[maemo-developers] Scratchbox and maemo development (was Re: sb2 wishlist)

[From: Lauri Leukkunen]

On 4/12/07, Quim Gil <quim.gil at nokia.com> wrote:
> Oh my.
>
> On Thu, 2007-04-12 at 11:29 +0300, ext Lauri Leukkunen wrote:
> > I'm not entirely sure how to interprete this. You're basically saying
> > that Nokia is considering dropping scratchbox and the debian
> > integration model and replacing it with openembedded?
>
> I wonder how you conclude this from my previous email.

Easily :)

But let's drop that for now.

> >  It would be nice
> > if you guys would say things as you mean them instead of going around
> > giving these vague statements.
>
> I'm saying things as I mean. If I would have an answer I would come to
> your office and then send an email here. But I haven't an answer, nor
> the rest of the maemo team. Surprise: nor we want to have any answer
> about SB/SB2 that hasn't produced together with you.

Yeah, but I'm not involved in the Nokia internal developer platform or
SDK discussions. The only thing I see (or mostly don't see) is what
goes on out in the open.

> The sensible thing to do when there is no clear/agreed strategy is to
> start defining a clear/agreed strategy. That is what my previous email
> tried to do.

Yup, and I seized the opportunity to raise the number one question
from SB2 perspective?

> From my understanding Scratchbox2 alone doesn't tell much for a
> strategy, specially considering that now maemo is still based on pure
> SB. We start having a strategy if we agree on i.e. plans to have a
> stable SB2 and plans to migrate there. I expressed this by typing
>
> > - SB vs SB2
>
> But wait, scratchbox is not the only toolkit maemo developers might
> need.

No it's not, but on the other hand Maemo is a rather important
potential user for SB2, and for historical reasons I'm giving it a lot
of attention. But at some point the attention needs to go both ways or
it becomes simple obsessiveness and that's counterproductive.

> In fact people out there are already coming up with good ideas,
> tests and products to develop applications for maemo without using SB.
> Should we ignore them? Should we support them partially? Should we
> integrate those tools in the official maemo offer?

That may be, I've not followed those developments too closely. But for
the Scratchbox project Nokia has some responsibility, at least I think
so. If SB2 is so uninteresting that the biggest user of SB1 doesn't
want to invest any effort to it, it kinda looks bad for the whole
concept, no?

Considering that with the effort that has been put into updating SB1
during the last few months you could've finished SB2 a few times over,
it's just puzzling.

> Once this picture is clearer then we can get into details about how to
> support the SB2 development and documentation. We could/can even support
> SB2 before getting this basic strategy in place. Let's talk about this,
> but from a wider perspective this is not a solution per se.

We're talking about maybe one or two man-month effort. After that
Nokia can evaluate how mature SB2 is. These things will be done with
or without Nokia's involvement. I just don't have the time at the
moment and it would be good for SB2 if others took a bigger role in
its development. As a holder of about 99% of copyrights in SB1, I
think Nokia should be there.

Otherwise it will be simply yet another proof that you can only trust
individuals, companies can't be trusted with important things.

/lauri