I think that your attitude is perfectly reasonable-- and it's great that you are open to exploring these potential projects! That's all we can really do at this stage. I've had a mentorship experience where a significant amount of work had to occur up front to make sure that the mentee developed a reasonable scope and content for the work before anything else happened. I think that's normal and really a valuable part of the program.
The one difference I will pick with you is about code management. I think we have to be flexible and say that we just don't know where things might end up now-- they could go anywhere from nowhere (we hope not!) to actually merged. Let's be agnostic until we get a bite on our line! --- A. Soroka The University of Virginia Library > On Mar 10, 2017, at 8:34 AM, Andy Seaborne <a...@apache.org> wrote: > > Adam - we don't seem to have many JIRA marked GSoC2017. > > > I have signed up as a mentor because it has to happen now if at all. > I've marked JIRA I *might* mentor for: GeoSPARQL (JENA-664), > Fuseki+transactions (JENA-700). > > Both are about doing a solid prototype, rather than straight-to-codebase, > which is too much work (risk) to do on the fixed timescale. > > Neither project is easy. > Neither project is just development. > > They both have big learning curves and a student will need to have some > background to bring to the problem. > > My assumption is any prototype that gets developed on github and the Jena > accepts the code and places in a repo for stuff not in the codebase (maybe > even the SVN area which we still have). Sound reasonable? > > To be clear - I am not guaranteeing to mentor either project at his stage. I > want to know the project will have a decent chance of success before > accepting. > > Andy
