< Earlier Kibitzing · PAGE 240 OF 240 ·
|Sep-03-14|| ||OhioChessFan: I think a Scrabble score is a property of a word.|
|Sep-03-14|| ||Sneaky: I admit I asked in a sneaky way by inserting "the chemical" but I made up for it with the hint about capitalization.|
Yes, it's the word used in the highest theoretical scrabble play worth 1,778 points by hitting all three triple word scores and making numerous side words at the same time.
|Sep-04-14|| ||beatgiant: <Sneaky>
There must be lots of other long chemical names with an X and a Z in them. Is this one unique because the American Scrabble Association happens to accept it for play?
|Sep-04-14|| ||johnlspouge: < <Sneaky> wrote: Without using Google, try to guess what is an outstanding property of the chemical OXYPHENBUTAZONE? >|
Strictly, as <OhioChessFan> points out, it is "OXYPHENBUTAZONE" that has the property, not OXYPHENBUTAZONE. IMHO, the misdirection that enhanced the question does not detract from this excellent piece of Scrabble trivia.
Do you have any other information, such as <beatgiant> requested?
|Sep-04-14|| ||al wazir: I was only guessing.|
|Sep-04-14|| ||Sneaky: <Is this one unique because the American Scrabble Association happens to accept it for play?> I like Scrabble a lot, and I've known some people who actually play tournament Scrabble, but I am unclear on how they handle very long words. The official Scrabble Player dictionary (SOWPODS) does not contain words that are incredibly long, because they are so unlikely to appear during a game. However I believe there is a provision to allow very long words if they can be verified. |
I believe that OXYPHENBUTAZONE is in fact the winner, in spite of millions of other chemical names, because it needs to have several properties: it must have the X and Z and hopefully some other high scoring letters (P, B, Y). Plus it must be exactly 15 letters long. Plus there has to be some way subtract 7 tiles from the word to leave you with only legal words. So it's not just a matter of finding a chemical with Xs and Zs in it, it has to be very carefully planned out.
To see the Scrabble play in its fully glory, check out this graphic:
Notice how it turned "PACIFYING" into "OPACIFYING", turned "RAINWASHING" into "BRAINWASHING", turned "ONETIME" into "ZONETIME", etc. It truly is a brilliant construction. Whoever came up with it is the Sam Loyd of Scrabble.
|Sep-04-14|| ||al wazir: <Sneaky: Plus there has to be some way subtract 7 tiles from the word to leave you with only legal words.> Not exactly. Some of the 8 letters previously played can be *isolated letters* left on the board from words played in the *transverse* direction.|
|Sep-04-14|| ||beatgiant: <Sneaky>
<it needs to have several properties>
So did someone run a computer analysis over a dictionary of chemical names to establish that this is the highest scoring one? Those conditions don't seem so stringent to me.
|Sep-05-14|| ||zanzibar: I'm thinking of writing a Monte Carlo simulation for tournament play. Maybe with the idea of exploring the Confidence Level idea of generating a TPR.|
Regardless, the problem is that both ELO and Glickman give me good statistical formula for a score as a function of rating difference.
But a score can be made with both wins and draws.
So the question is how to generate a simulation, accurate wrt Glickman, but allowing draws.
The best data I've found on the issue is here:
An empirical study.
My idea for the MC is then as follows:
(Let N = W + L + D be total number of games.)
1) Roll the dice for a draw, using chess-db data.
2) Then roll another time for the W/(W+L) part.
Let s be the score via Glickman (or Elo). Let p be the draw percentage. A little algebra gives
W/(W+L) = (2s-p)/(2*(1-p))
Hopefully s lies between the (1-p/2) and p/2 boundaries (all wins, all draws).
