why the huuuge repo?

I see that the clone for OF is really huge - it is 1.2Gb on my Linux box. Most space is occupied by library binaries. Why not to store them outside of repository and fetch only when needed?

because that’s the development repository, for all OSes, everything should be in there. The release versions are smaller.

But many projects survive without submitting binaries to repository itself. Why not to leave them outside and fetch only binary stuff required for specific platform?

That’s what the release versions are for: http://www.openframeworks.cc/download
64bit linux download is ~28MB.

Having to get all the binaries for all the libraries afterwards, from other sources, would introduce loads of complications - mandatory net access, version mismatches, …

yes, we have think of removing the binaries cause each version remains in the history making it really huge. The main reason to have them there is to make it easier but it’s beginning to be a problem. We need to find a way to remove them but not only the current ones but the whole history without loosing the history of everything else. Then probably have some script that downloads the binaries from some other place