On Fri, Jan 29, 2021 at 4:20 PM Oscar Benjamin <[email protected]> wrote: > > Thanks Aaron for sorting this. > > I've removed the ODE systems idea from the ideas page as it has been > mostly sorted out by Milan Jolly last year. There is still work to do > but it is not such a high priority after Milan's work. > > The change to 175hrs which is basically 5 actual weeks which places > quite significant limits on what could be achieved. We should think > about how to break up a lot of these ideas and make them more > manageable. For example the highest priority ideas are probably things > like the Risch algorithm, Cylindrical Algebraic Decomposition, Karr > etc. Each of those is way beyond 5 weeks though - we shouldn't bother > listing them there unless we can distil the idea down to more > carefully described steps that are actually achievable.
Yes, although I would argue that that has already been a problem. We haven't had much interest in these problems because they are too unapproachable. I don't know if they can be split into smaller problems, and if doing so would make them more approachable (there's also the issue that they are mathematically hard). Most likely for such things we would have to accept that a partial or incomplete implementation is acceptable. We already have had that to some degree with RUBI, which has had several GSoC projects but still isn't actually usable. > > I think that it would be good to move all of these ideas to a page > like "old ideas" and say that if anyone wants to rejuvenate the idea > then it should be updated and reframed in a way that is suitable to > the new GSOC arrangement before being moved to the new ideas page. > Bigger goals like the Risch algorithm should be moved to something > like a "big ideas" page. Any ideas listed for GSOC need to be specific > and achievable. Open ended ideas should go somewhere else I think. Feel free to refactor the page however you see fit. I deleted the old years pages because they were only confusing. > > As an example there is a project listed for factorising multivariate > polynomials: > https://github.com/sympy/sympy/wiki/GSoC-Ideas#multivariate-polynomials-and-factorization > That description must be fairly old and is too open-ended I think for > GSOC now. I recently proposed something related that is more direct > and achievable: > https://github.com/sympy/sympy/issues/20874 > Implementing that would solve a high-priority issue and I think that > for the right student concrete progress would be achievable. The factoring idea lists some algorithms, which presumably still aren't implemented but would help. I think it's useful to list explicit algorithms that can be implemented. Listing specific algorithms and specific papers helps to make the more mathematically sophisticated ideas more approachable and less open ended. Aaron Meurer > > To any potential students who are interested in these ideas please ask > about them on the mailing list before making a proposal. The list as > it stands does not adequately reflect the fact that some ideas are > much harder than others and also some are much more important than > others. Also many of the idea descriptions are out of date and do not > reflect the current state of sympy. > > Oscar > > On Fri, 29 Jan 2021 at 22:14, Aaron Meurer <[email protected]> wrote: > > > > The GSoC 20201 Organization applications are now open. > > > > **If you are a student interested in applying, please send a separate > > message to this list.** > > > > The deadline for the application is February 19. See > > https://summerofcode.withgoogle.com/how-it-works/#timeline. > > > > First off, anyone who is interested in mentoring, please add your name > > to the list of potential mentors at > > https://github.com/sympy/sympy/wiki/GSoC-Ideas#potential-mentors. I > > encourage students from previous years to consider mentoring this > > year. It is important to add your name there, as Google looks at how > > many mentors we have when considering our application. Even if you are > > unsure if you can mentor, or would only be interested in mentoring, > > please add your name there (whether or not you actually mentor will be > > up to you and will be based on if there is an actual application for a > > project you want to mentor). I cleared the names from last year, so if > > you mentored in the past you should re-add yourself to confirm that > > you can mentor again this year. > > > > Note that there is an important update to the GSoC rules this year. > > GSoC projects are expected to take up 175 hours over the three month > > period. Previously it was 350 hours. This should make things more > > accommodating for students who have exams in the middle of the summer. > > This means we should reduce the scope of any larger ideas, and also > > take this into consideration when reviewing proposals. > > > > I have updated the wiki to remove the years from the ideas page, > > application template, and student instructions pages URLs (if you want > > see the previous versions of these pages, look at the wiki git > > history). They are at > > > > https://github.com/sympy/sympy/wiki/GSoC-Ideas > > https://github.com/sympy/sympy/wiki/GSoC-Student-Instructions > > https://github.com/sympy/sympy/wiki/GSoC-Application-Template > > > > Please go over those pages and update them. The ideas page in > > particular needs to be updated, so that any ideas that are not > > relevant any more are updated or removed. If you know of an idea on > > the page that is no longer accurate, please update it. If you have an > > idea not on there, add it too (if you are unsure about the idea, you > > can discuss it here first). > > > > I am also in the process of updating the org application itself at > > https://github.com/sympy/sympy/wiki/GSoC-2021-Organization-Application. > > Comments and edits on this are welcome too. > > > > Aaron Meurer > > > > -- > > You received this message because you are subscribed to the Google Groups > > "sympy" group. > > To unsubscribe from this group and stop receiving emails from it, send an > > email to [email protected]. > > To view this discussion on the web visit > > https://groups.google.com/d/msgid/sympy/CAKgW%3D6LPGT6OqD0CqEhDfOmimWxkvCneqp3N6Dxz8z_CugAKig%40mail.gmail.com. > > -- > You received this message because you are subscribed to the Google Groups > "sympy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/sympy/CAHVvXxTN6Teriv3KnT2PyDM7ZOdxPtGGT3yFQPY-v7jMb8qKaQ%40mail.gmail.com. -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAKgW%3D6LkjbNJjmuSNsiBiDoZs%3DzR9P%3DAx160d2MC-%2BQXGNLChg%40mail.gmail.com.
