Camm Maguire: > [..] > > It is a pity that as sage offers a viable alternative, it has > nevertheless been passed over. I hear concerns about human resources, > but then when I offer to help make it work and request a status report > all I get is predetermined conclusions to not make it work. >
I am sorry we did not give you a proper status update - but to be honest that is because we did nothing significant to actually report to you. We had so many other things to do, there was no time to do anything on Maxima, of the depth that this task (using GCL) would have required. Please believe me when I say this - you can check our mailing list, the wiki page history, and our git logs, for our activity over the past several weeks and even months. I did in fact spend several (5-8) hours researching the topic and concluded it was not feasible for me to even *start trying* myself - but see below for further comments. > [..] But relying on internal structure which the > developers do not commit to maintain as a stable api is just asking for > this duplication of effort and *waste* of human resources. I know this > from experience [..] That is Sage upstream's decision to make, whether to waste their own time or not. It doesn't affect us Debian maintainers - the ECL patch is minimal. As I understand, they have judged it to be more beneficial to risk this, because linking directly with the internal library allows them to implement more features. I totally understand that this is generally something to be avoided. But in specific circumstances, one can make it work. From looking at the bug reports, Sage upstream seem to be quite "on the ball" on this one, and have helped catch several regressions in Maxima. > I assure you that I can get the print interface to work. In fact I > will pledge to be solely responsible for ensuring that tests of this > interface pass at the standard designed upstream should we agree to go > this way. It would make sense in such a case for me to join your group > and co-maintain sage with you. > We'd be more than happy if you wanted to do this and get Sage working with maxima-with-GCL, and keep this working! But the key question is, what do you plan to do regarding the FASL library and the ECL sage/interfaces/maxima_lib interface? In my limited understanding, one would have to modify quite a lot of internal Sage code, to make it instead use the Maxima print interface. This is definitely something to be discussed further with Sage upstream, and will take quite some time. Again, if you want to do this then we would be perfectly happy to accept any results that come out of it! But we didn't think it was within our capacity to do this, given all the other things that needed to be done. > [..] Thanks for this continued discussion! X -- GPG: ed25519/56034877E1F87C35 GPG: rsa4096/1318EFAC5FBBDBCE https://github.com/infinity0/pubkeys.git
