View Single Post
Posts: 2,076 | Thanked: 3,268 times | Joined on Feb 2011
#295
Originally Posted by qwazix View Post
Qgil stated clearly that any more code will not be opened. And maybe there is no need to. After the time it will take either to raise the funds and close a deal with Nokia (assuming that is ever possible) or to reverse engineer the drivers for the N900, most of the N900's will be dead and the device itself will be pretty much obsolete. So we might just as well code new drivers for a device of our choice. What's the most open device out there? How much blobs can we scavenge from the android kernel to work unmodified?

Other than drivers do we really need anything else opensourced to go along? Of course the more code is open the better, but the most valuable of programs are open or we have nice open alternatives, the drivers are at least redistributable so the things that remain closed are device specific like mce etc. They won't do us any good in the long term anyway...

We do have a mail client, application manager, browser, telepathy, phone client (from meego), camera app, text editor, office suite, file manager media player.
We are missing a calendar (maybe free, alternatives exist, it's just a calendar after all) messaging application, gallery and all those IIRC exist on meego, so it's a matter of porting.

So much for the user applications.

Now for the core, we lack drivers, mce, battery management and more. But I don't know how much effort we should put in those as they may be have to be developed again if we ever move to a new device (in some time it has to happen)
This is I believe misunderstanding when it comes to view on source. Open-source projects do not die. Sooner or later (having the source open) there will be enthusiast or a group of such that will have a go at the code. Drivers are the limiting factor. It's cool some suggest the money would be better spent at 100% open device, but we know results of such endeavours. Very much disappointing/lacking. Having a fully working (at N900 level at least) stack would be something worth focusing energy/money on. Having all those: we're gonna bring you heaven by different approach projects that fail sooner or later gives us nothing. Freemantle (or bought up mantle) could result in cool things. Would be a matter of reverse engineering drivers for one device that is most open, most cool at the time. Now we have to do all reveng for all devices/drivers or pray. Porting open component would be just a recompile.