Has anybody thought about this, or seen someone else do a simulation with draws included?
|Sep-10-14|| ||Sneaky: <Not exactly. Some of the 8 letters previously played can be *isolated letters* left on the board from words played in the *transverse* direction.> Yes, I know, that was hard to explain succinctly. My point was that it's not as easy as simply finding a long word with lots of high scoring letters.|
<So did someone run a computer analysis over a dictionary of chemical names to establish that this is the highest scoring one?> I'm actually not sure. I'm sure computers were used, and not just on chemical names but on all 15-letter words, but it's unclear to me if we know for certain this is the highest possible scoring play.
|Sep-12-14|| ||Sneaky: zanzy: What you write is interesting but it's a little over my head because my understanding of Glickman is so fuzzy. I thought it was just a modification of Elo that increases with the k-value in cases where a player has few games, or hasn't played in a while.|
Also forgive me but, what is the specific question you are really trying to answer?
|Sep-13-14|| ||Sneaky: Speaking of words, I was reading the newest inductees into the Oxford English Dictionary the other day. Here are some of the more interesting additions. |
The word "mansplain" is now in the OED defined as "(Of a man) explain (something) to someone, typically a woman, in a manner regarded as condescending or patronizing."
The word "douchebaggery" now officially is defined as "obnoxious or contemptible behavior." They regard it as "informal".
One that I find fascinating, the word "thing" has received another definition; it already had dozens, now it has one more. The new definition: "An established or genuine phenomenon or practice." The example they provide, "He looks like he's wearing boxers underneath his trunks; is that a thing?"
Oh, and thanks mostly to Netflix, OED has now made "binge-watch" officially a thing.
|Sep-14-14|| ||al wazir: All arbles that are neither beebles nor clupes are dishfinks.|
No dishfinks are elchers.
Beebles that are not dishfinks are not arbles.
No clupe is an arble unless it is also a dishfink.
Some elchers are beebles, some are clupes, and some are both.
Can any elchers be arbles?
|Sep-14-14|| ||beatgiant: <al wazir>
<Some elchers are beebles, some are clupes, and some are both.>
Is this list exhaustive? Or can an elcher be neither a beeble nor a clupe?
|Sep-14-14|| ||al wazir: <beatgiant: Is this list exhaustive?> I could tell you more, but I've already told you more than you need to answer the question.|
|Sep-14-14|| ||Karposian: <al wazir> <Can any elchers be arbles?>|
The answer is no.
Arbles = beebles/clupes/dishfinks.
So, if elchers can be arbles they must also be beebles, clupes or dishfinks.
<No dishfinks are elchers.>
<Beebles that are not dishfinks are not arbles.>
<No clupe is an arble unless it is also a dishfink.>
So, the answer must be no, they cannot.
|Sep-15-14|| ||Sneaky: I get so confused with the jibber-jabber I always have to convert things like this into more concrete terms.|
<1. All arbles that are neither beebles nor clupes are dishfinks.>
All chess-players that are neither poker-players nor smokers are drunkards.
<2. No dishfinks are elchers.>
No drunkards are handsome.
<3. Beebles that are not dishfinks are not arbles.>
Poker-players that are not drunkards are not chess-players.
<4. No clupe is an arble unless it is also a dishfink.>
No smoker is a chess-player unless he's also a drunkard.
<5. Some elchers are beebles, some are clupes, and some are both.>
Some handsome people are poker-players, some are smokers, and some are both.
<6. Can any elchers be arbles?>
Can any handsome people be chess-players?
What if this hypothetical handsome chess player neither smoked nor played poker? Well then according to (1) he'd have to be a drunkard, which makes him not handsome by rule (2). So that's not the case.
So perhaps he's handsome, plays chess, and smokes? We know by (4) that you smoke and play chess you must be a drunkard as well, and since drunkards can't be handsome by (2), that means a handsome chess player must not smoke.
Maybe the handsome chess-player plays poker? According to (3) "poker-players that are not drunkards are not chess-players", so our hypothetical handsome chess player can not play poker-either, because if he did that he'd have to be a drunkard, and (2) outlaws that.
I conclude "no." There is no such thing as a handsome chess player. QED.
|Sep-15-14|| ||al wazir: That was just a warmup. Now I am thinking of a different set of objects, with the following properties.|
All of them are floogits, glurks, hooshes, or igs.
Some floogits are glurks.
Some glurks are hooshes.
Some hooshes are igs.
Some igs are floogits.
All that are both floogits and hooshes are jeggers.
All jeggers are both floogits and hooshes .
No jeggers are both glurks and igs.
