(French here) I'm not surprised at all. I've seen a documentary once about Scrabble competitions and the guys were discussing after the match about the missed possibilities. One guy was complaining that he could almost get one long obscure word I've never heard about which would be worth a lot of points and the others were discussing about it. The reporter then asked them what the word meant and no one had any idea. Not a single French player could explain what the word meant, it was just in the official dictionary, that's all.
When you organize a Scrabble competition, you need to have an official dictionary or people start inventing non-sense words to get points. In French for example, you could add the prefix 're' (meaning again or back) to almost any verb and everyone would understand, this is just an example but it's really easy to create nonsense words that people understand from their root. And since you have an official dictionary, it's not about words anymore, it's about memorizing the dictionary.
As a slightly competitive Scrabble player, words like qi have ruined the game. You should need to know the meaning of the word to play it. This applies for short words and long words alike.
> You should need to know the meaning of the word to play it
How would you go about determining in a regulated fashion whether someone knows the meaning of a word?
Would you have them offer a definition to be checked against an official dictionary?
Would the definition given have to match the official dictionary definition word for word?
Is so, wouldn't this simply force high-level scrabble players to memorize not only the official words but the official meanings themselves?
And if perfect matching wouldn't be required, how much variation would be allowed and how would such variation be measured?
There's no sense in falling back on the "just use the word in a sentence" proof, either. Because that's no proof at all, since, even if we dispense with all the problems consensus introduces, if the usage is contested, then some official dictionary definition will have to be referenced in the end.
And then you're going to have to deal with the above mentioned issues anyways.
The desire for a more natural, simple and organic Scrabble world in which any word played must be theoretically backended by a player's knowledge of its definition is, well, just asking for trouble.
This suggestion isn't meant to apply to serious or competitive play. It's an attempt to keep things more fun among friendly players of differing skill levels. Anything that 'forces' serious players to memorize more meaningless information is a benefit, as far as I'm concerned.
People don't seem to like it when something they cherish is reduced to symbols. I got a good laugh out of this comment at the Guardian:
"I think it renders the whole thing pointless. It ceases to be a word game. I think that, as in tennis and cricket, there should be the right of challenge, and if the person doesn't know the meaning of the word, the go doesn't count and the word is removed." ~ MichaelBulley
I am sure Mr Bulley would prefer it if Chess players could challenge their opponent to prove he or she is a either a Monarch, a Lord, Knight or at the very least a commoner who lives in a castle.
I don't understand your analogy. The Scrabble problem is people not knowing the words they use. The only Chess equivalent could be people not knowing or understanding the moves they make. It has nothing to do with someone moving a queen and not being a queen themselves.
Scrabble is not about words,anymore than monopoly is about business.
Scrabble is arranging a random selection of tiles into combinations in such a way that score is maximized.
The tiles could have hexadecimal numbers on them and could be played against a randomly generated dictionary of scored combinations. Much like you can play chess with dragons figurines instead of castles or storm troopers instead pawns.
It was as much to do with words as chess has to do with war in the middle ages.
> The tiles could have hexadecimal numbers on them and could be played against a randomly generated dictionary of scored combinations.
This is an intriguing assertion. The words of a natural language are far from randomly generated. While I can invent new words like 'wobblery' or 'fandogonkified' they still follow patterns familiar from other words.
If I understand the Chomsky idea correctly, we are born with structures in our brain ready to accept a new language. A human might have a hard time memorizing a dictionary of randomly generated words in hexidecimal -- he or she would have to resort to mnemonic techniques (peg lists, memory palace, etc) to memorize that dictionary but it could be done. Further, it could be done by any human who was dedicated to the task.
Whereas the system you describe involves brute force, much like the Deep Blue approach to winning chess. And it's been shown that a group of people collaborating together can defeat computers at chess! I think a lot of interesting research could be done here (if it hasn't already) and it would not surprise me if your assertion could be proven wrong (but I generally agree with it).
