On 21/06/2010 6:42, Dirk Meyer wrote: > Hi everyone, > > yes, I'm not dead. It was a very busy year and I had no time to hack on > Freevo. But I plan to add a Freevo-day each week in my calendar to start > coding again. > > On 17.06.2010 17:58, Adam Charrett wrote: > >> Freevo 1.x is still going (reasonably) strong although slightly long >> in the tooth and Freevo 2, while further forward than a couple of >> years ago, isn't yet ready for release. >> > Sadly, no. > > >> So in the interim I'm thinking about the following changes to the >> Freevo 1.x code base with the idea of calling it Freevo 1-NG (not >> quite 2 but no longer 1.x). >> >> 1. Getting rid of pygame and replacing it with WebKit and pygtk. >> This means all the skins would be written in HTML and Javascript >> (lots more people able to do that than there are for our current skin >> format)) and we could then have video in the same window! >> > Well, this is one reason I started 2.0. But stripping the GUI code out > of Freevo wasn't that easy. That part is mostly done in Freevo 2.0. I'm > not sure I like the idea of 1.0 NG. It seems to me that it is waste of > good man power. 2.0 is closer to being finished than you think, maybe > your skills are better used on the 2.0. > > Let me go into details of the SVN head. There are two Freevo versions. > One is based on mevas (imlib2 based rendering) and the version I use > right due to the fact that the clutter-based Version has no video > playback, yet. But it is deprecated. > > Now let's take a look at the clutter-based version. The internal changes > I wanted to make for 2.0 are done. The core is ready for a release. The > whole gui code has moved out of the core. Besides the idlebar plugins > nothing depends on the GUI. If you want to build a gui based on webkit, > it shouldn't be that hard. If you want to use GTK, you are already done > -- sort of. The code is based on clutter which is wrapped in kaa.candy. > Clutter is the 3D-toolkit Gnome 3.0 will use for mutter and other > eye-candy stuff. > > If you want to port a plugin from 1.x to the 2.0 version, you have a > green-light from me. Again: the core is ready and the basic plugins such > as video, audio and image are working. The video plugin lacks actual > playback support. I was waiting for Jason to develop a new kaa.popcorn, > our player module to render the video on a clutter texture. Since Jason > has the same time problems as I have, I'm not sure I should wait for it > any longer and my next hacking days will be spent on video support: > > 1. Use gstreamer to render the video inside clutter. There already is > support for it in a clutter-gst package. Not what I wanted to have, but > I'm not sure I have time for the proper way. > > 2. Use mplayer as fallback. > > 3. Drop Xine support. I'm not sure kaa.xine actually works with the > latest version of Xine and it has a lot of code we do not need anymore. > > After I'm done with that, the 2.0 mevas directory will be deleted and I > will install the kaa.candy-based version on my media PC. > > What Freevo 2.0 really needs is TV support. That is kind of broken right > now. But to fix it, you do not even need to understand Freevo. There is > a directory tvserver that hold everything you need. As kaa libs you need > kaa.base and kaa.egp. That's it! > > So let me turn post this into a call for help for 2.0. If someone wants > to code one of the following parts, let me know: > > 1. Port cool 1.x plugins to 2.0 > 2. Fix the tvserver code > 3. Create a webserver (not done yet) > 4. Help on working on the video playback, vpdau etc. > 5. Test various components by using them in extra helpers, e.g. use > kaa.beacon, our media db outside Freevo. > > Regards, > > Dischi > > I've posted a reply to Adam's initial email and I do believe that it is worth continuing work on the 1.x or new 1-NG branch, at least for one more release. It's been several years since freevo 2.0 first came about but despite a few developers using it there really is no one else. It's fair to say it's still a bit off and in the mean time Duncan's lead on freevo-1.x with the contributions from the community has helped keep the freevo project in general going.
There was a long period for a while there before his involvement where it wasn't looking good. The work on the 1.x branch has helped retain users and developers and I think it still has a part to play. In the end it's up to Adam whether to work on 1-NG or 2.0 but I might suggest that it would be good for Adam to spend at least a bit of time committing outstanding patches to the 1.x branch that Duncan hasn't had a chance to get to so that a new release can be done for users of 1.x which will help keep users and contributors on board. While that's going on, if you do have time to start hacking on freevo-2.0 then IMHO it would be worth updating the 2.0 wiki. The info above is great and I know doing documentation is the last thing you want to do but it would really help others seeing up to date information in the wiki if you do really want to encourage people to start shifting their focus onto the 2.0 branch. I can then put an announcement up on freevo.org (alongside a 1.9.5 release from Adam perhaps) pointing to the updated docs and asking for contributors to work on the areas you mention above. With Duncan no longer having time to spend on 1.x this seems like it might be a good progression path to bridge the changeover to 2.0 and get the larger community interested in it and freevo in general again. John ------------------------------------------------------------------------------ This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first _______________________________________________ Freevo-devel mailing list Freevo-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freevo-devel