Re: GNOME 2.23 Schedule

[Olav Vitters <olav bkor dhs org>]

On Wed, Mar 19, 2008 at 12:12:59AM +0200, Felipe Contreras wrote:
> On Tue, Mar 18, 2008 at 10:16 PM, Olav Vitters <olav bkor dhs org> wrote:
> 
> </snip>
> 
> >  Yes. Ubuntu is acting as a translator between user and developer. Which
> >  is what I am suggesting you could do as well.
> 
> Right, and as a Fedora user I find it interesting that in order to
> find things to improve in GNOME I have to go to the Ubuntu brainstorm.

I said elsewhere I'd appreciate if you'd help of such things. Changing
this into 'I have to go in the Ubuntu brainstorm' is again a negative
way of seeing interpreting this. Brainstorm is indeed just Ubuntu, so go
ahead and start something more general (you'll get help.. after some
legwork is done).

> >  Ehr.. 'unmount resolution': I assumed you couldn't unmount because of
> >  some process and you want to kill it (e.g. bash process cd'ed in that
> >  dir). You could even think of something that wouldn't need the process
> >  to be killed, but would instead save the state locally ('gedit was using
> >  that USB, don't worry.. it'll be written next time you plug it in').
> 
> You don't need the process to be killed; you need the list of
> processes using that and the option to kill them.

But killing is still far from what a user should be concerned about.
E.g. what is the best choice to show the user. I'd be interested in any
work that figures that out (e.g. tests it using some usability lab).

> >  > Anyway, the fact that it's an important issue remains.
> >
> >  I don't see that as a fact (in the meaning: 'stop everything, work on
> >  this now asap'). I agree that there are a lot of things that should be
> >  fixed/improved.
> 
> Of course not, the developers decide what gets done or not.
> 
> important != work on this asap
> important = carrying or possessing weight or consequence

Understood.

> The fact that 3,000 seem to want this feature makes me think it's
> carries weight.

IMO 3,000 users clicking some button isn't a decisive factor. There
needs to be some prioritization, etc. Plus the difference between what
users say they want and what they really want (isn't about ignoring.. is
that if you follow a request from a user literally you aren't always
providing the thing the user wanted).
I'm trying to make clear that such work is appreciated, not easy as it
might appear firsthand, but is wanted (well, by me at least).

> </snip>
> 
> >  > Perhaps the roadmap is not a good place for that, perhaps some feature
> >  > list, perhaps even the bug reports could help, but you need to *see*
> >  > the votes.
> >
> >  Votes don't work. Really, the most popular bug on Mozilla Bugzilla has
> >  around 750 votes. There are millions of (Firefox) users out there. The
> >  750 is not significant (nor a representation of the Fx target group).
> >
> >  It only results (on Mozilla Bugzilla) about 'XXX votes, why isn't it
> >  fixed?!?', etc. Plus discussions on how many votes each person should
> >  have.
> 
> Some voting systems might not work, the most basic ones that I have
> seen work have the option to vote negatively.

My worry with voting systems is that they likely not represent the
target group. E.g., if you have a good representation of e.g. all tv
viewers, you could pretty accurately determine who many millions watched
some programme by only checking at most a few thousand people.

With the online stuff I've seen, you will be ignoring a lot of users.
Think of e.g. GNOME usage within companies, Extremadura (forgot the
numbers, but was at least 100,000 users), etc.

However, if someone sorts this all of this out.. all the better.

> Still, only-positive systems at least gives you an idea.

If there are enough votes (100,000), yes... but could very well be that
it was on the frontpage of some popular website. Like getting autographs
for some 'cause'.. you'll get them even easily (e.g. every once in a
while someone 'proves' this by getting loads of them for some cause that
doesn't make any sense.. not dismissing Brainstorm here btw).

> >  > Take for example the maemo bugzilla, you can easily see the most wanted things:
> >  > http://tinyurl.com/2eajzg
> >  >
> >  > There are potential new developers out there but I doubt they would
> >  > want to work on a feature nobody else cares about.
> >
> >  So that is why I gave that Jokosher example.. shine by doing.
> 
> Jono Bacon started it, which I bet was major undertake and which
> required a lot of knowledge most of which probably he already had.
> Also he probably knew enough people that would like such a thing.

'Knew' in the sense that he is vocal. Which means you need to talk about
it, etc.

> Most potential developers don't have that clear idea about how to contribute.

That is the intention of GNOME Love and various related things. Think
Google Summer of Code and that task thingy. I suggest to look at:
http://live.gnome.org/GnomeLove

IIRC for GSoC the admins do try to select the most promising ideas.

> >  Which application? Pidgin? I don't see that as GNOME, but as a Gtk using
> >  application. For GNOME, see http://www.gnome.org/start/unstable for the
> >  list of GNOME applications. Then there is also the GNOME infrastructure
> >  using stuff (on ftp.gnome.org). If an app isn't even on GNOME infra..
> >  then d-d-l is likely not the mailing list you want.
> >
> >  I'll refrain from commenting about Pidgin specifically.
> 
> I don't know what the objective of that comment was but for just FYI:

Commenting on Pidgin would be offtopic, so I said I wouldn't.

> libpurple + telepathy-haze = msn on Empathy, and later perhaps
> telepathy-msn-pecan.

Ok.

> >  > I will develop somewhere else.
> >
> >  ?!? and now suddenly this is about developing? I don't get the
> >  intention. Whatever you want to improve, go ahead and do it. If during
> >  that you need help / have questions: shout!
> 
> I'm a GNOME user and FOSS developer that would like to know which are
> the features users want. Knowing that is a step closer into making a
> consequential contribution.
> 
> Now I know d-d-l is not the right place for that, neither is the GNOME
> community, I'll better start on Ubuntu brainstorm, like apparently a
> lot of people are doing.

Well, I did not mean to imply the GNOME community is not the right
place, more that it isn't just up to developers (is restrictive to limit
to developers). GNOME is bigger than that.
Note that there have been various usability studies done (see usability
archives), but those focus on usability issues within existing
applications (AFAIR)... although one result was that gnome-main-menu
(new app).

-- 
Regards,
Olav