/got the 
> email. If it requires discussion, then it's worth filing a bug about 
> IMO.

You're welcome to that opinion, but given that the user has little way of 
knowing which tweaks are going to require follow up questions I don't see 
that it's practical.  And in most cases the follow up questions were 
eventually answered in a simple fashion, so there wasn't a lot of going 
around and around on the same issue.

> Also, getting them to file bugs puts a small barrier in the way, so that 
> only things they care enough about to spend 30 seconds filing a bug 
> report on get into the system. This might eliminate quite a few 
> frivolous requests.

Not likely in this case.  These folks write the checks that keep the roof 
over my head, so I'm truly not inclined to discourage their filing of 
bugs.  I have clients that I have to file all the bugs myself and they 
just get cc'd on the bugs they care about, so having clients that are 
willing to file stuff directly into bugzilla at all is a godsend to me.

[ Would it be wrong for me to point out that you seem to be following 
Joel's bad example of influencing behavior through software design? ]

>> In other words, tweaks and tasks are like bugs, but they're light weight 
>> and they don't require as much overhead. Practically speaking, expecting 
>> folks to make a new bug for every tweak would be ridiculous and a 
>> significant overhead for the bug reporter and the developer.
>
> It depends how good your customised enter_bug.cgi template is. ;-)

To some degree, yes, but mostly no.

-- 
</chris>

There are two ways of constructing a software design. One way is to make 
it so simple that there are obviously no deficiencies. And the other way 
is to make it so complicated that there are no obvious deficiencies.
  -- C.A.R. Hoare