Developers phone also fit for early adopters?
Derek Pressnall
dpressnall at gmail.com
Fri Jan 26 20:11:43 CET 2007
> The actual point was a different one: there is no reason to shy away
> just because it is not perfect yet.
Actually, there is. Studys have shown that in relation to the Bazzar
software development model that even though "release early, release
often" helps at some point, most successfull projects have there
initial release done when things are developed to at least a certain
level of functionality. Also, that very early development is usually
done by a single or a small hand full of developers that have a close
shared vision and clear seperation of tasks. (I don't have a reference
to the studys at the moment, but I remember seeing it before).
The reason is this: At the initial stages, the software is rapidly
changing (not just being added to, but quite often whole sections are
modified / re-coded, api's change, etc). So early code releases
aren't useful outside the core developers for this reason. Also,
until there is a reasonable framework, it would be counter productive
to the core developers to have people constantly hammering them with
questions / "advice" on code that is still in flux (until they are
ready for such outside help).
As for the "staged" hardware release, it takes a while to get
production ramped up. From what I understand, the first few samples
that come off the line are what is reserved for the core team, and it
is more important for them to get what they need to develop than it is
for outside developers to get samples at this point. It is entirely
likely that the devices shipped out Feb 11 are engineering samples
(usually produced on a small scale development-oriented line
equipment). Any last-minute hardware bugs are worked out during this
stage, which gives time for any tweeks to be made prior to the Mar
11'th rollout (when the big production lines will be cranked up).
Also be aware that the engineering samples are very expensive to
produce (I'm sure they are more than the $350 price point for the
final product), so it only makes sense for them to go to the core team
at this stage. Compare to the price of the Qtopia greenphone ($700),
which only had 1000 units manufactured in the first run.
Note, that the above is only my speculation based on prior experiences
working at manufacturing companies, I have no insite into how FIC is
actually doing things; though I would expect that their processes are
similar to what I've observed in the past.
More information about the community
mailing list