Talk:Sudoku

From Wikipedia, the free encyclopedia

This is the talk page for discussing improvements to the Sudoku article.
Article policies
Archives: 1, 2, 3
Former featured article Sudoku is a former featured article. Please see the links under Article milestones below for its original nomination page (for older articles, check the nomination archive) and why it was removed.
Main Page trophy This article appeared on Wikipedia's Main Page as Today's featured article on June 10, 2005.

This article uses British english dialect and spelling.
According to the relevant style guide, this should not be changed without broad consensus.
This article is within the scope of WikiProject Japan, a project to improve all Japan-related articles. If you would like to help improve this and other Japan-related articles, please join the project. All interested editors are welcome.
A This article has been rated as A-Class on the assessment scale.
Low This article has been rated as Low-importance on the importance scale.
Japan To-do:
Image:Crystal 128 gadu.png Good Job
"The exhaustive Wikipedia article on sudoku goes into even greater detail" — New York Times, May 1, 2006
News This page has been cited as a source by a media organization. The citation is in:
Sudoku is included in the 2007 Wikipedia for Schools, or is a candidate for inclusion in future versions. Please maintain high quality standards, and make an extra effort to include free images, because non-free images cannot be used on the CDs.
Archive
Archives
  1. January — September 2005 (part)
  2. September 23 (part) — November 2005 (part)
  3. November 2005 (part) — May 2006


Contents

[edit] Construction Section Unclear

I'm just a casual visitor here, but I found the Construction section of this article to need improvement. I might have tackled that but I'm not sure what was trying to be accomplished here.

FIRST SENTENCE

"Building a Sudoku puzzle IS performed by predetermining the locations of the givens and assigning them values only as needed to make deductive progress."

Need definition of "givens". I think I know what they are but this paragraph is confusing about them. Are they the numbers that appear in a new puzzle? Does this include their locations?
Couldn't we say something like, "Building a Sudoku puzzle, in simplest terms, consists of assigning predetermined values (givens) to an empty puzzle board in a way that leads to a unique solution".

SECOND SENTENCE

"This technique gives the constructor greater control over the flow of puzzle solving, leading the solver along the same path the compiler used in building the puzzle."

Greater control as compared to what? What does the computer have to do with this? (I know a computer program can create Sudoku puzzles, but surly they can be created manually as well.)

THIRD SENTENCE

"Great caution is required, however, as failing to recognize where a number can be logically deduced at any point in construction—regardless of how tortuous that logic may be—can result in an unsolvable puzzle when defining a future given contradicts what has already been built."

Is this trying to say, "don't screw up"? We could say, "It is possible to create unsolvable puzzles so be sure you can solve the puzzle before releasing it into the wild."

FOURTH SENTENCE

"Building a Sudoku with symmetrical givens is a simple matter of placing the undefined givens in a symmetrical pattern to begin with."

What are "undefined givens"? If they are undefined, then they are not "given". Thus my confusion over the term "givens".

CONCLUSION

After reading this section I have to say I still don't know what the rules are for constructing a Sudoku. Must it have one and only one solution? What is this stuff about symmetrical puzzles? What are the conventions around construction? I can't answer these questions. I came here to find the answers.

Peace,

R39525 (talk) 20:53, 8 May 2008 (UTC)

[edit] Another opinion

I guess I wasn't here for the free-for-all, but I'd still appreciate links like www.sudoku.com, www.websudoku.com and www.e-sudoku.com, or even in Spanish www.sudoku.eseven though they are "selling" a product. They still have useful information in them, and a person looking at an article about Sudoku might benefit from them; maybe even buying a game (like I did).

