On 30 Aug 2016 at 12:10, Robert Ramey wrote: > I don't think any student should get to participate more than once. If > that's to harsh, one could modify it to be "more than once on the same > project" > > I understand that it might take more time than one summer to finish a > project. In this case the student should just accept this and accept the > responsibility on his own or pick a less ambitious project. The idea that a > project might take more than a summer and hence should be supported again > is abhorrent to me. It converts the process into more of a > grand/entitlement mentality which fosters, complacency and stasis rather > then disruption. > > I don't think a project has to end as an accepted boost library to be a > success. In fact, I doubt there are more than a very few students who are > actually capable of accomplishing this. I the project accomplishes it's > original (stated) aims - it's a success. > > students should be discouraged from taking on large ambitious projects like > new meta programming library. (LOL - I know) > > students should be encouraged to take on smaller, less sexy but more useful > projects - How about new documentation for the boost units library, Plan > for Boost usage of CMake. I would mentor these. > > students who actually complete their projects as anticipated should be > recognized explicitly - free ride to C++Now or CPPcon and opportunity to > get speaker slot. > > I don't really know much about the program, but it seems to me that the > Google people are in sync with my views on the subject.
I actually agree with you about what kind of work GSoC students ought to do for their own good and the good of Boost. Janitorial work, rather than cathedral building is much better for everybody concerned. But in the end this is a very competitive market, they can get paid 6k/month to do janitorial work at some multinational summer internship. Getting them to choose 5.5k for an entire summer tends to mean we get the cathedral builders, the ones happy to exchange income and big multinational work experience for much less recognised open source work and at one third to one quarter the pay. So as much as I agree with you, I've gone with the market forces here, and I've seen my role as trying to best facilitate the enthusiasm of the best of those we get rather than trying to fight market forces. I would add that cathedral builders tend to stick around in the community after GSoC ends to look after their babies. Janitors tend to move very quickly on. I don't agree with you about one summer only. It is *very* hard to find good students. Out the past four years, about 10% of new applicants I would call "reasonably good". Enforcing a limit of one summer only would reduce the calibre of the available student pool severely. It would also substantially raise risk, and I've noticed most mentors prefer a known less brilliant student over an unknown potentially more brilliant student. I can see why - they'd like fewer potential surprises. Niall --- Boost C++ Libraries Google Summer of Code 2016 admin https://svn.boost.org/trac/boost/wiki/SoC2016 -- You received this message because you are subscribed to the Google Groups "Boost Steering Committee" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
