Non-words in dictionaries

Not all words in dictionaries are real words. Some are bogus. Sometimes this is due to errors. In 1931, during the preparation of the second edition of Webster's New International Dictionary, someone wrote "D or d, cont / density" on a notecard, indicating that "D" and "d" should be added to the dictionary as abbreviations for "density". Through some misunderstanding, this was misinterpreted to mean that a new word should be added to the dictionary: "dord". And it was. Five years after the dictionary was published, an editor uncovered the error and had the definition deleted from future printings.

The second way that a non-word can show up in a dictionary is through someone copying erroneous words from another source. The history of dictionaries contains many incidents of unscrupulous dictionary-compilers stealing entire word entries from other dictionaries.

And this brings us to the third way that fictitious words can show up in a dictionary: They can be deliberately put there. It is common practice among dictionary publishers to insert completely made-up words so that when someone plagiarizes their work, they can catch them red-handed.

In 2005, one of the fake words in the New Oxford American Dictionary was revealed to be "esquivalience" which supposedly meant "the willful avoidance of one's official responsibilities".

In 1903, a dictionary of musical terms was published, and at the very end was this definition:

zzxjoanw (shaw). Maori. 1. Drum. 2. Fife. 3. Conclusion.

This definition was reused in Mrs. Byrne's Dictionary of Unusual, Obscure, and Preposterous Words: Gathered from Numerous and Diverse Authoritative Sources which was published in 1974, but with an alternative (and seemingly, completely fabricated) pronunciation.

If the definition "conclusion" and the incongruous pronunciation weren't evidence enough, a few other facts should have tipped people off that this word is a hoax: a) The Maori language does not contain the letters "j", "x", and "z". b) All Maori words end in a vowel. c) In traditional Maori culture, there are no drums!

Other compilers of information also introduce false information for the same reason. Phone book publishers have included fake names and phone numbers. Encyclopedias are published with false entries (like the New Columbia Encyclopedia's entry on Lillian Virginia Mountweazel, a fountain designer and photographer (famous for photographing mailboxes) who died "in an explosion while on assignment for Combustibles magazine". And map-makers insert fake streets (called "trap streets") or draw their streams as if they were slightly squigglier than they actually are.

What lesson can we draw from all this? Be skeptical. Don't just accept everything you read. Not even in dictionaries. Not even on this blog! And certainly don't believe words when they start with "zzx". That would be dordish.

Three-dimensional word games may be on their way

I've spent some time trying to think of a way to make a three-dimensional word grid for Scrabble or Bananagrams purposes. (The added challenge that an extra dimension would pose in such a game is very tempting!)

Cubical blocks with the same letter on all sides would almost work. Except that it wouldn't allow words to be build sideways from an upper level without other blocks acting as supports.

There are in fact, a lot of other games from the past that have been referred to as "three-dimensional Scrabble" or something similar, but they never seem to be successful, truly three-dimensional implementations.

Upwords is sometimes included in the category of 3-D word games, but it's physically more like layered Scrabble with corresponding differences in strategy. I mean, yes, technically it's three-dimensional, in that you need three-coordinates to locate a piece (row, column, height), but at any one time, it's only the topmost letter that matters. It's not really what people want when they think of 3-D Scrabble.

This patent from 1997 is for methods of making a three-dimensional Scrabble game.

A figure from the patent documents illustrates what I would consider closer to a true 3-D word game:


This patent laid out a bunch of ways that three-dimensional constructions might be done (connecting letter spheres together with magnets, velcro, and various mechanical tricks). As far as I can tell, none of these ideas ever made it to production.

Another patent was quite recently applied for for a three-dimensional word game. The inventor is listed as Joseph Elie Tefaye with Games R Us of Australia.

