From: ····@panix.com
Subject: NYC Tuesday 27 July 1999: Lisp, XML, Scalable Scripting, and the World OS
Date:
Message-ID: <7nk2e7$2rf$1@news.panix.com>
Jay Sulzberger will speak at 6:30 pm at Prolifics, 116 John Street, on
the 20th floor, on the Island of Manhattan, thanks to the kind invitation
of Bruce Ingalls and the cppsig http://www.cppsig.org .
Map: http://www.cppsig/images/prolifics.gif
Today's http/html protocol stack, even with Java, Office, QT, Javascript,
PHP, and Perl running at both ends of the connection, is too often awkward
for the programmer to use, and annoying for the user to program. XML is
the chosen panacea of the influential W3 Consortium. XML today is almost
entirely simply a well defined syntax for S expressions, which are both
the input and output of the functional core of Lisp, an ancient
traditional language of logicians and programmers. But XML today lacks
the other half of Lisp: the interpreter that, given an S expression, hands
back the value of the S expression. This value is always itself an S
expression, which may again be subjected to evaluation. One of the
advantages of this style of computation is that, to use the offputting
jargon of the Functional Cabal, if all evaluations are really pure
evaluations without "side effects", then using arbitrarily many computers,
rather than just one, or a few, is made considerably easier. Most of the
world's computers today sit idle most of the time. The coming year will
see the construction of the universal XML interpreter, which with an
improved browser, will give us the kernel/interpreter and a shell for the
World OS.
Lisp in action:
http://www.whitehouse.gov
The great XML hoard:
http://www.oasis-open.org/cover
For XML, and indeed every document, down to the last bit, as an
un-evaluated S expression:
http://sanpietro.red-bean.com/~craig/grover
http://www.prescod.net/groves/shorttut
Suggestions for how to evaluate XML:
http://www.exmlscript.org
http://www.risource.org
http://dportal.co.uk/pxml
Databases that are already patently S expressions:
http://www.hsdi.com/qddb
http://www.multivaluedatabases.com
http://www.framerd.org
And in a related story, qscheme is an up and coming contender for
the title "Fastest Interpreted Scheme in the World":
http://www.sof.ch/dan/qscheme/index-e.html
I gave this harangue last year at a meeting of LXNY and I was lucky enough
to be heckled by two distinguished New York Kibologists, and defended by
an entire family of extremely charming hackers, some of whom flew in from
JPL. Perl Mongers and Kibologists are always particularly welcome to come
on down and help deal with the excess cheese. Blurb below.
Jay Sulzberger <····@panix.com>
Software: See below.
Hardware: Laptops running SCM and/or STk will be useful.
Who should attend: Beginners and hackers.
Scheme is a dialect of the Lisp family of languages. Scheme was
discovered in 1975 by Guy Lewis Steele Jr. and Gerald Jay Sussman.
R5RS is the latest official specification of the programming language
Scheme. The Introduction of R5RS begins:
Programming languages should be designed not by piling feature on top of
feature, but by removing the weaknesses and restrictions that make
additional features appear necessary. Scheme demonstrates that a very
small number of rules for forming expressions, with no restrictions on
how they are composed, suffice to form a practical and efficient
programming language that is flexible enough to support most of the major
programming paradigms in use today.
If at least one beginner shows up early at this meeting Jay Sulzberger
will present an introduction to functional style programming, using Scheme
as an exemplary functional language. We will explain functional style
programming by presenting the forgetful functor from a spreadsheet
interpreter to a Scheme interpreter. Seeing this functor makes it clear
that if you know how to write a spreadsheet, you already know how to write
Scheme. Time permitting, connections with hyperbolic partial differential
equations, homology of relational databases, geometry of the partition
function, and quantum weirdness will be made. If no beginners show up the
attending members will implement an Intercal interpreter.
Hecklers welcome.
We shall use the SCM and STk interpreters, which may be found at
http://www-swiss.ai.mit.edu/~jaffer
http://kaolin.unice.fr/STk
respectively.
SCM is by Aubrey Jaffer and STk is by Erick Gallesio.
We recommend also downloading the SLIB library and the JACAL system.
The most important thing to download is the zipped bundle of html-ed
documentation, which includes R5RS, from Aubrey Jaffer's site.
General information about Scheme may be found by starting at
http://www.schemers.org
and circling outwards.
John McCarthy's home page is at
http://www-formal.stanford.edu/jmc
under which may be found the first paper on Lisp.
Correspondence about this meeting should be sent to ····@panix.com .
Please include a complete Scheme, Intercal, or Basic expression in the
subject line.