On 10 March 2014 14:29, Krystian Bacławski <[email protected]> wrote: > @Bill Page: I'll add comments on top what Waldek said. I think he gave a > pretty comprehensive answer - he filled up the gaps I couldn't have > addressed. >
Thanks. I think that together you have answered my questions very well. > On Sun, Mar 9, 2014 at 1:56 AM, Waldek Hebisch <[email protected]> > wrote: >> >> Of course OpenAxiom also added features. > > > Where can I read more about it? > There is of course the OpenAxiom source code ... Really I think it is wrote reading especially because Gaby has spent a lot of effort to make it a lot more readable than the original Axiom source. The debate about readable source code versus embedding obtuse source code into a "literate" document is one of the long standing issues which lead to the creation of the FriCAS and OpenAxiom forks of the original Axiom project. >> >> One thing from OpenAxiom >> that could help are interfaces that expose compiler data >> structures. I mentioned them to Krystian in the past, >> but IIRC he found them not so helpful. I hope he can >> explain this better. > > I think it somehow escaped my notice. We should definitely discuss it again. > In long term I would still prefer to continue work of FriCAS version of SPAD > compiler - as Waldek has the knowledge and time to guide me through the > sources. > Understood. >> And Aldor compiler is written in C, while new work is done in Spad, >> so hopefully new typechecker will be shorter and simpler. > > > The hope is that people who are active SPAD / Aldor developers will be able > to contribute changes into the type system. IMO C language is too low-level > language to develop compilers effectively, even gcc is moving towards > higher-level C++. > I am sympathetic to the use of higher level languages. Apparently that was also the goal of the original Axiom developers. >> And ultimately goal is to go beyond Aldor. > Yes. I think that is very challenging goal. It seems to me that a similar goal has been pursued by developers of language like Haskell, unfortunately with only rather mixed success. I think where Haskell advanced the most was in the application of ideas from category theory. It me that is where a new language to replace Aldor and SPAD should focus. Category theory as the foundation for the entire Axiom/FriCAS library still makes the most sense to me. > > I'm not in the position to drive the changes in type system of SPAD / Aldor. > I consider myself to be only a beginner. However - having experience with > couple other languages like C++, Python, Ocaml, Haskell I see what could be > added to the type system to make it look modern and appealing to new users. > I am not sure whether it mates sense to strive for just a "modern" look and appeal. There are already a lot of players in that sort of game. Where Axiom is different is in the depth with which it applies abstract mathematics. It is hard to imagine that this approach can have any sort of mass appeal - except of course among those most mathematically inclined. :) Cheers, Bill Page. -- You received this message because you are subscribed to the Google Groups "FriCAS - computer algebra system" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/fricas-devel. For more options, visit https://groups.google.com/d/optout.