The system GP describes doesn't involve brute force, it's just an accurate description of Scrabble. It's about arranging tiles with point values in a way score is maximized. That the scoring positions are not uniformly random but follow some implicit language rules is something that helps people memorize / derive the proper arrangements, but otherwise does not impact the game in any way.
What is the difference between memorizing a sequence of steps and having a couple of sequences of steps culturally contextualized in a symbolic language?
I mean it's literally one language against another one. You can call it 'meaning' all you like, but 'meaning' is representative. You call it 'understand' but to this guy memorizing the entire dictionary in order might mean 'understand'.
You might say you need to have felt 'being angry' in order to know what 'angry means' and to determine this as understanding the meaning of the word is ridiculous because it goes both ways. People make their owns meanings up, the only reason we agree is because we exist in relatively the same space time (with respect to the size and age of the universe) and information only varies so much at the distributed rate it tends to vary across minds, people most people are similar in how their minds function and construct pattern relations.
I could have sworn that it was a requirement to be able to define any word placed, but maybe that's just a house rule I internalised while growing up. Or a rule since dropped from the 1960s edition we played with.
The funny thing is that it probably has the potential to become so well known that people forget its roots, and then people wouldn't get the original joke in that Simpsons episode anymore.
Surely it would be incredibly subjective to require a definition?
Whereas in Scrabble as it stands, a move is either legal or illegal. It's very black and white. Well, once you decide on the dictionary you're playing to!
I stopped playing Scrabble against my father because he was frustrated that I had memorised every two and three letter word. Even being able to define many didn't matter; if they came across as though they were memorised rather than learnt, he was annoyed. I don't see the issue. Everyone who plays is playing by the same rules.
A common thing is to give all players a list of all of the weird little 2 and 3 letter words. The problem with them is not that they're illegal, just that knowing them is very gamey unfun busywork, but gives a big advantage to everybody who spent time memorizing them.
Wouldn't that significantly cut down on the valid moves possible from a given board configuration? You'd only be able to play perpendicular to preexisting words.
I imagine that the parent considered that one could play a longer word that happened to make shorter words just not _only_ play a two letter word; so parallel words would be acceptable as would things like in-fills that only took one letter but caused 2 words to be formed.
Perhaps a better, succinct, rule that tackles the same problem would be "a minimum of two tiles must be placed".
You're right, I didn't think of that - but if you're playing with someone who doesn't know too many two-letter-words, that might come up less frequently anyway.
Having played a lot of Scrabble with my wife I can feel for your father a bit. I think its probably less about having memorised them, and more about two players of dramatically different skill levels playing the same game.
Eventually I came to terms with it and just considered anything within a two hundred points or so of my wife's score as a personal win, but it can still be frustrating to feel you're playing your best and losing to someone who's barely trying.
The moral of this story I think is not to play an English graduate at Scrabble unless you're OK with losing.
And it makes the training less pointless. Words without meaning are useful for nothing besides winning Scrabble, knowing strange words WITH meaning however has a chance to be useful outside the game.
People's confusion about whether being too methodical about winning the game is against the game spirit are missing that games, especially spectator sports, have conflicting goals.
As a player, your goal is to win the game, period. But for everyone else, including spectators and the whole support industry that is built around the game, it doesn't matter who wins. It matters only that the crowd is entertained. A player being too good makes the game boring, and so it is discouraged.
You can see this dynamic more or less clearly in pretty much every sport (F-1 being a notable example). The rules of the game are so constructed as to "level the playing field", to mostly eliminate any kind of advantage a player may bring to the field - like better tech or supplements. The crowd needs to see fierce competition, where everything depends on the talent of players. But then again it can't be too obviously random. If people feel they can't predict the result, they won't enjoy it. Doesn't matter if the game is inherently random or not; it only needs to not look like it. Then there's danger. Safe sports are boring, but actual accidents will detract people from enjoying the show, so again you need to create a sense of imminent danger while minimizing any real chance of a problem.
