Joshua Stone <[EMAIL PROTECTED]> writes:

> I would be very interested in this project, but looking over the list, I 
> am starting to realize that this is a bigger problem than taking a 
> library, fixing a bug or two, and submitting a patch.  Maybe you guys 
[...]
> With many of the errors being an artifact of Wiseman's implementation 
> choice, different implementations would generate different lists of 
> errors.  Packages that Wiseman successfully installed may not work on 
> other implementation-platform combinations. (see the 404 error trying to 
[...]
> This makes me think that we really can't do a "you-pick-10" model, 
> because this is a hyperdimensional problem.  Perhaps we could start with 
> a matrix of implementation-platforms and packages and begin with a more 
> accurate assessment of the situation.

In a different thread, Humberto Ortiz Zuazaga <[EMAIL PROTECTED]> writes:

> implementation version. Can we capture version information in our
> compatibility matrix?
[...]
> This means that any gardener's solution for tracking dependencies has to
> be better than apt, which is a tall order. Does anyone have ideas on a

Joshua and Humberto raise some good points.  But my plan for the
"Fixing ASDF-INSTALLable packages" project was simpler and much less
ambitious.  I was trying to avoid biting more than we
can--currently--eat.  I was thinking about an acceptable tradeoff
given limited resources, not a global maximum.

I envisioned a--how can I put it?--"sparse matrix", in which we fill
as many Lisp implementation/operating system/library version entries
as allowed by the available contributors.

This does not of course cover all cases, but it may not be necessary.
When someone looks for a library, I assume that, before issuing an
ASDF-INSTALL, he checks the system requirements and installs it only
when his environment is supported.

So, we could initially focus on as many of the environments on which a
library is advertised to run as possible given the available
resources.  Perhaps the upcoming Common Lisp packages directory
project could help users select only the libraries they can actually
use.

If all this looks like a "worse is better" approach, it's because it
probably is.

My hope was to help address with limited resources "most" of the Lisp
problems reported by novices.  Fixing "most" of the ASDF-INSTALL
problems, and the benefits from other gardening projects, could
collectively improve the overall situation of Lisp novices in a
reasonably short amount of time.

On the longer term, there would still be the possibility of further
iterations on the ASDF-INSTALL problems with a comprehensive
compatibility matrix, extensive test suites, etc.


Paolo
-- 
Lisp Propulsion Laboratory log - http://www.paoloamoroso.it/log
_______________________________________________
Gardeners mailing list
[email protected]
http://www.lispniks.com/mailman/listinfo/gardeners

Reply via email to