Originally Posted by demolition (Post 1016893)

The request this petition makes is most audacious, to say the least, and in the unlikely event it works, it would be quite a triumph. On the assumpton that Nokia (I include all Maemo-device partners, as well as Nokia itself, in that word) will not release any PR-type updates, releasing code would allow the community to make improvements.

I believe, however, that a more refined question may yield some results because without more focus and weight, the reponse will always relate to IPR and so-on. Incidentally, much of that page relates to Maemo4, not 5 and doesn't appear to include drivers/3rd party s/w (point in top para noted).

The reqest could be refined by breaking it down into a list of the areas of functionality where source code or system-access is necessary i.e What source should be released. Many items on this list exist as wontfix and open on the Maemo bug site. For each item on the list there needs to be an associated:
.

1. Why source should be released:

   Although the existing licence alteration scheme has not proved very sucessful, while devices are in use (and phone contracts are still current), there is a case for code-release or official updates (cough!). The release of code might be based on one or more of the following reasons (there may be others too):
   
   (a) Fix those parts of the OS (inc. drivers) that prevent the device operating as advertised, reasonbly expected* or that make it in anyway not wholly fit for purpose. If provable, There might be some grounds for consumer-rights action in this area. The graphics driver (I know it's not just Nokia Corp. involved) is one of the most obvious areas where there's a difference b/w advertised and actual performance - there was no mention of frame-dropping for video playback or image-shearing in the press, was there?
   
   (b) Allow the OS to be extended/improved. Probably not really in Nokia's (or ms's) interest to keep old devies alive unless such an extension/improvement addresses something listed in (a) OR shows a proof of concept that can be used (by Nokia etc.) in future devices.
   
   (c) For application s/w development. Again, unless such s/w addressed something listed in (a) or showed proof of concept, which might help Nokia etc., this argument has the least weight (I beleive).
   
   (d) Allow the device to operate to its full potential e.g. formal BT input support or UI improvments to allow direct printing. This might be possible if release of code wouldn't harm sale of future devies or Nokia etc.; it would be an easier decision for those areas only concerning Nokia. Would it be a valid argument to suggest that doing so would promote the sale of non phone Nokia devices (future or existing)?

2. How source should be released:

   Does the _full_ source (even of requested sub-sections) need to be released, or would header files with precomiled (obfuscated if necessary) lib/[binary-type] files be sufficient? (I'm Assuming that interface and implementation can be separated.)
   
   For OS development (fixing problems), clearly the whole source is necessary, not just the interface; if the implementation of a given class/function is completely rewritten, then the existing (closed) version maybe done away with altogether. Again, for application s/w development one only needs to include a well commented header file to make new applications.

3. To whom source should be released:

   From all the people on the "Yes list", I'm sure many do not have s/w in the Maemo repositories (though many may report operations as testers) and whilst being a vocal member of a community is important, for everyone to have access to the source is, I think, not necessary.
   
   Also, as someone mentioned, who would be liable if "improvements" were found to make the device dangerous to use, or a dependecy clash caused 50% of devices to brick?
   
   Who would sign-off any OS fixes/improvements/extensions?
   
   One idea might be: if projects requiring currently unreleased source had to register via the council (or a dev sub-committee, or similar) and developers needed to register with the project to obtain these sections, some trace of what was going where could be made, then perhaps even the most sensitve parts might get released and fixed?

How often has such a list been presented to Nokia? It would be useful to have summaries of these meetings posted in the News and Community sections.

Some parts of Maemo are more sensitive than others so each could be handled differently. And, a lot of Maemo is Debian, so much of what is needed for application s/w development (not OS development) is already open. That which is closed, seems to be mostly drivers. The advantage of breaking it down would be control, the disadvantage would be spending time on an OS which cannot bring in any more revenue. Though, there maybe an obligation to do so if suitable proof can be found.

I think most N900 owners (and probably N8x0 & N770 owners too) believe the(se) device(s) is/are wasted opportunities, where something truly great could have been realised but for some reason Nokia failed to pull it all together. By releasing relevant pieces of source the community (developers, testers and doc'ers) fixes/improvemts could be sought and users would be happy.

On the one hand by providing good customer support, in the form of updates and source releases etc., Nokia would encourage users to fly its flag who would in turn be more likely to buy another Nokia product. While on the other hand, by offering a very short life-cycle Nokia are forcing users to switch devices, which might seem like a good way to get revenue but makes users less likely to care about who made said device; also, not very 'green' as Nokia always proport to being.

I'm sure some will pour scorn over this post, if for for nothing else, for its length. However, this is not a rant, just a discussion from an occasioal coder and current Maemo user.

The device(s) and OS(s) are very good so attempting to smooth the rough edges is, I believe, a worthwhile venture; clearly many others do as well, as shown by the interest in the CSSU and the growth in the number of programmes in the repositories, even since Nokia told some commercial developers to give up on Maemo.

Re: my vote: I'm still on the fence because, as it stands, I don't think it's a viable (and in some ways reasonable) request. Also, I'm not sure a yes/no poll for source-release is the best way to achieve bugs fixes and making Maemo more extensible because there's more to it than that (what, why, how, to whom, etc). But, I do agree with the sentiment: either fix the errors or let us (Maemo community) fix them ourselves.

*reasonably expected: only pertaining to existing features e.g. fix not being able to turn device on while charging.

Originally Posted by Daneel (Post 1017114)

You are our self proclaimed leader so step up to the plate and go through the buglist and identify a list of sourcecode items you would like to see released, in order of priority, with reasoned arguments as to why, how and to whom.
That is actual work you know, something that takes time, effort and might result with something useful, not this dumb to oblivion poll of yours.

Pretty please.

Originally Posted by Hurrian (Post 1017124)

*ahem*NT/2000sourceleak*ahem*

Originally Posted by Texrat (Post 1017576)

I was originally 100% behind this but after being informed how low the code quality is, I think moving to MeeGo and developing a Maemo UX/UI is the way to go.