On 5/10/06, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
> > can you be more specific ???
>
> I can remove all software for which I cannot access and re-publish the
> source code and all of the hardware capabilities (power management,
> DSP and whatever other little doodads Nokia packed in there) are still
> accessible; I can compile a kernel and a corresponding initfs and root
> fs from source. Etc.
Those are the goals but i have to admit we have not yet reached there. We are
working on most parts e.g you can compile your own kernel, create your custom
rootfs,
remove stuff etc with package management. There are still certain piecies which
are
either not in our direct control (e.g TI/DSP related stuff), and some which we
just cant give out (e.g battery related software).
As far as the DSP stuff goes, if a developer has linux-dsp-tools,
what's the issue? The actual code which implements the DSP tasks
should be distributable under just about any license, no?
For us it is important to strive a balance at product creation and at the same
time persue the openness goal
I certainly understand that being a software developer by profession
myself. However, to some extent, the two goals are either at odds, or
the second one is actually more important, to whit:
If Nokia has a product which is viable in the marketplace as it
currently stands, it actually doesn't need the community and can do
only what it has to do legally to comply with the licenses of the Free
Software it ships. In which case pursuing openess is just a
distraction and internal development effort focussed on that goal is
probably wasted. This of course puts the community in a position where
it has little option but to reverse-engineer the parts that Nokia
won't release and pursue whatever course makes the community's
software usable and effective, even if it means forking.
On the other hand, if Nokia is hoping that it can leverage the
comminity to *create* the software that makes the device a compelling
product for consumers, then it's in a bind; it must please the
developer community *first,* so that we'll be motivated to do the
lifting required to get the product to that point; speaking
personally, my own enthusiasm for the device is directly tied to its
openess, and my interest in contributing is entirely linked to this
(hence my not developing software for the various Windows-based
tablets and handhelds or purchasing said devices).
To put this in a slightly different perspective, compare it to the
situation in desktop and server operating systems: it is widely
accepted that it's the open source development process itself which
makes Linux a stable server platform, so it's strongly in the
self-interest of vendors shipping said hardware to support *the
community* well. On the other hand, desktop users are used to crappy
software and expect all the latest whizbang USB devices to work with
their machine out of the box; it's less clear why vendors selling
products in that market should care about fostering open-source
communities because their bottom-line depends on getting products to
market quickly with (perhaps poorly-functioning) device drivers for a
million devices they don't control.
I'm not sure which category Nokia finds itself in with the 770 (I'm
prepared to admit that I also may be creating a false dichotomy here,
but in order to be convinced I'd want to hear evidence); however, in
order for me to expend any effort on making the device a compelling
one, I need it to be compelling for me, and for me what makes any type
of computer or electronics device compelling is that I can run it
w/out using proprietary software. I don't mean to lecture here, but
that's what I believed I was getting myself into when I bought the
770.
Anyway, I of course appreciate you taking the time to respond and to
engage in the discussion, I seriously hope that it eventually becomes
moot due to the device living up to the openess promise (and sincerely
hope it doesn't become moot because it fails in the marketplace before
that happens).
Dave
_______________________________________________
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers
