- From: Sandro Hawke <sandro@w3.org>
- Date: Fri, 20 Feb 2009 10:52:37 -0500
- To: Chris Welty <cawelty@gmail.com>
- cc: "Public-Rif-Wg (E-mail)" <public-rif-wg@w3.org>
Channel: #rif
Teleconference code: 7431 (RIF1)
I have some notes on topics I'm thinking about.... This could work as
an agenda, I suppose. I've added numbers...
-- Sandro
================
syntax choices
[1] * BAREWORDS - if they might become a defined term later (eg
NAF) then it silently, sometimes undetectably, changes
the meaning of existing documents. So best to have
a declarations for local, maybe alias.
(also, okay to have variable names declared instead of
prefixed with ?)
[2] * infix operators -> parens
parens -> ambiguity in:
p(a b (c+d) e(c+d) f (c+d))
resolve with either
-> delimiters between all terms and formulas
(comma in arguments, semicolon/period in group?)
-> care about the space before an lparen
[3] * QNAME, CURIES, and BARE_IRIs
disallow ()[]{} which are allowed in URIs
but allow - + = , ...
it's a curie if the part before the colon is
defined via a prefix declaration; otherwise
it's a URI.
[4] * string literals (quoting), comments, numeric literals
[5] * MPS design: add anything that:
-- I know how to parse comfortably
-- has a well-defined abstract syntax (thus xml syntax)
-- someone in the WG really wants and plans to use
-- no one in the WG expects to want to mean something else
[6] * keep formula (boolean-expressions) separate in grammar
from forms/terms (object-expressions). Frees = for use
in named arguments....
[ maybe use use { } for reified formulas ]
[7] * frames as terms, nested frames, ...
[8] schema-checking the parse tree... dialect identification.
[9] "include"
[10] str-less-than ?
issue-67
Received on Friday, 20 February 2009 15:52:45 UTC