comments Neil Van Dyke 26 Jan 2006 16:23 UTC

A few quick comments...

> Libraries Organized by Category
>
> Many popular open source library collections are organized by
> category, sometimes with subcategories within the category, and then
> libraries within the (sub)category.

I think taxonomies for classifying libraries for human location are (or
should be) more ephemeral than the universal identifiers of the
libraries themselves.

I wrote some remarks on this a few years ago:
http://lists.gnu.org/archive/html/guile-devel/2001-04/msg00285.html

> Many projects, groups, and businesses have registered their own domain
> name. Your domain name is a natural choice to use as the base for your
> universal identifier... when, of course, the the domain is relevant to
> the resource you are naming.

Domain names are great for dividing up the universal namespace nowadays.
Languages like Java and XML use domain names for this purpose, and it's
not a bad choice for Scheme, either.

> HTTP URL's
>
> Many people and projects, even if they don't have their own domain
> name, do have their own home page on the web.
>
>     http://swissnet.ai.mit.edu/~jaffer/SLIB.html
>     http://www.gnu.org/software/kawa/
>     ...
>
> This SRFI supports using HTTP URL's as the base of a universal
> identifier when that is convenient.

If "http"-scheme URLs were used for identifying libraries, I'd think
that each URL would actually be a reference to a Web object that could
by retrieved via HTTP to define the referenced library.  If we only
wanted a name for a library, I'd suggest not using URLs.  Is that what
you had in mind?

--
                                             http://www.neilvandyke.org/