On Apr 11, 2008, at 15:30 , Chris Cannam wrote:
>> I don't think the choice between Cocoa and a Qt4 port is as clear >> cut as you >> think it is. > >> From my personal perspective it could not be any more clear cut. A > Qt4 port has some value; a Cocoa version is worthless to me. For you, yes. But note that in the business perspective you described, things are different. Anyway, I get your point. >> But keeping RG tied to Qt4 kills any chance it would have to get >> popular on OS X, because no OS X dev will be interested in >> contributing to an app which uses tools which are years behind what >> they're used to. > > That's not an argument that I can care about, because it's talking > about people I don't know, who may not exist (OS/X developers who > might make substantial contributions to this program), whose goals I > probably would not share. And in any case, you're not exactly > presenting much supporting evidence. And even if it were true, it > wouldn't be enough to justify this. Of course I have no evidence of possible OS X devs contributors, that's just me hoping. As I said, if I don't get any outside interest I'll just drop it. As for evidence of Cocoa being more advanced than Qt, I don't think it's worth discussing. >> Regarding possible future evolution, between a platform which has 5% >> of the PC market and is sustained by a large company, and a toolkit >> which has an unknown fraction of 0.8% of the PC market, and has just >> been bought by a company which primary interest is mobile devices and >> not the desktop, I'd rather bet on the first one. > > Even though the first one leaves you totally locked in to a > proprietary platform and the second one (which also has a fraction of > the OS/X and Windows markets that you didn't bother to mention, and > which is itself open source) does not? I'd rather be in a flourishing proprietary platform rather than an a dead-end free one. And let's be realistic, if the desktop version of Qt stops being actively maintained by Trolltech for some reason, Qt will stagnate. Given it's size it has to be maintained by full time devs. >> - approximately 0% of RG's functionality : getting RG to sound notes >> in a properly scheduled way took us quite a bit of time, about a year >> if I recall correctly. > > That's because we did a bad job of engineering it, not because of the > platform. If we did it again on Linux it would, or should, take much > less time as well. We'd still have to write a sequencer, and to depend on jackd and fluidsynth, wouldn't we ? > Anyway, I don't think I've anything further to say on this. I would > much prefer that this doesn't get mixed up with the Rosegarden > project; it would be potentially very confusing. Fine. I'll use another name as well. I think you should keep the rgplayer code around, though, it might come in handy as a reference should you eventually get RG to run on OS/X. > On a personal level, > I think it's a sad thing to lose you; I've enjoyed working with you, > it's been a long time, and good colleagues are hugely important on a > project like this. Thanks, the feeling is mutual. > I just wish you'd see sense and realise that the > answer to your problem has been right in front of your nose all along. I can't make any sense of sticking to a platform which pretty much doesn't care about us, sorry. Seems I'm in good company anyway : http://www.codinghorror.com/blog/archives/001089.html :-) -- Guillaume http://telegraph-road.org ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone _______________________________________________ Rosegarden-devel mailing list [email protected] - use the link below to unsubscribe https://lists.sourceforge.net/lists/listinfo/rosegarden-devel
