Ralf Hemmecke wtor:
>
> On 03/01/2012 12:03 AM, Bill Page wrote:
> > Although "extend" is part of Aldor, my point of view is that this is
> > rarely needed and maybe only rarely a good idea in Axiom.
>
> Since "extend" does not yet exist in FriCAS SPAD, we are talking about
> the empty set, but anyway, I firmly believe that "extend" would be a
> good idea. Even if it is only used to straighten up the current
> bootstrap process.
>
> In FriCAS we have all the sources, so in theory, one could simply modify
> the original sources and recompile. The problem is that such a
> modification might have effects on distant other domains. In some sense,
> I think, "extend" makes the code safer, by not allowing to break
> existing code.
>
I do not think this is true -- the only difference between 'extend'
and extending by modifying code is that in case of 'extend'
compiler have no information about extensions while in the
second case compiler knows more. If in the second case
you get error at compile time, it means that extension was
wrong and would probably give troubles at runtime. If
modified code compiles, you have essentially the same
possibility for breakage as in case of 'extend'.
Of course, by modifying code you can do much more than via
'extend' -- above I only consider changes that could be
done via 'extend'. In other words, if you get any extra
safety, it is not due to 'extend' itself, but rather
because 'extend' limits you to subset of possible
changes (basically adding new signatures to domains and
joining domains to categories). If it is all want you
need than there is no safety gain. If you need more
then since you can not use 'extend' there is no
comparison.
--
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.
