the "gamma" proposal - details that need working out

[Sebastian Kügler]

On Monday 09 July 2007 14:15:13 Tom Albers wrote:
> At Monday 09 July 2007 13:55, you wrote:
> > "Gamma" is a real release, but not targeted at new end users (so it
> > really *is* 4.0, and *not* a release candidate. That thing has to be very
> > clear.
>
> Now I'm confused. This is the text from Troy:
> "
> Our date for 4.0.0 would then be something like Jan 23rd or Feb 1st, which
> we really need to decide now so that we can start booking hotels and
> conference rooms and so forth in mountainview. "

This is probably caused by the difference in mine and Troy's understanding.

> > - A 4.0 release, especially in the case of KDE will have glitches. Those
> > glitches might be acceptable for our current user base, but they're
> > certainly not to new target groups that we're looking at for KDE4. We're
> > pushing KDE4 into the mainstream market which is much more critical about
> > small glitches. (This is based on research into target group
> > characteristics, which I'm happy to share if you want.)
>
> Is this any different that a new major release of any other project. See
> people avoiding Vista as the plague currently?

Well, in a way. We're trying to address this problem proactively. I don't 
think that Vista has anything to do with it, the problems that people have 
with it (to my very limited knowledge at least): Very little new, changes are 
mostly not what the user wants (lock-in), messy products (5 different 
versions), price. None of that applies to KDE.

So it is very different. The expectations of the new users we're aiming at are 
really different. Neglecting that by saying "Well, all dot-oh releases are 
crappy" (paraphrasing :-)) doesn't solve that problem. 

> > - Right now, it's quite hard for usability and accessibility people to
> > review KDE4, too much is moving. Same goes for translators. We need a
> > real release for those people to contribute in a meaningful way. (The
> > lack of people participating in HIG hunting shows this quite well.)
>
> After beta 1 everything is frozen solid. After that there are still months
> to go. Translators still have a good period to translate, and they can
> start already (as usual).

Only little do, in fact only those that are able to run KDE4 (alpha) are. I've 
discussed that with Olaf, and he said that the availability of the current 
versions is bad, and that reviewing applications that crash is just not 
possible (which they do now).

> > - KDE 4.0 will miss quite some applications, we might be able to ship a
> > basic set, but certainly not the comprehensive desktop we'd like to have
> > ready.
>
> I was not aware of that. Which apps are missing?

Just a few examples I can come up with right now: k3b, digikam (well, porting 
in process), PIM apps. Note that I'm not necessarily talking about stuff in 
kdebase or the default we're shipping, but also a lot of 3rd party stuff we 
want to give a chance to catch up. Another thing is plasmoids and extensions 
that make the end user experience a rich one.

> > - Marketing KDE4 towards the end-user, but giving distros virtually no
> > time (well, one week) to prepare a finished product renders it vapourware
> > (it's not available at that point).
>
> They have several releases to practise this. This is hardly an argument.

It's not the main argument, but giving some time to polish and integrate is 
crucial. Basically, lots of the 3rd party arguments above also apply here.

> > So technically, from a developer's point of view, it doesn't really
> > change all that much.
>
> Yes it does. As soon as we say that we are going to release 'gamma's'
> instead of real official releases and the official launch is early
> februari, a bunch of developers will sit back and watch E.T. again. I
> really feel it is the wrong signal.

You're right. We should make very clear what the real aim is, this really is a 
side effect we don't want. This is a reason why we shouldn't change release 
numbering, because for developers this *does* matter.

> > I think that the proposal is very sensible since it solves a
> > number of very real problems we're facing.
>
> If I understand it correctly, the 'problem' we want to tackle is that you
> want to release at a date close to a 'release party'-date.

Only partly. It's more about shifting expectations for new target groups. And 
having a really good "KDE End User Release" we'll make a fuzz about.
-- 
sebas

 http://www.kde.org | http://vizZzion.org |  GPG Key ID: 9119 0EF9 
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Remember, God could only create the world in 6 days because he didn't have an 
established user base.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 481 bytes
Desc: This is a digitally signed message part.
Url : http://mail.kde.org/pipermail/release-team/attachments/20070709/89daf05f/attachment.pgp