On Thu, Sep 11, 2003 at 01:52:17PM -0500, Manoj Srivastava wrote:
> On Thu, 11 Sep 2003 12:47:39 -0500, Steve Langasek <vorlon@netexpress.net> said:
> > On Thu, Sep 11, 2003 at 10:39:18AM -0500, Manoj Srivastava wrote:
> >> >> And the contention of a lot of people seems to be that this is a
> >> >> trivial flaw, and we are not insulting your intelligence by
> >> >> implying that you can't come up with a description for your own
> >> >> package.
> >> > It has nothing to do with intelligence. It is about willingness
> >> > to help not just point out the weaknesses. If we all start
> >> > filling such bugs then we will soon end up in a nice mess.
> >> And the maintainers just waits and has things handed to them on a
> >> silver platter?
> > Yep, you nailed it -- none of us are doing anything else with our
> > time besides whining on mailing lists and waiting for complete
> > patches to roll in. Couldn't possibly be working on bugs that
> > actually *are* important to our users.
> Curious that you took a rhetorical question about the
> requirement of solutions with bug filings to apply to yourself. Also,
> the language construct I used does not in general imply that the
> question was addressed to all developers; it specifically targeted
> the proposition it was addressing.
> The fact that you think it applies to you implies three
> things, that you
> a) think that poor descriptions are not important to users, despite
> the usability issues
> b) do not consider the issue package selection, which is one of the
> blemishes of our distribution, to be important, since
> providing information to help select packages is not
> c) are too busy to habdle all flaws of your package, unless people
> who file these bugs also hand you the solutions.
I think that many of these bugs have been filed at an overinflated
severity, and should be prioritized below many bugs that affect users of
the packages in question -- rather than simply offending *non*-users of
the packages who would most likely never have noticed them at all if not
for this discussion. The premise that short descriptions hinder an
administrator's ability to decide whether or not to install the package
is false, predicated on the assumption that all pieces of software are,
or should be, equally accessible to all users. The standard set by
Policy 3.6.1 is as follows:
The description should describe the package (the program) to a user
(system administrator) who has never met it before so that they have
enough information to decide whether they want to install it. This
description should not just be copied verbatim from the program's
documentation.
The burden is on the bug submitter to establish that a curt description
does not provide the administrator with enough information "to decide
whether they want to install it" -- a metric that does not necessarily
require that the administrator actually understand what the package
*is*. There are plenty of packages in the distro for which more
detailed descriptions could only serve to entice users to think the
package would be useful to them when it isn't. Without such evidence
that a real user had a real problem making a decision about the package
in question, these bugs are by and large classifiable as
"unreproducible".
And as it is, yes, I am too busy to handle all flaws of my packages.
This shouldn't be a particularly surprising admission; there are plenty
of statistics which show the number of open bugs in Debian growing over
time. Given the number of bugs calling my name, therefore (including
some legitimate requests to improve package descriptions), I don't
expect to spend any more time on this particular thread.
--
Steve Langasek
postmodern programmer
Attachment:
pgpNmzqNWsClQ.pgp
Description: PGP signature