[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

Testing the limits of nactation

Posted By: Nack Ballard
Date: Wednesday, 5 January 2011, at 10:53 p.m.

In Response To: Testing the limits of nactation (Timothy Chow)

If the primary goal in Nactation were to minimize the programming needed for computers to understand and convey single-character plays, then just 11 letters (BEACON, plus RDJI, plus maybe one letter like M to handle multi-quadrant doublets) with proper use of italics, underlining, emboldenment and colors, and a strict following of the rules already in place (or with fewer rules), would be sufficient.

Other letters and supplemental symbols have been added...

    (a) so that capitals can be used most of the time, while italics and so on are seldom needed,
    (b) to make the language more desriptive; provide room for style in describing plays, and
    (c) to help circumvent the problem that operating systems don't distinguish between regular capital letters and other letter forms for file-naming purposes.

It is ironic that the extra letters and symbols help so much with (c), yet make it more difficult to translate strings to and from some future computer program that doesn't impose the same severe file-naming restrictions.

Assumption is also added as an option to make Nactation more human-friendly. Granted, if assumption is used, it requires a person to renactate a couple/few plays of a game/match transcription before feeding them to Matt's (hypothetical) software. So be it. Nactation evolved with (a), (b) and (c) in mind, not to optimally accommodate computer programming. Creating a system that is both human-friendly and tailored for programming is not as easy as it might look with only one character per play. :)

That said, it is certainly possible to create a program that will do what Matt says (with the caveat in the previous paragraph), just as it is possible to create a chess program that understands the "baroque" rules of castling and en passant, or a backgammon program that understands that if only one number can be played it must be the higher number (only much easier than a games-playing program). I doubt that it would be nearly as problematic as the "bug-swarming" software projects you have witnessed over the years.

Noteworthy is Matt's observation, "...one glance at a column of opening replies expressed in [nacbrac] format highlights interesting patterns and discrepancies." Much of my own early game study has been done this way; it's remarkably efficient. A program could of course aid in generating such lists.

Nack

Messages In This Thread

 

Post Response

Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:

 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.