This figure shows the essence of the idea:


The design is reminiscent of the two-dimensional Typ-Dom, a predecessor of Scrabble invented in Austria in 1936. In Typ-Dom, the letter tiles are shaped like jigsaw puzzle pieces, so that they all link together. It's an interesting way of keeping letters in place without a board. This actually strikes me as superior to having a board in some ways: You can extend the grid as far as you want in any direction, and if your table gets jostled, the tiles won't be thrown out of position. In fact, you wouldn't even need a particularly flat surface to play on.

The extension to making Typ-Dom into a three-dimensional word game just required two main ideas: 1) Using a base to build the three-dimensional word grid from, and 2) modifying the pieces so that one can link up with another at a 90 degree angle. It looks like this might work out nicely.

But if you can't stand waiting for someone else to manufacture the perfect 3-D word game, you could always make your own, Star-Trek-style:

Online Speed Scrabble: a review

I discovered that there is a web site called Supernifty which has an online Speed Scrabble game that can be played for free. Naturally, I decided to try it.

To play, you have to go through a straightforward registration process which requires only an e-mail address, a password, and a nickname. Once you've logged in you are greeted by a listing of available games that you can join, or you can create your own game. The feature that I particularly like here is that you can give a name to your game by typing it into the blank before creating the game. In principle, this gives players the ability to set up games for different skill levels or with different rules. Or if someone wants to practice on their own, they could call their game "Solitaire Only."

The game is an implementation of the "Take Two" variant of Speed Scrabble: You start with seven tiles, and once someone has formed a valid grid, two more tiles appear on the top row of every player's board, and play advances to the next "round". There is some scoring system where you get credit every time you win a round. Whoever wins the tenth and final round wins the game (and gets a lot more points). Points are also based on how long your longest word is in each round, and on how competitive your opponent is.


Diligent readers of this blog will know that I am not a fan of Speed Scrabble, but this online version does away with a few of the nuisances of the offline game: Starting with a small number of letters and no ability to dump tiles makes it hard or sometimes impossible to form a valid grid early in the game. If you encounter such a situation in Supernifty Speed Scrabble, you can just skip that round by pushing the "Give up on round" button to get two more letters. (At least you can do this when playing alone, I did not run into this situation when playing with other players.)

I was a little surprised at how often the initial set of letters could be formed into a grid of words, and I started to suspect that this was by design when I got letters that seemed to coax me into forming a symmetric grid for a couple of rounds (as shown). Whether or not this is built into the program is something that I can't tell yet...

The game is implemented in Javascript, rather than Flash, which may be the reason that the interface is more minimalist. I liked how smooth and clean it felt, and since many of the games I played were solitaire games, it was overall kind of a relaxing experience (except the few times that the timer started to run down and I had to franticly drag tiles around the board.) My only frustration with the interface was that it's not possible to select a rectangle of tiles (as in the electronic version of Bananagrams) and move it around. This sometimes leads to having to shift a bunch of words from one place to another, one tile at a time. Aside from that issue, it is otherwise an intuitive interface, so you can just start playing straight away. If you'd like a more detailed description, try the official description of Speed Scrabble.

There were only a few people drifting on and off the site while I was playing, and often I was the only player on the site at all. If I am interpreting the scoring system correctly, the list of high scores corroborates the idea that there are just a handful of hard-core players and some others who have played on many occasions. But the site just came online in August of 2009 and an iPhone client for the game has been recently released, so the number of players may be on the rise.

If you like online word games and you want to try something different, I can recommend Supernifty's version of Speed Scrabble.

Bananagrams lecture on March 11, 2010

Abe Nathanson, the creator of Bananagrams, is going to give a lecture, titled "The Success of Bananagrams", Thursday, March 11th, at 5:30pm at Providence College in Providence, Rhode Island. The lecture will take place in Moore Hall II.

The map on the event page actually has the wrong location. I've plotted the actual location on this Google Maps map with the lecture hall dead center:

I'd really like to hear this lecture as I'm sure it will be interesting. If anyone gets the chance to attend this talk, please send me a report.