Re: Does the bug tracker actually work?



On Sun, Aug 06, 2000 at 03:30:48PM +0200 or thereabouts, Martin Baulig wrote:
> The problem is that such reports do not help at all; they even do harm. It
> means that someone needs to read them, find out that there's no useful
> information at all in them and then close them. This time could be used to
> fix real bugs.
 
On Sun, Aug 06, 2000 at 04:34:29PM +0100, Telsa Gwynne wrote:
> OK, I can see that point. I hadn't realised it would take so long to 
> skim and mark as closed, though. That does sound a pain.

Maybe this is where you oculd do with a full-time paid bug-analysing person?

As for stupid users, if you make the dsktop easier to use, you will
get an incerasing number of people who are significantly lss
computer-literate.  You will also start to get users for whom the
computer is a tool, and who have no interest in the technology itself.

These people tend not to be very good at reporting bugs.  I know my
partner doesn't distinguish between hard disk space and memory, and
yet is very intelligent.  He's simply not interested in how the
computer works.  When it doesn't work, sometimes he asks me if he
should uninstall programs he isn't using, to save memory :-)

As for making it easier to report bugs, one approach might be to
have a "this isn't working right" standard help item that first
gives a list of common tasks the tol does, then a list of common
tasks the tool *doesn't* do that people might be looking for, and
finally offers to help the user report a bug.

After a bug report is given, it might also be possible to do an
automatic keyword search for existing bugs that might be related;
this would help to reduce the number of duplicates.

If you encourage people to include a workaround, and maybe if
programmers add workarounds when looking at bugs, then other people
would have a good motivation to read existing bug reports.

If you need a text retrieval interface to the bugs, let me know :-)
My old lq-text package has been used in bug tracking packages
before, bcause of its high precision.  It doesn't try to do
natural language queries though.
<http://www.holoweb.net/~liam/lq-text/>

Lee [Ankh on irc.sorcery.net and irc.gnome.org]

-- 
Liam Quin - Barefoot in Toronto - liam@holoweb.net - http://www.holoweb.net/
Ankh on irc.sorcery.net http://valinor.sorcery.net/
author, The Open Source XML Database Toolkit, Wiley, July 2000
Co-author, The XML Specification Guide, Wiley, 1999




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]