unnecessary font Provides churn (was Re: rawhide report: 20090116 changes)

[Ville Skyttä]

On Monday 19 January 2009, Nicolas Mailhot wrote:
> Le dimanche 18 janvier 2009 à 23:35 +0200, Ville Skyttä a écrit :
> > I think it's about time the people responsible for this mess
> > start actually
> > fixing the compatibility issues and make-work they have been and still
> > seem to be continuously inflicting themselves.
>
> So, appart from spitting on some community work you didn't contribute
> to, do not contribute to, and never bothered to provide useful feedback
> on

Hmm.

- Reviewing the templates, discussing how to best adapt rpmdevtools to the 
changes, and doing practically all of the actual work (not that there was too 
much of it): https://bugzilla.redhat.com/477055

- Pointing out how some of the choices made in the fonts packaging guidelines 
make them (IMO) unnecessarily Fedora specific and thus may cause extra work 
for Fedora contributors: https://bugzilla.redhat.com/477606

- Asking for some naming rationale: 
https://www.redhat.com/archives/fedora-packaging/2009-January/msg00000.html

- Private mail asking for assistance with two common fonts packaging related 
issues I regularly run into in server environments, something that at least 
at the time was not really covered at all in anything I had read (haven't 
checked later if they would be addressed).

- My previous (admittedly quite blunt) mail in this thread and this one, in 
which I'm pointing out things that I find arrogant or that have room for 
improvement.  I can't imagine that it wouldn't benefit you and everyone else 
if you took these into consideration and wouldn't start using expressions 
like "idiocies" when facing negative feedback.

Call me whatever you like, or shrug these off, but I think the above is not 
too bad considering I'm really not too interested in fonts packaging at all 
(read more below).

> (even when personnaly CC-ed with weeks of advance), do you have 
> anything useful to say?

How about this: Thankfully practically every invasive change people warn about 
on this list is accompanied with a "I'm going to fix these issues in CVS for 
packages I have commit access to".  I don't know why we haven't seen such an 
offer from you (or from anyone else who's driving these changes; I don't know 
who exactly they are), and to me it looks like you've also turned down an 
related explicit request to help and fix what you probably of all people know 
best how to fix, and should have the permissions for: 
https://bugzilla.redhat.com/480477

I thought it goes without saying that everyone's interested in non-breakage 
and offers to help when breakage can't be avoided, but only as a last resort 
when it's clear that it can't be avoided and when it's clear exactly what to 
do about it.  Personally, that's exactly as far as my interest towards fonts 
packaging extends to at the moment.  I don't know what made you think 
otherwise and send a personal Cc for the initial discussion [0] and why you 
still continue to refer to that after reading 
https://bugzilla.redhat.com/477606#c5 

> Because idiocies like "[not] tried very hard at all" won't get you very
> far. CVS, git, wiki, mailing list, and bugzilla logs tell their own
> (true) story.

Ahh, the i-word.  Anyway, in that case I can only shudder while thinking what 
it would have been without that.  Yet I still refuse to believe better work 
couldn't have been done with this with just a bit more research and taking 
other packages and packagers better into account, and that it's unfair to 
expect that kind of better work.  The real results for one example affected 
package can be read for example from the vdr-skins %changelog for 20061119-4, 
20061119-5, and 20081124-1, as well as bugs #477478 (includes 2 changes) and 
#480477.  Of those 6 changes I count 5 (everything except the dejavu-lgc -> 
dejavu change in #477478) that from my POV have resulted/will result in no 
benefits whatsoever, they're there just to catch up with backwards 
incompatible font packaging changes.

> Not trying very hard is writing long hate mails instead of doing
> one-liner changes you've been given all the necessary info on.

The reason I'm investing this much time writing about this issue is that I 
still for some reason hope that doing so might prevent some similar issues 
happening in the future.


[0] Not that I mind the personal Cc; at some point on various lists I started 
getting so many of them that nowadays my mail is sorted to folders based on 
mailing lists in recipients, no matter what the other recipients are.  The 
consequence is that I do read the mails, but in the context of the list in 
question and usually don't even notice if I'm personally Cc'd or not in 
messages that also went to a list.