Apple is going to beat all competitors

[Josef Wolf]

On Fri, Sep 07, 2007 at 08:27:36PM +0200, Andreas Utterberg wrote:
> 
> What the v2 neo needs is a nice oi, the best would be if its possible to add
> compiz fusion, beryl effects to it. That would really boost the interest to
> the mass, just look at the development speed to the berylproject had, and
> the very big community around it in a very short amount of time.

Since you mention development: IMHO, it is essentially to make it as
easy as possible for interested developers to jump into the pool.

Last time I checked (I must admit, it was almost two months ago) you
needed to jump through several hoops to get running: first learn about
OE and its bitbake recipes, then there's the overlay tree.  When you
dive into OE, there come more dependencies, and they tear you further
into even more dependencies.  All the information is spread through
across the wiki.  I started searching for details, but along the
way I got lost.  I realized that it would take weeks to set up and
customize the build environment (I prefer to really _understand_
what I set up, not simply install and go).  I don't consider me to
be a beginner, since I work on embedded systems for almost two decades,
and linux (albeit, not embedded) is my main OS for about 15 years.

I'd rather concentrate on implementing something on the beast instead
of searching for days for information how to set up the build environment.

Coming back to your point above: I contradict your opinion.  Neo is _not_
about mainstream and it is defintely not targeted for people who would
buy an iphone.  Neo is about freedom.  People buy iphone for effects,
but people don't buy Neo for some compiz/beryl effects.

> Also important to get the device out for sale as soon as possible,
> the sooner the better,  that  would get more people involved because
> i think may is waiting on the v2 to jump on the train, im one of
> them :-). Better to get it out now with more bugs, then later with
> less bugs.

I'd contradict here, too.  Getting out early (even with serious bugs)
to developers (or people who count themselves as developers while not
beeing an official developer) is OK.  But before going to mass market,
some stabilization needs to be done.  We don't want hear "this device is
totally unusable" from the masses, do we?

IMHO, at the present time, the most importrant task is to make it for
interested people as easy as possible to jump into the game (this
includes easy installation/customization of build-environment).  Given
enough critical mass, more and more features (eventually compiz/beryl)
will be added to attract the masses.