djdas djdas at djdas.net
Sun Mar 7 16:46:19 CET 2010

Neil Jerram wrote:
> On 6 March 2010 22:24, djdas <djdas at djdas.net> wrote:
>> As the team leader of the Neophysis Dev Team I'm proud to announce you a
>> new distribution for our loved(?) Freerunner.
> Well, of course it will be great to have another choice for the FR, if
> it is rock-solid and supports all the important phone functions.
Yes it's our first target, a fully functional phone.
>> - the phone stack should be oFono (reliable, multi-platform, fast)
> Is oFono already rock-solid as regards GPRS?  i.e.
> - doesn't drop the GPRS connection prematurely
> - allows simultaneous use of GPRS and phone calls
> - deactivates GPRS cleanly (when requested) and can reactivate it later.
Honestly we didn't try GPRS functions yet, because we focused on the 
phone part for the Armeniacum release, although oFono team implemented 
the GPRS functions and the Calypso plugin already works with 
multiplexing, so I assume there shouldn't be many problems.
> Also, is oFono better than FSO as regards
> - never missing incoming SMS messages?
> - handling split (CSM) messages?
Yes and yes :) never missed any SMS and long ones are automatically 
handled by oFono: you can send/receive a string of whatever length 
(don't know what is the limit but at least with a 3 SMS long string, I 
had no problems) without taking care of the splitting/joining.
>> - a completely cutomized boot sequence to start in 30-50 secs maximum
> Nice - but more for the implication that you have carefully considered
> all the daemons and cut out what you don't need, than for the actual
> boot time (which doesn't matter much if one rarely needs to reboot).
I always turn off my phone at night so for me a fast boot is mandatory 
;) and yes, we removed all the unuseful stuff (at least on a device like 
the FR)
>> What you can expect: well, it's in alpha stage, the only working thing
>> is the dialer for placing and answering calls (no contact-list, just
>> numbers :P) no SMS (we're working on them ATM), no suspend/resume, it's
>> a simple phone :D it boots in about 40 seconds with the modem registered
>> (NO PIN SUPPORTED ATM!!!), it works only with moko11 firmware; to
>> shutdown, simply push the power button for 8 seconds (read-only
>> filesystem so you can't break anything).
> Are the limitations because of middleware problems (i.e. oFono) or
> just because you haven't written the UI yet?
The second one ;) we are completing the SMS handling part these days and 
working on the contacts handling, the rest is because we didn't 
integrate FSO daemons yet but will get on soon :)

>> Last but not least thanks to anyone who will believe in our project and
>> will want to join us in this new adventure.
> I will believe in it if it produces a rock-solid and full function
> distribution soon.
Soon is a word which fights with "time and resources", we are doing our 
best and we are full of ideas but work on it on our spare time so please 
be patient :) and please let me be a bit ironic: in three years we 
didn't get a rock-solid full function distro yet, in 6 months you can 
almost phone ;)
> For me, it raises the question of why people feel that they need to
> start off their own thing, instead of contributing to an existing
> project.  I think one justification for this is if the new project
> really produces a result quickly that is clearly better than the
> existing projects.  If that doesn't happen, I'm afraid I don't see the
> point; the end result is only that we have N+1 not-quite-good-enough
> distributions, instead of N.
looking at distrowatch.org it seems many people think differently :) 
joking apart... we started because we thought something faster and 
reliable than the current distros (almost for the phone part) could be 
possible by rethinking from scratch every single component of the 
system...future will say if we were right 6 months ago ;)
> FWIW, the same question obviously applies to oFono too (vs FSO).  I
> read some of the IRC logs about that, and the oFono guys only seemed
> to have one technical reason for starting their own thing - because
> they thought the FSO D-Bus API was too complex and exposed too much
> unnecessary detail to applications/users.
I think so and I agree with oFono guys. Furthermore oFono doesn't want 
to substitute or concur with FSO, it's just a phone middleware, while 
FSO is aimed at the whole system so I don't think we can compare them.
> Which on the one hand is not persuasive - because I'm sure it would
> have been possible for them to add their proposed simpler API to FSO,
> in parallel with the existing API - and on the other hand is a concern
> - because it makes me wonder if an oFono-based distribution would be
> able to support applications like Cellhunter?
I think FSO is a great idea but I think a middleware should be optimized 
(read as: wrote in C and easy to handle). We will use FSO for the parts 
which do not need "realtime" handling and this is why, for the phone 
part, we choose oFono.
> So, in summary, I have misgivings about your choice to start your own
> thing from scratch, but I wish you well and hope that Neophysis will
> make rapid progress.
Thank you, I simply kindly ask you to try it and then post your feelings ;)
> Oh, and finally:
> "source code of our libraries and daemons is not yet open to anonymous
> readers but only to active developers"
> I hope you understand that this means that Neophysis is NOT free (or
> open source) software.  I encourage you to fix this very quickly!
Technically speaking ATM it's free but not open source, but this is only 
because our git is full of test branches, source code is not yet well 
commented and organized, we plan to open it next week or at most the 
next one (for sure before the end of March), just the time to give a 
quick clean :)
We in no case want to be closed and we need contributors and developers 
so it's not an "issue to fix" but a target :)
> Regards,
>         Neil
Best regards,

More information about the community mailing list