Perhaps a compromise where the reader is warned that the links sell a product is all that is needed. --KSnortum 20:30, 26 June 2006 (UTC)

In theory, I would have no objection to those links being added (and they were there for quite a while). However the more links there are, the bigger the magnet the External Links section becomes for others to add their own contributions of very dubious quality. So, on that basis I'd vote to leave it alone. Besides, http://www.sudoku.com is already in the References section. --angusj 22:32, 26 June 2006 (UTC)
I would point out that many "External links" sections point to places that sell things. For example, the article for ComedySportz contains external links to the individual cities. They most definitely sell admission. It would seem better to me to throw out the dubious links instead of the baby with the bath water. --KSnortum 23:54, 26 June 2006 (UTC)
The management issue of the External Links section has nothing to do with whether a particular site is commercial or not. It's all to do with offering quality links and how that quality can be adjudicated. Without the current vetting (which is admittedly ruthlessly uncompromising) this section will quickly become another "free for all" which it was previously. --angusj 04:41, 27 June 2006 (UTC)
I added Sudoku links to websites i got by Googling "Sudoku". They were the top couple of choices. I hope they will be ok. dposse 02:20, 29 July 2006 (UTC)
And I took them back out, based on the "magnet" effect starting to kick in. If people can get those links off the top several choices in a Google search, they're not doing much useful here, except to give other people "cover" to fill that section with spam links again. MDWthing
All i did was to google Sudoku] and i picked out the top couple of links, which are great websites. I guess we have to find a way to keep idiots from spamming this article. dposse 16:13, 2 August 2006 (UTC)


Hi, what do you think about www.one-sudoku.com? Should it be here? 85.160.70.226 01:22, 7 August 2006 (UTC)

what's the case for including it? (I've removed it at least for the moment) MDWthing
i thought it might be interesting as a site with sudoku with score by nations.85.160.73.115 23:18, 7 August 2006 (UTC)
do you think it shoudln't be here? it's not commercial... 85.160.68.138 01:08, 9 August 2006 (UTC)
No. It's far from exceptional. (Also, when adding comments here it is customary to add them at the bottom of a section. It helps others follow the chronological flow. That's why I moved this mini-section.) --angusj 03:35, 9 August 2006 (UTC)

The grammar in this article needs a great deal of work. Search on the page (Crtl F) for instances of "have been" to find all of the passive voice sentences.

[edit] Linkspam is Inappropriate Labeling

User Angus J. seems a little overzealous of removing any and all links that could potentially bastardize the quality of the article. According to "What to link to," it appears that most of these links are valid entries. While there may have been an issue of link spam, (I only found this because I thought, That would be a really good obscure title to use to onobtrusively googlebomb a page. while reading the list of edit summaries as I was looking for the legendary free for all link inclusion that apparently occurred over a year ago.) Apparently that was exactly what it was used for.) the main worry has been simply that there would be too many links (even though it would be valid links). Calling this linkspam is insulting. Reverting link adds and flagging people for linkspam would be about as appropriate as it would be for me to revert Angjus J.'s linkspam and reversion edits and start flagging him for vandalism (See Blanking) It would be better to use a categorization system, as originally proposed. The current External Links section is far too limited as is anyway. I'd also like to point out that considering the fact that there is recurring dissent by people new to this article (often overruled by the regulars), it seems that the current policy isn't following Consensus. -Caudax 07:00, 6 July 2006 (UTC)

OK, I understand that you strongly disagree with my editing. However, I was *not* acting alone in drastically pruning the External Links section. Nor was I the person who added the comment hint message at the top of this section asking people to propose sites here (and gather a consensus) *before* adding new links. The External Links section *did* get totally out of control because contributors displayed a total lack of objectivity in the sites they added. (Presumably they were mostly links to their own sites.) Links should be to *exceptional* sites only, since there are literally thousands of Sudoku sites. You evidently decided to disregard the request to propose sites here first, when you added http://www.chrysocome.net/sudoku. I believe that link is far from exceptional and very well demonstrates this lack of objectivity I mentioned above. However, if you can persuade a reasonable number of contributors here to decide what should be in External Links then I'd be more than happy to 'butt out'. Also, if a number of other regulars here feel I'm being too heavy handed then I'm equally happy to 'butt out' and leave this to others. --angusj 13:30, 6 July 2006 (UTC)
I think that the pruning of the Ex Links section is in general a good idea (and I do appreciate the time that users, including yourself, are putting into it. It's just important, imo, to explain the reasons for doing so (as you've been doing) and also to assume a good-faith attempt to improve the page (as you sometimes do not seem to). 65.115.38.32 17:42, 18 July 2006 (UTC)
Yes, I'll take on board your suggestion to assume 'good-faith' initially. Thanks. --angusj 23:12, 18 July 2006 (UTC)

