[maemo-developers] [maemo-developers] Taste the Herring!

[From: David Weinehall]

On tor, 2006-11-30 at 10:53 +0000, ext Andrew Flegg wrote:
> On 11/30/06, David Weinehall <david.weinehall at nokia.com> wrote:
> > On tor, 2006-11-30 at 10:36 +0000, ext Andrew Flegg wrote:
> >
> > [snip]
> >
> > > If Maemo is to be an open and community participated effort, rather
> > > than just an open source Nokia SDK[1], the leaders of the project
> > > (i.e. Nokia) need to start informing the community of their plans for
> > > the platform.
> >
> > When did Linus last announce the release date of the next kernel (and
> > actually follow it)?  Ohhh, but I guess you don't regard that as an open
> > and community participated effort either. =)
> 
> Touché, however I said "plans" not "release date" (although I did ask
> about the release date as well). Plans involve other stuff related to
> frameworks, component versions, capabilities etc.

OK, our release plans for the next maemo release:

We will try to fix as many bugs as we can, updating the components to
newer versions where feasible, and do so within the next 6 months.

This is pretty much what we can tell you at this point.  We cannot
specify component versions, because we don't know that ourselves (bugs
or performance issues might force us to up- or downgrade).  As far as
capabilities go (I assume you mean features rather than the security
related kind), that's something we simply cannot divulge without top
management decisions, since that would commit us to a direction we might
have reason to deviate from.

We can of course express expected directions of development though,
and I suspect that Carlos will announce such things fairly soon.

Oh, and we will release the alarm framework within the same time frame.
That will have version 0.4.x.

As for people asking whether forthcoming releases will be free or not:
all our releases have been free (as in gratis) so far.  I don't know of
any plans to change this, and I suspect that our management realises
that such a move could kill the platform.  It wouldn't surprise me if
Nokia some day started selling add-on software; I haven't heard of any
such plans, but it would make sense (and I cannot see how that could be
objectionable).

If/when new hardware is released, the software released with that
hardware might not necessarily work on older hardware, although the
Maemo parts likely should be easy enough to back port (and the handy
hacker might even manage do disassemble the software images for the new
hardware and use some of the proprietary software from those on the
770).  The reason is simple -- making new software images work on old
hardware would mean that some hardware upgrades would simply never be
feasible.

And believe it or not, Nokia doesn't make any money from giving away
software for free.  It makes money from selling hardware.  So if we
cannot leave the option open for hardware upgrades now and then,
we don't have a feasible business case at all.  The Internet Tablet
concept isn't as simple as an mp3-player, where the same features can be
sold forever on and on with just a change of plastic covers and/or
storage capacity.

Now all of this should be fairly evident, and I'm sure that if you sit
down and think for a while, you'll realise that a company like Nokia
simply cannot commit to a five-year plan (or even one-year plan) for
something like this.  What if someone suddenly came up with an idea that
would render the Internet Tablet concept totally dead in the water?
What if someone launched this really cool and new cpu that required
1/10th of the current power consumption but required rewrites of the
software?  What if upstream development of some component we're using
suddenly stops -- we cannot maintain everything ourselves.  What if...

If the 770 had been Nokia's only product, or at least one of our major
products, we might have been able to make this kind of promises, since
we would've had a far bigger workforce, and could have easily hired
several hundred developers just to maintain the SDK (I'm pretty sure the
Symbian SDK numbers are in that ballpark, though I'm not sure, since
I've never worked in the Symbian division, nor ever developed for that
platform).

But face it, the Nokia 770 is a drop in the ocean.  Nokia is the largest
cellphone manufacturer in the world.  But we're not even a blimp on the
radar screen when it comes to handheld devices.  And the market for
cellphones is gigantic, the market for Internet Tablets is, so far,
minuscule.  Until this changes, any medium- and long-term platform
commitments will, I suspect, remain impossible to make.

Then again, all of this are my own ramblings (even though I'm posting
them from my Nokia account), so apart from the commitment on releasing
the alarm framework, you'll have to take it with a grain of salt.


Regards: David