| |
BGonline.org Forums
Computer-friendly Nactation methodology for variant/family symbols
Posted By: Nack Ballard In Response To: Computer-friendly Nactation methodology for variant/family symbols (Matt Ryder)
Date: Sunday, 6 February 2011, at 11:08 a.m.
It is more elegant and more compact to describe a play with a single character, and I believe that should be the mainstream form of Nactation. I would like computer usage to conform to those values rather than the other way around. I'm willing to teach a second-character alternative for rare plays when it seems clear that the necessity exists for computer interface, but I think we should be resourceful and on the alert for other solutions as much as possible. I don't want to ignore potential conflict, but I also don't want to uglify a system in order to solve problems that will occur seldom or never.
I appreciate your inspirational vision of copying Nactation sequences from sources like the forum into a computer program. However, people nactate with different levels of skill, and for people who haven't learned all the conventions, assumption is a vital tool. Moreover, as much as I always appreciate the effort, a small percentage of letters that people choose are (inevitably) downright mistakes.
As it will be a long time if ever before "everyone" is fluent, I think it is better to accept as part of the process that more skilled nactators are going to (occasionally) need to alter some of the letters (perhaps adding a character in rare cases) in order to make the string computer-comprehensible. It is a trivial matter for me to do so, and in time it will be for you as well (though I was a bit surprised you didn't jump on the opportunity to practice by participating in the recent Nactation Study).
Contrary to the way it may seem, I wouldn't say that I "promote" assumption. I think of it as a set of training wheels that not only encourages people to go for a ride but indeed to make it possible for them to do so. When I judge that a readership is able to grasp a technically correct letter or symbol, I generally present it as the preferred character. (I have been holding back some in that regard only because the tutorial update is not yet published.)
Regarding the "21S-11 [E H31]," I believe Kye merely went to bgonline.org and misclicked on 21$-11, which does show [E H31] for that reply. There was no assumption being made, and it doesn't "signpost bigger problems."
I'm not sure what solution you are hoping will present itself for nacbrac'ing data that is charted in traditional data form (such as Tom's large compilation). It seems to me that either it all has to be hand-translated to nacbrac or you have to write a computer program to translate it. The latter choice seems more practical, though I can't help liking the idea of you practicing your Nactation the first way. :)
In my opinion, there should only ever be one 'best' symbol to describe a play...
I consider multiple Nactation options to be an advantage. You may be under the impression that assumption is being thrown around all over the place, but that is far, far removed from the truth. For example, consider 52D-32S-22N-51S, recently posted here. With her roll of 51, White just played Bar/24 13/8. That play can be nactated S, Z, D, 8 or t. None of those are assumptive choices; ALL FIVE are technically correct, bulletproof, unambiguous. Should I design an intricate ranking system to dictate to users which of those to use in case a computer is going to translate? Why? I see no reason not to write the program so that it can understand any of the five; it has to know the definitions of all the characters in any case.
For the four 21S-11 candidates, your Nactations are perfect. For reference, your E could be f, and your P could be I. Those are also perfect. That said, If e, f, E and P are fed to the computer, then even IF the program finds an underlined letter indistinguishable (do we know that for sure?), there is no need to use a second character. As a nactator skilled enough to know that, e, F, E and P are the choices I would make.
Finally, here's a detail of which you should be aware. E is not the fifth member of the E family; rather, it is the first member of the E family. E is closely related to E conceptually (which is why the definition is designed that way -- it's easy to learn), but from a programming perspective it should perhaps be interpreted as a different letter.
Cheers,
Nack
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.