Talk:Word square

Example 12x12 Square: clarification needed

As per the talk page for "The Book of the Sacred Magic of Abramelin the Mage", there are a couple of issues with the description of the 12x12 square in this article. I know nothing about this book, so perhaps someone else can help?

1. The phrase "According to Dmitri Borgmann's book..." suggests that Borgmann found this square through some interpretation or analysis of the original, and that the existence of the square is not a clear-cut matter of fact. But according to the above-mentioned talk page, the square appears in full in the original.

2. Also according to the above-mentioned talk page, "none of these words are listed as being the names of angels, incubi, demons or kindred souls" (as is claimed).

3. The word "hoax" implies some deception, but it is unclear what that deception is and who perpetrated it. If it simply means that the words were deliberately invented so as to form a word square, with no attempt to pretend they are "real" words, then "hoax" is the wrong description. Clarification is needed.

Matt 11:25, 2 September 2006 (UTC).

ANSWERS! 1. Many squares appear in Book Three of the English version: they are spells: this one turns Animals into Stone. 2. What I have seen is consistent with the "words" being pure invention. 3. It is a hoax if anyone pretends it is a word square. It is also a hoax if you do not believe it will transform Animals. Please see my update on the main page. Rex Gooch 9 Nov 06

Editing the Word Square page

Some time ago, I updated this page with discoveries of the past few years, which have not merely involved very important new squares, but also a quantitative understanding of what is and what is not possible.

Those changes have all been removed without any notification. Moreover, a false statement seems to have been used as a reason.

It would be courteous to let me know if you disagree with anything. My email is rexgooch@ntlworld.com

I think the deleted text you are referring to is this:

[Additional comment, 2 Mar 2005: various 10-squares constructed in 2002/3 are in fact accepted by the experts. Typically they contain two place names: I "proved" that the chance of making a ten-square wholly of words from a giant version of any dictionary is less than exceedingly small! Look for my DESCENDANT square, agreed at the time of this note to be the best. I have also published an 11-square, though this requires a mixture of languages. The big advance in the last five years has thus been to delineate what is and what is not possible: that's a lot tougher than just finding something. Rex Gooch]

I deleted this because it is written as chatty first-person research/opinion, which is not the Wikipedia style (and it didn't seem at all easy to fix without specialist knowledge, otherwise I would have done so). The idea of Wikipedia (in theory!) is that the articles should be of a style and quality similar to that of major published encyclopedias. If you looked at the "word square" entry in, say, Encyclopedia Britannica, would you expect to read the above, or anything like it?

Wikipedia editors are generally not notified when changes are made to text they have written. It would make the system unworkable: every time anyone changed anything they would have to look back through the whole edit history, try to figure out who had contributed to the text they were changing (which could be many people), and spend an inordinate amount of time notifying those individuals. It can be very annoying when people change or remove stuff you've written, but it's just one of those things. However, what you can do is create a Wikipedia account and username for yourself and add articles to your "watchlist". Any changes to the nominated articles are then flagged up to you. (At least, this is how I understand it to work... I've never actually used it myself.)

Having said all that, it's great to have some expert input into this page, and the new text is a major improvement to the article. Here are my detailed comments on the new text (version as of 4-Nov-06):

"The acceptability of a word square depends upon its size...

Not really. From what follows, I think the point is that the larger the square, the more difficult it becomes to avoid using obscure and questionable words. However, this opening statement doesn't say that - it says something that is patently untrue.

...for a 4-square it is reasonable to require all words to be recognisable by a child.

Not really. I guess what is meant is that it's easy to construct such a square. Requiring the words in a 4-square to be recognisable by a child is not reasonable at all.

...Hyphens, apostrophes, capital letters, and words last in use hundreds of years ago are to be avoided if possible. Some even mandate all words are to come from a named dictionary. However, such restrictions sometimes produce unwelcome results: for example, most people would prefer a well-known place name like Manchester to an obscure Anglo-Saxon word.

Doesn't follow. The restrictions mentioned would exclude both "Manchester" and "an obscure Anglo-Saxon word".

...Regarding 9-squares, it is just possible (but only in 2003) to find a 9-square whose words are all contained in the combined OED, Webster 2, and Chambers.

I'm not entirely clear whether this means that such a square was known to be impossible given the list of words in the editions prior to 2003 (someone did an exhaustive search), and only became possible when words were added in later editions, or whether such a square was first found in 2003 (and may have been possible for some time previously).

A 10-square is naturally much harder to find ... and its finder has been promised immortality.

I'd be very tempted to delete the joky statement "has been promised immortality".

... It cannot be constructed from dictionary words alone, but can be if we permit words in gazeteers.

So, someone has done an exhaustive search covering all published dictionaries, yes?

...there is strong proof (both mathematical and experimental) that no combination of dictionaries would suffice.

It is impossible to "prove" this mathematically. All that statistical methods can do is tell you it's very unlikely. I suspect that "strong proof" actually means "strong evidence".

Jeff Grant had a long history of producing ever better 10-squares, then Rex Gooch produced a great leap forward (Word Ways, August 2002).

