On Sat, Sep 19, 2009 at 4:47 PM, Bjarke Hammersholt Roune <bjarke.ro...@gmail.com> wrote: > > People have been writing mathematics software for a long time and so > people have been accumulating experiences about good ways and bad ways > to do things. I'm wondering how much Sage development is informed by > that experience and however much it is, I think it is worthwhile to > think about how to increase that. Is this something that has been > discussed here? > > Given some feature, how to find out what experiences other people have > had in implementing it before? One way is to have the people with > those experiences directly involved in developing Sage or at least > have them on sage-devel. A second is to identify these people and ask > them what they think. A third is to examine other systems and see what > kinds of advantages and disadvantages their designs entail (open > source systems allows to read the code, too). A fourth is to search > the literature for anything relevant. An with all of these except that > first is that they delay development by adding steps beyond coding the > first seemingly reasonable thing that comes to mind. That is both good > and bad. > > I thought to open this topic here because of a recent discussion on > sage-flame, and I'm hesitant to do so since it has the potential to > generate lots of talk to no benefit. On the other hand, it seems to me > that if a good way can be found to benefit even a little more from > past experience, that could have a significant impact on the quality > of Sage. My impression is that asking the people with the relevant > experience rarely happens, unless those people are already on sage- > devel and choose to speak up, and doing more of that seems a low- > effort way to benefit a little more from past experience.
I suspect that this discussion should be on sage-flame, but we'll see. Your disguising a dubious claim in the form of a question: Since Sage developers just do the first thing that pops into their head, how can we find a way to fix this problem? However, implicit in the question is the assertion that people who implement code for Sage often do the "first seemingly reasonable thing that comes to mind" instead of "asking the people with the relevant experience". This strikes me as a naive and generalization about a subtle and complicated development process. I've cc'd this to sage-flame. I recommend that all responses be there only. William --~--~---------~--~----~------------~-------~--~----~ To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
