From: Steven Balensiefer (alaska@cs.washington.edu)
Date: Wed Apr 14 2004 - 13:26:19 PDT
The Essence of XML
by
J. Simeon, P.Wadler
I'd like to start by stating that I've never done any programming in a
functional language, and therefore I'm quite certain that some of the subtle
points of this paper were lost on me. I also must admit that I only have a
vague notion of the "niche" that XML is intended to fill.
Despite this inadequate frame of reference, I thought that paper did a good
job communicating the various details of the formal model the authors
constructed. The definition of the self-description and round-tripping
properties was new to me, and precisely capture the intuitive notion of a
data representation structure.
I thought the examples were very illustrative, with the exception of the one
in the section entitled "Derivation by restriction on simple types". I may be
wrong, but it definitely seemed like they mis-typed the element
configuration/laser/height as miles, when it should have been feet.
There were a couple things about the paper that confused me. First of all,
the definition of the substitutes judgment has no mention of an element
substituting for its baseElementName, which is used later as a hypothesis. It
seems like a simple omission, but I think it should be included for
completeness. I also would have liked more discussion about the consequences of
allowing ambiguity in the validation and erasure of complex types.
Overall, I'd consider it a good paper and note that the proof of their formal
model satisfying both self-description and round-tripping is an important
result. Furthermore, it was written in a way that even someone completely new
to functional languages could get a feel for what was being said.
This archive was generated by hypermail 2.1.6 : Wed Apr 14 2004 - 13:26:21 PDT