| |
BGonline.org Forums
Assumption in Nactation
Posted By: Nack Ballard In Response To: How to nactate 52S-54-g 13/9 8/3* (Matt Ryder)
Date: Tuesday, 4 January 2011, at 11:41 p.m.
Surely the all-encompassing guide to the rules of nactation you're compiling would supersede "assumptive" nactation and render it defunct?
Not at all!
It is necessary to distinguish computer use from human use, though. I wouldn't suggest trying to program the art of assumption.
As far as I'm concerned, "assumptive" nactation is to be avoided for many reasons:
* It's predicated on a set of fundamentally subjective assessments about the value of one play over another. The transcriber's opinion might easily differ from that of his audience.
True. A lot of factors (including judgment) go into determining what can safely be assumed and not safely assumed.
Consider 64S-44 played 24/16* 13/9(2). For a computer, I should use B, which is strictly correct. For a human audience, I would likely apply the hit assumption and use D (or 9) for the play. A computer would interpret D as the midpoint-breaking blot-leaving zilliwhopper 13/9(4). It is conceivable that a human (perhaps a Martian beginner?) might as well, but my greater concern is that someone misinterprets B as 24/20(2) 13/9(2), as diagrammed in section 3 of the tutorial, not yet knowing that the entire far side is fair territory and a hit overrides the extra point.
* It's predicated on the notion that the plays are all conceptually flawless. How might one assumptively nactate a blunder in which some commonly accepted backgammon principle is overlooked or ignored?
One must weigh the pros and cons. Suppose 21$-21S-11 arises in a game and I nactate the play as U. How do you interpret it? My intention is to assume 6/5 as an obvious ace, and nactate the rest of the play, which is 24/21.
Technical use of the U family would be U = 24/22(2), V = 24/20, and u = 24/21 24/23. So, a computer would interpret U as the multiwhopper 24/22(2). For a computer, I would nactate 24/21 6/5 as e.
If someone actually played 24/22(2) in that position, I would nactate it as @. Usually, such an alternative resource exists. But if not, then I would use U and mark it as a blunder by adding a dot after it (or circling the play or striking a line through it is clearer when writing). The player had a brain spasm and overlooked the blot on his 5pt, whatever. This is the price one very occasionally has to pay for the benefit of keeping the option of assumption open the rest of the time.
* It's challenging (nay, near impossible) to codify programmatically.
Right. It's not worth trying (except perhaps in some cases I will mention).
I realize that if someone hands you a Nactation string and a couple of the plays have employed assumption, you will have to renactate the input of those two plays.
* It might introduce ambiguities. If one symbol (say, f) is used in the assumptive case, and another in the non-assumptive case (say, capital E), how might the audience reliably distinguish instances of assumption? There doesn't seem to be an easy way to tell which Nactation situations imply assumptive techniques, and which don't.
For a computer, no. For people, it depends largely on whether the nactator has exercised sound judgment. Personally, I prefer to use assumption sparingly, when the difference in equity between the play made and its technical interpretation is huge, and the portion of the play being assumed is forced or it's a hit and/or "patently obvious."
Typically, I implement assumption only when the roll is doublets and the play is some 3:1 allocation of a BEACON letter (the underlined subfamily), so as not to burden non-maven readers. In the first example I cited above, 64S-44, I might use B (technical) or D (assumption), depending on circumstances. In the second example, I usually use U (assumption).
Even for some non-doublet situations, assumption is efficient (and could even be programmable). Consider 31P-62S-62 (or it works for 31P-62S-22) played 13/7*/5. By assuming the hit, you can use L (Lift) for the non-hit part of the play. You have to create a blot in order to lift it and there has to be a point on which it can land, which limits the number of L possibilities anyway. Sure, O is fine, but L seems friendlier — more easily recognizable.
For the most part, assumption should be used for computers only for partially forced plays, or for hitting plays that do not introduce a conflict. For humans, though, especially those who haven't yet learned how (or prefer not) to nactate complex doublets (which, granted, only arise on a small percentage of plays), assumption can be a valuable tool. OTOH, as a science it is imperfect and there is potential for abuse; I would therefore encourage people not to over-assume ("Only a donkey would make that .04 mistake, so I can just pretend that play doesn't exist"); it is a justifiable concern.
Nack
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.