Re: [jokosher-devel] The road to 1.0



Stuart Langridge wrote:
OK, as you will have all surely seen, Jokosher 0.2 is out, and we
should all be very proud! It might take a few days to settle down (I
noticed Laszlo helping some people out with a problem or two on the
IRC logs last night), but after that we need to start thinking about
what exactly goes into the big 1.0 release. My plan for this is:


Okay there was talk of calling it 0.9 instead and having 1.0 come out after we get feedback from feisty beta users. Is this still the deal or do we want full 1.0 no bugs release in time for the feisty freeze?

 > Release 1.0 early-ish next year, say around March. This should
hopefully give it time to get into Ubuntu Feisty; we need to check
when the freeze is for that so that we can beat it. Feisty's the next
big distro release that we aim for, I think; there won't be a Fedora
release before that. A March release will mean a freeze in early-mid
February, which gives us less than three months to get all the
features actually written. Still, we have Laszlo The Code Machine to
kick us all in the arse and get it done :-)


Feisty releases april 20th, and the universe freeze is most likely going to be march 20th. I propose Wednesday march 14th for the release date, but only because its exactly 1 year from my first Jokosher patch. If you would rather have it on the weekend, that's cool too.

We should decide on exactly what features need to go into 1.0.
Obviously bugfixes as bugs arise, but what else? I think everyone
wants to have Network Instruments and the VoIP stuff, and we get to be
fully Gnome buzzword compliant by using Telepathy as well :) We should
have a meeting about this. What does everyone think to Sunday 26th
November (this Sunday) at 2pm GMT?


I was under the impression that telepathy stuff will have to wait until after 1.0 because otherwise we would be depending on CVS stuff forever.

I would like to implement a couple small things from http://jokosher.python-hosting.com/wiki/LaszlosIdeas, in particular tag support, and as-you-record-waveform drawing. There are two major features that I would like to see:

* Event looping --The ability to ship with a sample project that is less than 10mb in size requires the ability to loop events. But before we do that, we need support in gnonlin (which would have to be in the next release, so we don't depend on CVS). See bug:
http://bugzilla.gnome.org/show_bug.cgi?id=370836

* Pipeline permissions -- This isn't really a feature that the user will see, but it more under the hood. We should add the appropriate stuff to Project.py so that no other class can directly access the pipeline. This is because other classes will currently add and remove elements all willy nilly while its playing. Our current solution to this is to unsensitise all the appropriate GUI elements when we are playing (like we did with the add instrument dialog). This will not scale. I would like Project.AddToPipeline() to take in an element and if the pipeline is playing, it puts it in a queue, which will be added when it won't cause an error. The exact implementation of this will need input from all the devs.

And yes, 2pm gmt on sunday sounds fine.

Laszlo



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]