I mentioned F-1, because it's a notable example of a sport where all rules are constantly adjusted to ensure that you can't win by having better tech.
Personally, I hate this whole showbusiness and the engineering of enjoyment. I would appreciate a game with rules set up so people would be free to develop better strategies and better tech in order to dominate.
Actually, the question would be more "what is judged by a competition ?". Levelling the playing field is not only a reason for pleasing the crowds (which happens sometimes) but also a way to assert what the competition is for. A car racing organization can decide that to judge the pilot skills is more important than judging the technology level, and therefore setup rules for that.
On the other hand, we could see in the foreseeable future another competition judging only the technology and letting a standard I.A. do the driving (Google car racing team ?). But spectators would maybe not enjoy much such competitions because they have to relate to competitors.
Same thing : at the Olympics, it is generally considered that the important part is the human effort, not the technological one, so they set up standards to judge only this human part. This is not a marketing strategy, but a philosophical choice. And this is interesting because this can be debated as the lines between the technology and the human will become blurred (remember the polemic when it was discussed if the blades of Oscar Pistorius were or not giving him an advantage ?).
But my view on that is we need to keep a strong 'human only' part of sports, even those where tech can play a role. I don't mind that there is a whole range of competitions with the whole spectrum of human / tech association in any proportion, but I would not be very interested long if the human part was very minor.
For the same reason Tour de France riders ride standardized bikes. You can actually buy faster bikes than the ones used in the Tour de France (it's difficult to also get the necessary legs though).
Similar to sailing in the olympics: Everyone has the same model boat.
Not like the America's cup boat race where you need a huge R&D budget to design a boat. Then someone shows up with a catamaran with a wing sail and they have to regiger the rules again.
The Americas cup had rules about boat configuration. So technology started driving the race. There was in the 1980s the australians showed up with a secret weapon the "winged keel" on the boat "Australia II". It helped a lot and raised the game for yachting.
1988 rolls around, and a New Zeland team is trying for the cup. Giant boat. US shows up in a catamaran. Catamarans are just faster..
The boats don't even seem like they should race.
http://www.yachtphoto.com/data/photos/176_1r1988_02.jpg
Lawsuits about the boat rules after the race was over. The rules didn't say anything about having 2 hulls.
God it pisses me off that Oracle has enough money to blow sponsoring shit like this while simultaneously working to destroy the software development industry. "We've got enough money to buy useful technologies and run them into the ground, and to run lawsuits that are incredibly harmful to the software industry for decades, and enough left over on the side to race expensive boats for fun." Fuck you, Oracle.
> I would appreciate a game with rules set up so people would be free to develop better strategies and better tech in order to dominate.
There are games like that. But they don't tend to have as many spectators or business support - and they don't tend to be as much fun - because they're not optimized for those things.
>As a player, your goal is to win the game, period. //
When I take part in sports or play games my object is to enjoy the game and if it's competitive to make a good account of myself. I used to play football (we were runners up in our University cup competition, so not terribly shoddy) and later played competitive volleyball - one can do well and not win, one can win and play poorly.
Just as not all companies are solely built on a motive of profit above all else so too not all competitive players play solely to beat their competition, some wish to enjoy the game and better themselves more than they care to be labelled "winner".
> As a player, your goal is to win the game, period. But for everyone else, including spectators and the whole support industry that is built around the game, it doesn't matter who wins. It matters only that the crowd is entertained. A player being too good makes the game boring, and so it is discouraged.
I think this misses the point that in many games, for many players, they simultaneously have the role you describe as "player" and "spectator"; while they want to win the game, they want the game to be entertaining and not boring, as well.
Actually curious: is the goal of the F-1 rule changes to level the playing field? I'd heard it was to keep automakers on their toes, guiding the competition towards things that the general public wants, too, like gas efficiency. If I remember correctly, one of the more recent years saw a huge tank capacity reduction, ostensibly to force automakers to create powerful engines while maintaining some semblance of efficiency.
