[
Lists Home |
Date Index |
Thread Index
]
I confess that I'm baffled by your examples. I don't read or write a
lot of lisp code nowadays, but I the real problem is that you are
using a lot of functions that are not defined. In several cases you
give LISP code for what seems like it should be XSLT code, and the
markup examples don't seem related to the points I was making.
I'm not that interested in exploring at length how XSLT might have
been different. I just wanted to help to make your argument clearer
to and to point out some of the reasons that a syntactic focus is
useful for some applications.
The notational problems may be confusing me, but I don't see the
place in your response where you address the issue of putting out
prefixes that are _not_ associated with a namespace, either because
the object document is intended not to be namespace conformant, or
because the declarations are expected to be made in another document,
which will create a parsing context.
For instance, in protocol design (according to Microsoft) there is
sometimes the need to create session-unique namespaces -- They have
used relative URIs for this, but one could also late-bind a unique
URI at transmission time.
I'm also not arguing that this is impossible under your approach, but
that it involves extra complexity, where as a "syntactic approach"
takes all of this in stride without special features, but at the cost
of permitting namespace-related errors in output when one is not
striving for "special effects".
If you want to make proposals that will be widely considered, you
probably either need to change your notation to one that is more
commonly understood, or define it clearly and hope that the people
you want to communicate with are willing to learn it.
I think I understand where you are coming from, and I've spoken my
piece, so I am now going to return to my dogmatic slumber. I don't
think your ideas are bad, just a different tradeoff. Good luck. This
has certainly helped me think through some issues about type-safety
and namespaces.
-- David
--
-------------------------------
David Durand | 12 Bassett St.
david.durand@ingenta.com | Providence RI, 02903-4628 USA
VP, Software Architecture | 401-331-2014 x111 Cell: 401-935-5317
ingenta plc | FAX: 401-331-2015
http://www.ingenta.com/
|
