> Furthermore, there's quite a lot of merit to having such a CVS be > initially quite independent vis-a-vis patches that might get put in: > It allows everyone to get a good feel for the quality of the code > that is being proposed to come in from outside SAP. The SAP folk > could review the patches being put into SF, and see what, of those, > are suitable to push upstream. And by evaluating the patches, they > are also evaluating the patchers, getting a feeling for who's code > they would want to put in.
I think at least most of sapdb is complex enough that patches should be generated and reviewed by several people. Just allowing quick CVS commit is too dangerous with software of this complexity. After that applying the patches by some maintainer is not too hard and so far there does not seem to be any sign for a patch bottleneck. BTW lots of other free software projects developing complex software have similar policies (e.g. linux kernel, gcc, freebsd etc.) -Andi (who submitted so far one patch which was totally wrong -- review is needed)