There are many, many objectives for which the F1 rules need to account. Competitive balance is one of them. Providing interesting and relevant technical challenges for automakers and teams is another (and was a major driver behind the recent switch to hybrid engines). In addition to those, you have safety, spectacle, cost-containment, logistical considerations, and about a thousand other goals that need to be incorporated. The FIA takes incredible flak for whatever they do because there are myriad tradeoffs between these objectives, and it's impossible to balance them all in a way that makes everyone happy.
Modern day sport is more business than anything. Competition/show business is a weird thing - you need to have rules and you hope you don't get someone so good that the match becomes boring (there are exceptions though - Rhonda's fights last a few seconds to few mins at max). Money makes things weird. That is probably why some incredible athletes don't compete - Laird Hamilton for example.
You got me here. At the time of writing of my original comment it didn't occur to me that capitalism is an example of a game thst doesn't set the rules to level the playing field. I guess I was thinking about games with well-defined victory conditions and relatively short play time.
A chess buddy of mine did exactly the same: he was in a mental institution for a while and completely bored, and also a scrabble maniac. There was nobody to play with except for an old lady who only spoke French. He memorized the French dictionary (in < 2 days, I can add); only the words, as the semantics had no value. He was less bored after that.
A friend of mine did something similar with the game Scramble With Friends on her phone. She got bored kicking everyone's asses on the English version, so she started playing Danish, which she'd never seen. She just tried various combos, remembering what worked. She was globally ranked in the top 1% in like a week.
Except a lot of natural language grammar is contextual, so there is no way to have a conversation that makes sense without an understanding (internal model) of the meanings of words and really together with the cultural context: literature, pop culture, current events etc. You may be able to produce valid sentences that still sound bogus though.
Searles Chinese Room argument although fundamentally flawed (the entire room or house is the conscious part) it says that although you can have a person spitting out the right translations the person is does not speak Chinese.
Um, I think there's a challenge with this analysis of the Chinese Room argument. The argument presupposes a perfect lookup table of (sentance -> sentance).
So the point of the argument is that given the correct responses, there's no way to tell if there is a contextual understanding.
Now you can say that such a table cannot possibly exist, but you would have to say why.
The Chinese room isn't a conversation thats the whole point. It actually unwillingly represents the neurons in my brain that doesn't understand Chinese either.
Searle is making the wrong analogy because all he shows is whats going on with the input/output when he really should show the entire house as the intelligent part.
This is absolutely beautiful. I'm interested in the comment that high level scrabble playing is more mathematical than linguistic; does there happen to be anyone here who could recommend some related reading?
At a high level, most people have memorized a large amount of the dictionary so the key is minimizing the score the opposition can get when you play a work, while still maximizing your score.
it is mildly statistical, but this basically pales to the amount of memorization you can do. It's good to set your rack up so that you have a good chance of getting decent letters in the subsequent turns, but the gradient of improvement is almost always in favor of "memorize more words" (admittedly, by going through the most common to least common groups of six letters plus one more) rather than math.
But in the strictest sense, statistics is more important than linguistics. It just doesn't really matter because memorization is so much more important than both of them.
The book Word Freak was pretty good and as far as I can tell, accurate.
> memorization is so much more important than both of them
Is that really true at this high of a level though? There are a little under 120K words in SOWPODS that are 9 letters or less (looking through some of those archived games, it seems like 10+ letter words are uncommon). So there's a hard upper bound on what memorization can do for you, and it seems plausible that that limit (or close enough to make strategy the dominating factor) would have been reached by a handful of people.
I mean, maybe? That's still a scenario where the vast amount of memorization you did is more important than the strategy you may come up with. There aren't top players who got there with superior strategy and a vastly inferior vocabulary.
> That's still a scenario where the vast amount of memorization you did is more important than the strategy you may come up with
But if both players have memorized about the same number of words, the game is then entirely about strategy.
