From: Robert Uhl
Subject: Those Who Do Not Understand Lisp...
Date: Sun, 27 May 2007 06:00:37 +0000
Message-ID: <m3d50m6c1m.fsf@latakia.dyndns.org>

...are condemned to repeat it.

This fellow seems to have rederived one of the reasons for Common Lisp's
packages and symbols:

  http://blogs.sun.com/bblfish/entry/duck_typing_done_right

He notes that different pieces of software mean different things by the
same English name: walk() could be a function telling a robot to walk,
or a function walking a tree, or sending a dog outside, or whatever.
His proposal is to give names (symbols?) contexts (packages).  Only
instead of 'foo:bar' he writes '<http://a.com/Duck> a owl:Class;'.

In other words, it's reinventing packages, only without the notion of
*package*, and using, and shadowing, and and and.

-- 
Robert Uhl <http://public.xdi.org/=ruhl>
This thread has gone on long enough that it's about the subject again.
                                                    --Lieven Marchand

From: bblfish
Subject: Re: Those Who Do Not Understand Lisp...
Date: Sun, 27 May 2007 06:35:00 +0000
Message-ID: <1180247700.262228.227680@z28g2000prd.googlegroups.com>

Thanks for the compliment, but it is not really I who invented the
Semantic Web. Tim Berner's Lee did.
They may have been inspired by Lisp though, as most other languages
have been.

The good thing is that if it fits so nicely with lisp, then it should
be easy to build good RDF triple stores in lisp.
This is indeed what a well known lisp company named Franz is doing:

http://www.franz.com/products/allegrograph/

Henry

On May 26, 11:00 pm, Robert Uhl <·········@NOSPAMgmail.com> wrote:
> ...are condemned to repeat it.
>
> This fellow seems to have rederived one of the reasons for Common Lisp's
> packages and symbols:
>
>  http://blogs.sun.com/bblfish/entry/duck_typing_done_right
>
> He notes that different pieces of software mean different things by the
> same English name: walk() could be a function telling a robot to walk,
> or a function walking a tree, or sending a dog outside, or whatever.
> His proposal is to give names (symbols?) contexts (packages).  Only
> instead of 'foo:bar' he writes '<http://a.com/Duck> a owl:Class;'.
>
> In other words, it's reinventing packages, only without the notion of
> *package*, and using, and shadowing, and and and.
>
> --
> Robert Uhl <http://public.xdi.org/=ruhl>
> This thread has gone on long enough that it's about the subject again.
>                                                     --Lieven Marchand