Re: Tomboy in Desktop

[David Nielsen <david lovesunix net>]

man, 24 07 2006 kl. 22:58 +0100, skrev Alan Horkan:

> > I have no gripe with sticky notes,
> 
> Doesn't sound like it.  You are advocating it be removed and the Tomboy
> author is not.

I honestly don't, my gripe is with duplication of functionality, if we
include Tomboy, which I believe was the discussion at hand. We would be
providing the basically same functionality twice, this seems to go
against what I've come to know and love about GNOME over the many years
I've been a part of the community. In fact the very feature that
converted me from KDE long ago got replaced, acme or whatever the name
was of the first no frills, no configuration, just work multimedia
keyboard thing. The functionality got even better by being replaced in
my opinion.

> It is all too easy to disregard all the work put into more peripheral
> tasks or documentation and ongoing maintainance.

I'm a translator, trust me I'm not disregarding tasks aside programming,
in fact I translated Tomboy just the other day and I'm about to put my
work through peer review within my designated team. 

> APIs get deprecated, but applications get removed entirely.
> Sometimes the option to keep using what you were happy with really is
> better than having to learn a new different application.

I have yet to actually meet a user who used Sticky Notes, but lets say
they exist and are plentiful, I would have expected at least one to
chime in and make an practical argument against removing it. However if
none do and no concensus can be made on replacing Sticky Notes what do
you propose then.. I would be perfectly fine with keeping Sticky Notes
on for a few releases although I'm afraid it will end up never getting
removed just like I'm afraid that will happen to gnome-cd-player. If we
can make a cut off date sufficiently far in the future would that work
for you, they both die say in 2 years. That should after all be plenty
of time to write documentation of all kinds and you have my word I will
help out all I can.

For these kinds of things it would be really nice to have the same kind
of system the kernel uses, we mark something off as deprecated with a
given date in this case we could say come branch time for 2.21 the
following gets removed, please stop relying on it today. 

If we can't outright remove a given program at least we should agree on
a policy to do so gracefully in all cases henceforth.

> I'm not suggesting Sticky Notes be made tomboy but the whole process of
> removing older applications bothers me.  I'd like to see them
> frozen/deprecated in some way and remain available and then maybe removed
> at some major milestone or at least after a few releases.  This really has
> nothing to do with Tomboy, I felt similarly strongly about keeping
> gnome-cd-player when sound juicer was proposed.

Yes and when exactly did you imagine we'd get to remove it? This is why
we need a policy and a strong leader. No even I'm not that arrogant as
to suggest that would be me - in the past Jeff has served this position
as our pantsless leader, though his current commercial ties might cause
for accusations of bias. If the foundation could afford it that would be
one position I'd love to see filled. Someone like Quim Gil has shown a
tremendous ability to coordinate and energize people so I'd probably off
hand propose him or someone with the same qualities.

Actually I've thought about that for some time now, ever since GUADEC
where I happened to catch some of the board meeting via the stream. It's
also apparent by listening to the community that they think we flutter
around to much and need direction. A strong leader and some road maps
for desired goals would do us some good and help bring back the
community's faith in us.

> Stability for those who do use it.  Why force users to change?  Dont try
> to assume you know best.

Why force people to do anything, why force them to use GNOME - I believe
the majority will elect to migrate for reasons of functionality and
stability of these new products. But we have forced change on them
before, gpdf -> evince, gtk1 -> gtk2, sometimes forcing users a bit is
needed for progress to be ensured. Imagine if we never removed code.

There are also other ways of ensuring stability than keeping code
extensive use of build testing would be just one (maybe we could look at
GStreamer here, they have a fine set of tools in use but we can improve
on them over time, I know Fredrico would probably like some automated
performance measurements as well e.g.). 

Stability is also for those that seek it, GNOME is pretty damn stable
but we can do better.. Did we ever turn the critical to fatal thing back
on again for development, it seemed to hit bugs left and right, great
testing feature for those of us who have the strange hobby of filing
bugs.

I apologize for any stop energy I might have released accidentally.

- David Nielsen

http://osnews.com/story.php?news_id=15266