I think that's the point of what the article is saying. Sure, memorization is critical to get to the point where you can compete at that level, but once you're there, memorization is just a given. What differentiates the players' performance is their strategy.
I feel like ‘vocabulary’ is a slightly misleading word in this context, since it implies some level of understanding. ‘Set of allowed combinations of letters’ is evidently the measure used.
We house-rule this. When playing Scrabble in my group of friends, we've added a rule that when a word is challenged, not only will the challenger look the word up in a dictionary, but the one who was challenged must correctly use the word in a sentence.
It's more fun that way (for us), and it gives us all an incentive to go learn both the spelling and meaning of obscure words.
This reminds me of my international student friends in college with perfect SAT Verbal scores but very poor writing/speaking/listening abilities. Boy were those guys smart.
I used to play online scrabble using a desktop app built by the Internet Scrabble Club (isc.ro)[1]. If you go to a website called cross-tables.com [2], you'll see tournament results and rankings of top players in the world. I've played online games against nearly half the players currently listed on the top 20 in North America and the world. I haven't played regularly for years, so my current Scrabble skills are pretty poor, even though I did beat many of those guys the odd time in the past.
One day I even played a number of "speed" games (3 minute games, those were fun!) against Trey Wright, the winner of the National Scrabble Championship in 2004. IIRC, I beat him in one game, by a single point, while he thrashed me in the other 9 or so.
In order to compete against these guys, you have to basically learn words for no other reason than to use them in Scrabble games. Word definitions are useless. It's a bit of an addiction, and quite pointless, but it's fun when you actually end up playing a bizarre 8-letter word (e.g. OENOMELS) and win the game because of it.
It's absolutely true that language has little to do with being good at Scrabble. I used to play against Thai players who had very little knowledge of English but who knew way more English words than I did or ever will.
Of course, when online, you pretty much have to expect that people will cheat. Most of the time, the cheaters are obvious, because good Scrabble players recognize cheating when compared to real human game play.
As for Nigel, the guy's not human. I don't remember if he was ever on ISC (many of the top players were at some point), but everyone in the Scrabble world knows him and knows his uncanny word knowledge and Scrabble abilities.
And if anyone's interested in reading about the Scrabble tournament scene and the associated lifestyles, check out Stefan Fatsis' book "Word Freak"[3], probably the best Scrabble book ever written (although it does contain quite a few f-words, thanks to the colorful personalities that it describes, which I wasn't crazy about).
They say he memorized "the" Scrabble dictionary. Is there a standard dictionary? Where can I see it? It would certainly help resolve conflicts next time :D
There's a few different dictionaries. There's the official tournament word list (TWL or OWL), which has had different versions. The latest, I believe, is OWL2014. I think the version before that was TWL2, IIRC.
But those books are only available to tournament scrabble players who pay an annual fee. It's a bit elitist, but that's just how it is (unless it has changed lately).
Other than that, you can pick up a regular Scrabble game dictionary, usually called OSPD, which will not have any off-colour words in it. Those can be bought in any book store and you can even search the OSPD on Scrabble's official website.
That picture looks like a mugshot. When that link opened up I thought, "Gee, you'd think they'd mention that he was in prison!"
That's really cool. I don't think that just because a person doesn't use a word regularly, that they don't 'deserve' the word or can't play the word on the board. There's THOUSANDS of words, nobody uses all of them.
I hate this attitude towards people who play games over time, and they start showing signs of game theory during play, and suddenly it's 'not about the game anymore', but it's always BEEN about the game. The goal is to win!
When you organize a Scrabble competition, you need to have an official dictionary or people start inventing non-sense words to get points. In French for example, you could add the prefix 're' (meaning again or back) to almost any verb and everyone would understand, this is just an example but it's really easy to create nonsense words that people understand from their root. And since you have an official dictionary, it's not about words anymore, it's about memorizing the dictionary.