Current situation of amarok, and of latex tools

Daniel Chen seven.steps at gmail.com
Wed May 13 18:39:16 UTC 2009


On Wed, May 13, 2009 at 2:07 PM, Vincenzo Ciancia <ciancia at di.unipi.it> wrote:
> I hope this will not sound like a complaint.

It does.

> The problem
> is there, and it's grave.

Grave for whom? For you? For what common use cases? These are things
that are factors to consider when affecting an entire release.

> In my opinion, the switch to recent versions of some programs has been
> done without the needed testing, (or, like in the case of the intel
> driver, without taking seriously the response from testers) and results
> in a completely broken or very badly usable system for many. In the
> latest release of ubuntu, I mean.

There has been no lack of calls for testing. Some of these calls have
resulted in timely and effective bug reports. Others, not so much. I
doubt testers' responses have been blithely ignored. I know I
certainly didn't regarding the audio stack (woe tho' it is). Some
regressions are more serious than others, and like most bugs, there
are sometimes quick workarounds. (Sometimes the entire audio stack has
to be ripped apart, but that's irrelevant.)

One pivot is "how many users will regress if we do X instead of Y?" It
is not useful to base a decision solely on popular outcry. There are
far too many hardware combinations. People whose Ubuntu installs seem
to work rarely pipe up and complain. How do you account for them?

> THIS IS BAD FOR UBUNTU!   THIS IS BAD FOR UBUNTU!   THIS IS BAD FOR
> UBUNTU!  THIS IS BAD FOR UBUNTU! THIS IS BAD FOR UBUNTU! THIS IS BAD

It's a very thin line between complaining and fixing the bugs, but
motivation may distort one's vision.

> I messed up my ph.d.
> thesis with it today, then in complete frustration reinstalled kile
> 2.0.1 from intrepid, which works like a charm.

I understand your frustration. I, too, have a day job. Are you
spending your free time fixing kile (and/or kdvi)?

> What is it doing there in a stable release? This program has not been
> tested. It is not stable. People does not like it yet.

Are you seriously saying that Amarok has not been tested?

Also, I'm unsure what part of "liking" (which is subjective,
regardless) an application actually goes into bundling it into a
release. As far as I know, Amarok has existed in some form for most,
if not all, Kubuntu releases. To remove it would constitute a
regression. To bundle an older version would result in complaints
regarding an outdated version. etc.

> fixed, but why shipping a broken program in a stable distribution?

> Now this can't be my fault. Nor yours: you wanted to
> get rid of unsupported applications and that's good. But it was way too
> quick as a move. Next time a bit more testing will help.

Or...you could step and take responsibility for some part of the
distribution/release process. The line between complaining and fixing
bugs really becomes thinner, then.

> The new intel driver was
> and is broken. Upgrading has been a grave mistake and users are seeing
> an ubuntu that deadlocks.

Again, for which users on what hardware? Do those users constitute the
majority of people using Ubuntu?

> It seems to me that too much trust was put in the fact that it'd
> have been fixed.

How much of this complaining would be moot if you had contributed upstream?

> The ubuntu procedure for testing, in jaunty,
> seems not to have worked in some points. Next release can be better also
> from this point of view. Perhaps by just listening a bit more to
> regressions (it seems my favourite topic?).

What constitutes a regression for you may not be a regression on
someone else's install. Are you more important than that someone else?
Whose install should break? "Neither" is an ideal but impractical
answer.

Dan




More information about the Ubuntu-devel-discuss mailing list