Is it possible for glurks to be igs?
|Sep-15-14|| ||beatgiant: <al wazir>
<Is it possible for glurks to be igs?>
|Sep-15-14|| ||al wazir: OK, here's the final exam in the course (from Robert Heinlein's _Space Cadet_).|
<After the test starts, he read, a score of '1' will result each time you press the lefthand button except as otherwise provided here below. Press the lefthand button whenever the red light appears provided the green light is not lighted as well except that no button should be pressed when the righthand gate is open unless all lights are out. If the righthand gate is open and the lefthand gate is closed, no score will result from pressing any button, but the lefthand button must nevertheless be pressed under these circumstances if all other conditions permit a button to be pressed before any score may be made in succeeding phases of the test. To put out the green light, press the righthand button. If the lefthand gate is not closed, no button may be pressed. If the lefthand gate is closed while the red light is lighted, do not press the lefthand button if the green light is out unless the righthand gate is open. To start the test move the starting lever from neutral all the way to the right.>
List all the ways (combinations of lights, gates, button sequences) by means of which a score can be made.
|Sep-15-14|| ||Karposian: <List all the ways (combinations of lights, gates, button sequences) by means of which a score can be made.>|
I can't see any possible combination that enables a score to be made!
So, my answer is: 0 ways to score.
|Sep-20-14|| ||Sneaky: OK here we go again:
<1. All of them are floogits, glurks, hooshes, or igs.>
All of them are doctors, lawyers, rich, or honest.
<2. Some floogits are glurks.>
Some doctors are lawyers.
<3. Some glurks are hooshes.>
Some lawyers are rich.
<4. Some hooshes are igs.>
Some rich people are honest.
<5. Some igs are floogits.>
Some honest people are doctors.
<6. All that are both floogits and hooshes are jeggers. >
All that are both doctors and rich are members of the country club.
<7. All jeggers are both floogits and hooshes.>
All members of the country club are both doctors and rich.
<8. No jeggers are both glurks and igs.>
No members of the country club are both lawyers and honest.
<9. Is it possible for glurks to be igs?>
Is it possible for lawyers to be honest?
If there is such a thing as an honest lawyer, he or she must not be a member of the country club, nor is he or she a doctor, all because of (6) and (7).
However I don't see anything else preventing an honest non-doctor lawyer, rich or otherwise, from existing.
So I'm going with "Yes", it's possible than an honest lawyer exists. I'm not betting the farm on it, but it's possible.
|Sep-20-14|| ||Sneaky: I ran into a very simple logic puzzle on the web a few months ago, and it aggravated me because I was tricked. Let me try to reconstruct it:|
<Some zeebles are binks. Some binks are choolas. Does it follow that some zeebles are choolas>
I used my typical method, trying to put real concepts down for the nonsense words. It completely led me astray. My crime was that I created subsets where the original statement implied no such thing. Specifically, I came up with this:
<Some animals are birds. Some birds are crows. Does it follow that some animals are crows?>
Stated like that, it seems like the answer ought to be "yes". But what if we restate it like this?
<Some tall people are German. Some Germans are dwarves. Does it follow that some tall people are dwarves?>
Stated that way, the answer is a glaring "no". And that is indeed the answer.
My problem is that when I read a statement like "Some zeebles are binks" I have to remember it doesn't outlaw the possibility of there being a bink which is not a zeeble. Binks might be a subset of zeebles, but I don't know. So when I wrote "Some animals are birds" I didn't stop to consider the notion of a bird not being an animal, even though I hadn't actually ruled it out.
So the lesson learned is that if you are going to go through the pains of substituting real words for the nonsense words, avoid words which imply subsets in the real-world. If you do, your real-world biases are bound to get in the way of your logic.
|Sep-21-14|| ||al wazir: <Sneaky>: I didn't mention it before, but I really admire your device of "concretizing" or "reifying" abstract and exotic terms. It's the sort of thing that I would do if I had thought of it. One of my problem-solving rules is to be quantitative where possible, or at least be specific; another is to try out ideas on *examples*. Your technique is in the same spirit.|
|Sep-21-14|| ||nok: Although his technique is indeed just confusing here.|
< Earlier Kibitzing · PAGE 240 OF 240 ·