I would be tempted to tone down the phrase "giant leap forward".

...An account of how the ten-square was found is in 'Hunting the Ten-Square' in Words Ways May 2004 (available on the Internet).

A link to the relevant web page would be good here.

But above all, word squares have been put on a scientific basis: we can, for example, say that 250000 10-letter words are needed to make a 10-square

This can't be said. It depends what the words are. I guess this is referring again to statistical methods, which cannot give exact answers about how many words are needed, just probabilities.

Eleven-squares are far harder again to construct, and cannot be done using English words (even including transliterated place names).

Again, this is a very definite statement about something that I would think it's impossible to be definite about. No-one has ever compiled a complete list of English words, nor is ever likely to.

Matt 11:11, 4 November 2006 (UTC).

Rex Gooch's replies 7 Nov 06 Thank you for your remarks. Please read the changed article which takes account of some of your comments.

4-square and children. Utterly reasonable to use simple words. Indeed, I would be lambasted if I used the average 4-letter word in my list. It is meet and elegant to use the simplest solution - like Occam's Razor.

The restrictions mentioned would exclude both "Manchester" and "an obscure Anglo-Saxon word". and that is what I said - that's the unwelcome result.

I'd be very tempted to delete the joky statement "has been promised immortality". It is a much-quoted statement by an expert. It shows the enormous perceived difficulty of the task. It was not a joke.

So, someone has done an exhaustive search covering all published dictionaries, yes? In effect. Few words are missing from the big dictionaries. Moreover, adding 1000 normal dictionaries would not be expected to add anything significant. Dictionaries that do not repeat almost every word in others are unsalable. You haven't grasped the hugeness and comprehensiveness of the word list.

there is strong proof (both mathematical and experimental) that no combination of dictionaries would suffice You are nit-picking. OK, so in a general encyclopedia, you want me to start many statements by "Given the null hypothesis that....then the chance that the data observed....."? Really? It is not normal in medicine (hot-bed of sloppy statistics), or in issuing death sentences, to say that proof beyond reasonable doubt is no proof. Not "very unlikely", but nugatory, exiguous. I'll bet you don't avoid going out because the Titanic might arise, travel, and drop on you. That is true with a low probability, but I am happy to say it will not - for sure. Anyway, we're in an area where mathematical and psychological probabilities diverge.

I would be tempted to tone down the phrase "giant leap forward". and I would tone it up! You again have not grasped the utterly fundamental change that has occurred. After 80 years of virtual stagnation, suddenly we have a 10-square, which is a great leap forward (just ask those who have spent decades of ther lives on the task). In an even greater leap, we have an 11-square and other large difficult squares. In a truly huge leap, we have progressed from trial-and-error to accurate prediction.

(what is needed for a 10- and 11-square) This can't be said. It depends what the words are Again, this is a very definite statement about something that I would think it's impossible to be definite about. Well, I said it, using my skills as winner of the Cambridge Science Prize, and 40 years of experiments and statistics. The maths made predictions, experiments confirmed them. You cannot get more solid than that. In science, that is. If you try to argue that a coin that has come down 50 times heads has an equal probability of doing the same again, you need treatment! That ten new words admitted this year to the OED would form a 10-square is not merely unlikely, it's in lunatic territory.

In the double square of order 8, I can't find any English references to the word "ENERGATE" in any dictionary nor even google ngrams. — Preceding unsigned comment added by 2601:645:8A80:8120:884A:1040:B49:2E9C (talk) 00:38, 9 January 2024 (UTC)

Diagonal word squares

Does anyone have any clue what the following means?

"Diagonal word squares are word squares in which the main diagonals are also words. There are three diagonals: top left to bottom right, top right to bottom left, and a palindromic one."

I guess that palindromic words along the diagonals are preferable because the diagonal can then be read in eaither direction, but I have no idea why there are "three diagonals". Matt 02:30, 3 April 2007 (UTC)

Immortality

Other than this page, I can find no instance of, or reference to, the "much-quoted" promise of immortality mentioned in this article, nor whom the "expert" is. Perhaps Mr. Gooch could elaborate? —Preceding unsigned comment added by Tennjam (talk • contribs) 05:43, 20 October 2007 (UTC)

I'm afraid he cannot now; Rex Gooch died on 13 March 2007 [1]. However, in a 2002 Word Ways article Gooch wrote, "I believe it was Jeff Grant who wrote that the first person to find a ten-square would achieve immortality." [2] —Psychonaut (talk) 17:57, 28 April 2013 (UTC)

Article needs to reference the great "Paser Crossword Stela" c. 1,500 BC Egypt

https://en.wikipedia.org/wiki/Paser_Crossword_Stela

This is a "word square", only each entry was a word or two. It was 67 by 80 squares and in Egyptian Hieroglyphic.

Dwarfkingdom (talk) 03:01, 3 January 2015 (UTC)

Adding External links

Hi. Is it possible to add this link as a resource: http://www.word-square.com/

It's a free online word square puzzle. I believe it might fit here... — Preceding unsigned comment added by 85.65.104.44 (talk) 18:12, 5 August 2017 (UTC)