Author: Steven J. Edwards
Date: 18:47:11 05/30/98
Fellow programmers:
I have been reading, and will continue to read, every posting here
concerning PGN 1998. To make sure that I see your article, please
include "PGN 1998" in the subject header.
1) I am providing the OCD toolkit to help propagate the standards, but
in no way is it required for using the standards. However, I hope to
make it so that others would want to use it and there are no impediments
to its usage. The OCD can also help implementors by providing a means
to help check their applications.
2) There is a question as to how flexible a PGN importer should be. My
opinion on this is that if a particular application will take nothing
but 100% pure, no-slop PGN, then it is a valid application.
3) The first problem with a permissive PGN importer is that it
unfortunately encourages the production of sloppy PGN exporters. And I
really want to avoid sloppy exporters; even if a sloppy exporter is
fixed, its earlier output may hang around on archives for a long time.
4) The second problem with a permissive PGN importer is that it
permissiveness is not easily formalized. This is bad because it can not
be standardized and different permissive importers will eventually cause
confusion to the end users.
5) NAGs are no more than a way to represent non-ASCII glyphs in the move
text. They cannot have a one to one mapping onto Informator symbols
because the former are not parameterized.
6) Foramlized parameterized symbols are handled by the Broket Forms.
These are chunks of EPD that appear between angle brackets. Basically,
any EPD operation that can appear in an EPD record may appear in a
broket form. For example:
<pv e4 e5 Nf3>
and:
<mate 5>
and:
<c0 "This is a swindle">
This is the place to handle parameterized items, so EPD will be expanded
as needed.
7) I would also take this opportunity to remind Fellow Programmers that
the first priority of Portable Game Notation IS data interchange and IS
NOT data presentation. While it is fine with me and many others to use
PGN as a sole means of communicating games to end users, it is certainly
not a requirement. Therefore, although cosmetic issues of data
presentation are important in some discussions, they are not critical to
the PGN revision process.
8) Oh, and of course the P in PGN means "Portable" and not "Post".
9) I am doing the PGN 1998 discussion here instead of rgcc. The noise
level there is too high for me.
-- Steven (sje@mv.mv.com)
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.