>> Fuck it. I'll do it. It'll be a "reverse kickstarter".
your in the states right? is there now other way to fund such and endeavor? artist grant, academia, etc?
Well, I'm an independent artist. I imagine your organization has the capability to foster residencies, grants, etc for producing new work, wether it be new art or new/updates to arts toolkits. I've finished up my school work at CMU and just need to hand in my written thesis but I'm still affiliated. There is some interest for running GEM on RaspberryPI, but I haven't gotten much more than that for a specific Mac GEM update.
I'm not trying to say "give me money or I won't do this" but the truth of the matter is that I don't use GEM for my work personally. As with most open source projects, I have less "skin in the game" and could imagine pecking away at it here and there between other projects on my own. However, if I'm given the opportunity to come some place for a week or two, engage with some students/present my work, and focus on cranking our code it's totally doable in a shorter amount of time and I would have the incentive to do so. Many of us write and contribute to open source software as part of the process of creation. Others do it because they simply like to write code. I'm in the former group, not the latter.
>> pick out the code later today.
any luck?
Just to confirm what Iohannes, et al. already know. pix_filmDarwin, pix_movieDarwin, pix_recordQT, & pix_videoDarwin need to be updated to use the newer Apple APIs: QTKit or AVFoundation. I've done an AVFoundation app already and we're using QTkit in the updated OpenFrameworks code, so I can probably reuse bits here and there.
I'm not sure, but Base/GemWinCreateMac.cpp may need to be updated to, at the very least, use newer APIs to be more future proof. Also, the build system will have to compile objective C++ files since any updates will involve creating Obj-C classes that the original C++ interface will need to wrap. I have a few autotools projects, so I should be able to handle that.
That would be the basics and I think it would probably take me two weeks to do it right, maybe a week to get something working.
In the long run, I think it would be a good idea to replace the custom window creation/handling code with a library like GLFW or SDL2 to lift a good part of the cross platform burden off the project, but that's probably for another discussion. We're using GLFW in OpenFrameworks and SDL2 supports all desktops as well as iOS and OpenGLES (Raspberry PI) ... would be nice to have GEM working on RPI without X Windows.