| |
BGonline.org Forums
Question from Nactation study
Posted By: Nack Ballard In Response To: Question from nactation study (Matt Ryder)
Date: Sunday, 13 February 2011, at 4:19 p.m.
It seems like a bit of a slippery slope. Using open-ended logic like that, almost any letter could be subverted to fit. R and S would work, as well as a host of others, depending on how liberal you are about which "part of a play" the letter applies to.
Yes, S or R would fit the suggested allowance (as would U, V, Z and even Q). If someone is confused and thinks S means 24/21 8/6 in such a position where no "real" S play exists, why deny it? (If you can provide even one concrete example in which a conflict or problem results, I'll modify the wording or the entire approach.)
As I hinted at before, I would be in favor of also adding other limited allowances such as: "If no R (including our partial-play allowance), then try U, then give up." This one would work well for say an early game 51 played 24/18 (previously R was common but the 18pt is no longer in the R field and only U is "proper"). However, much is going to depend on you. I am hoping that early on you see both the benefits and that there is no downside in making an allowance (beyond the work programming it).
Ultimately, I would like to see the program smart enough to handle the most frequently and naturally applied assumptions, and at the same time I would encourage people (if they don't yet know a proper letter or otherwise feel compelled to make an assumption) to tailor their assumptions so that they fit strict usage plus allowance. For example, I would point out that while it is non-optimal but okay to use @ (anchor) for 54S-61P-64 played 24/20 13/7, it crosses a line to use @ for 54S-61P-43 played 24/20 13/10 because while a human would probably interpret that as intended, a computer would interpret it as 24/20 8/5 (assuming we do not further sophisticate the interpretation algorithm).
Nack
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.