Ralf Hemmecke wrote:
>
> > FWIW, my suspicion is that this will be a forever incompatibility with
> > OpenAxiom -- just a note for those library writers who are aiming all
> > AXIOM flavours.
> >
> > -- Gaby
>
> I must say that I would like that library writers follow the tradition
> and do *not* use underscores at all in their names, at least not in
> exported names. From that point of view I don't care much about a syntax
> change.
>
> Although I think that Gaby's remark does not really introduce an
> incompatibility at Spad level (one shouldn't use "ri" and "r_i" in the
> same function anyway), I tend to agree with him. It's bad enough that
> there are 3 axiom-flavour currently.
People who want their code to run in all AXIOM flavours must be
careful -- it is possible but requires work. Portablity problems
due to underscores are trivial to avoid. Other may be more tricky
but are solvable. OTOH if someone wants random piece of code
written for OpenAxiom to work with FriCAS and vice versa, then
this does not make sense. Makeing sure that independently
developed feature is compatible with other compiler requires
substantial work. And doing something in compiler is significantly
harder then making changes to user code. Given small amount of
Spad code that is written this means that total amount of
work, that is work on compiler + work on user code would grow.
In FriCAS I deliberatly limited changes to Spad language
(to maximize compatibility with existing code), but I want
to improve Spad language and for that changes are inevitable.
More generally, having two (or more) pieces of code which are doing
_exactly_ the same thing makes little sense -- effort is duplicated
for no gain. In other words either merge the projects or allow
them to go independetly each in their direction. Concerning
merge, IMO merging back with Axiom is out of question -- the
project go in too different directions (one trivial (but easy
to understand) example beeing that I spent several months
removing trash from FriCAS while Axiom at the same time added
more trash). Concerning OpenAxiom, I think that differences
are much smaller. But we would have to get consensus in
tens (if not hundreds) issus like the current change to
underscore handling. And statements like 'this will be a
forever incompatibility with OpenAxiom' does not make me
optimistic that attempt to merge would succeed.
--
Waldek Hebisch
[email protected]
--
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/fricas-devel?hl=en.
