Programmer <programmer@verisimilitudes.net> writes:
> timo.myyra@bittivirhe.fi (Timo Myyrä) writes:
>> Could you give example just how this would make development easier compared to
>> using Quicklisp?
> It would be convenient to be able to start Common Lisp development
> from a fresh install and without fooling around with Quicklisp. I'd
> be using a trustworthy, secure, and well-written package manager
> instead.
>
>> The ports development only has those libraries that are needed by some
>> applications. It shouldn't be used to duplicate the languages own package
>> manager.
> It seems a bit odd to port over end-software just because I want the
> libraries present, but alright; I can start work on that.
>
> Solene Rapenne <solene@perso.pw> writes:
>> CL developers use quicklisp, not the system CL libs. So, adding more ports
>> requiring work and openbsd developers time for 0 benefits, that's a no-go.
> I don't use Quicklisp. Just because it's popular doesn't make it a
> standard.
>
>> If a Common LISP program had to be ported and required porting common
>> lisp libraries as dependencies to work, then, we would consider having CL
>> libraries in the ports tree.
>>
>> Starting from the end, aka porting libraries, "just in case" we will need it in
>> the future, would be wasted time.
> As above, alright.
>
>> You can help by updating ecl, sbcl, clisp or stumpwm if you want to hack common
>> lisp ports.
> I can work on that as well. Is there a list of anything that needs to
> be done with regards to these anywhere? I would be interested in
> getting SBCL and friends to work under OpenBSD on the same platforms
> supported under GNU/Linux, so that's something I can start looking
> into, certainly.
If you want to have a project idea, then fixing the threads on SBCL would be
prime candidate.
Timo
No comments:
Post a Comment