On Sat, 5 Nov 2011 12:25:27 -0200 Gustavo Sverzut Barbieri <[email protected]> said:
> On Saturday, November 5, 2011, Mike Blumenkrantz <[email protected]> wrote: > > On Sat, 5 Nov 2011 23:12:21 +0900 > > Carsten Haitzler (The Rasterman) <[email protected]> wrote: > > > >> On Sat, 5 Nov 2011 09:47:08 -0400 Mike Blumenkrantz <[email protected]> > said: > >> > >> > On Sat, 5 Nov 2011 22:31:01 +0900 > >> > Carsten Haitzler (The Rasterman) <[email protected]> wrote: > >> > > >> > > On Sat, 5 Nov 2011 10:34:31 -0200 Gustavo Sverzut Barbieri > >> > > <[email protected]> said: > >> > > > >> > > > I wonder if this should go in few days before the freeze. Really, > let's > >> > > > put such a big thing in a branch and let it mature for a while, > merging > >> > > > into trunk after 1.1 is released. > >> > > > >> > > this thing has been going on for months. you just didnt see the > previous > >> > > patches i got. > >> > erm without favoring one way or another, if WE haven't seen anything > until > >> > just now then why would it matter how long "this thing" has been going > on > >> > for? > >> > >> you were favored with some ecore_con rewriting just before 1.0... :) > > that was in order to fix a huge number of bugs, not add a giant api. > >> > >> you're going to have to trust me then. the code is pretty much dormant > unless > >> the feature is used. as such i've given it much more testing than the > people > >> who wrote it and spent basically 2 solid says worth of work making it go > from > >> barely limping to actually working reliably. we have 2 more weeks of bug > >> fixing to iron out the bugs. > >> > >> right now as best i know (please. i'd love to have other people who know > this > >> around), i'm the one with probably the most experience with audio. i > spent > >> several years as a teenager composing audio tunes via mod trackers. i > have a > >> decent understanding of what sequencing, tracking and sample > manipulation are > >> about - i am far from an expert at things like mp3/vorbis audio codec, > but i > >> do know the principles. if there is a problem in the api design (which > is all > >> in edc - and its right now a tiny partial implementation of my > requirements, > >> there is more to come) then i'd love to have people other than me know > enough > >> about audio, mixing, sequencing etc. to actually say something. speak up > if u > >> think u have audio under your belt. but as best i know, no-one has > indicated > >> they know this to date. :( so i guess as the for-no-defacto-audio-guy... > i > >> say this is good as it has, api-wise, followed the spec i gave prince & > >> govind - in its subset implementation. if there are bugs - they can be > fixed. > >> still have 2 more weeks of bugfixing. that's WHY there is a separate > merge > >> window vs freeze window. :) > > seems reasonable enough I suppose. would still be nice if giant patches > like > > this could get brought in more than a couple days before the merge > > window...maybe some sort of samsung policy > > Also, split patches for easier review. Okay, none of us may be as expert in > audio as raster, but very small part is about it. In a way, we can act as > users of the API in the review, also reviewing all the other bits. But as > is the patch is a major dump that would be rejected by us and every other > projects. > > I hope raster understands our points. oh don't get nme wrong - i'm not an expert... just based on what i know of the other e devs etc... i happen to know/have done more. in the field. :) seriously... did you LOOK? there is barely any api... look at multisense.edc... it's also all documented... i think you need to spend the time to look at how little visible api it adds. it is mostly adding internal stuff that makes it possible to mix and play audio at ALL - thats most of it, as well as encode audio samples into the edj file. thats really the vast majority of it right now. it's a tiny bare-bones minimal subset of a much more extensive edc api i specced that included envelopes, instrument definitions, looping, sequences (tracks), patterns, specific channel control as well as auto-dynamic-channel creation etc. etc. i can't send you the example right now - but i have a much fuller spec. what u see in this patch is tiny and adds almost nothing. take a closer look. -- ------------- Codito, ergo sum - "I code, therefore I am" -------------- The Rasterman (Carsten Haitzler) [email protected] ------------------------------------------------------------------------------ RSA(R) Conference 2012 Save $700 by Nov 18 Register now http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ enlightenment-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