I suggest adding http://www.inertiasoftware.com/top50/ to the links, because that is a big resource of sudoku quality sites (I am not from that site, just think it is great!)

[edit] The French Connection

At first glance it looks like Sudoku, but is it? I can't work it out, am I right in saying that it's a different sort of puzzle with a strikingly similar appearance, where you have to add the values of the boxes up rather than fill in a specific unique symbol? Jooler 01:49, 5 June 2006 (UTC)

Ahh the newly added picture from La France shows the symbols 1-9, so this is more clearly like Sudoku without the 3x3 boxes. Jooler 01:53, 5 June 2006 (UTC)
See also the discussion under Proposing External Links. (Perhaps it should be moved here?) --angusj 05:11, 5 June 2006 (UTC)

Here it is. Le poulet noir 13:55, 27 June 2006 (UTC)

New research has been published in Pour La Science, the French version of Scientific American, that shows how Sudoku was invented by French puzzle setters in the 19th century. An English language analysis of this article was published in The Times on June 3, 2006 and can be seen here: http://www.timesonline.co.uk/article/0,,2-2208881,00.html I reckon it is worth having that article as an external link. Any objections? Le poulet noir 15:21, 3 June 2006 (UTC)

There's no evidence that Howard Garns was aware of, let alone copied a French puzzle which was allegedly published over 100 years ago, so I think it's a big stretch to say he 'revived' the puzzle. Also, from the link you gave it's unclear whether the current Sudoku rule applied in this French puzzle (ie the digits 1 to 9 must be in each row, column and box). However, if/when it can be reliably confirmed that this Sudoku rule was specified in the French puzzle (yes I understand there was also a constraint on the diagonals adding to 45) then I think it would be appropriate to document this early French variation in the main article. --angusj 23:34, 3 June 2006 (UTC)
See http://en.wikipedia.org/wiki/Image:Sudoku.jpg for evidence that La France published the Sudoku with the rule that you have to use numbers from 1-9. (It's in French, but you don't need a very deep understanding to be able to translate it.) It reads: "Completer le carré ci-dessous en employant les neuf premiers nombres chacun neuf fois de maniere que les horizontales, les verticales, et les deux grandes diagonales donnent toujours a l'addition le word obscured total." The sub-squares are not marked, but the solution requires that each 3x3 box contains the numbers 1-9. Le poulet noir 19:03, 4 June 2006 (UTC)
OK, my French is limited but I see no mention in these instructions about the nine 3x3 boxes found in a Sudoku. (In this puzzle there are four 3x3 boxes, not the nine found in a Sudoku.) Here's my attempted translation of the instuctions: "Complete the square below by using the first nine digits, each one nine times, so that the horizontal ones, verticals, and the two large diagonals always add to the same total." Of course there are many similarities between this puzzle and a Sudoku, but I would argue that this puzzle is not a Sudoku. Having said that, I think it is certainly worthy of a mention in the main Sudoku article - as long as it's not called a Sudoku - which is what I see you have called it. It is only interesting and relevant here since documents very well that these sorts of number puzzles have appeared in newspapers for a long time. --angusj 22:17, 4 June 2006 (UTC)
Fair enough. I was about to modify it, but you got there first. I have added a few more details showing the similarity of the French puzzles to the Garns version. I also put in a picture of the puzzle from the 1892 edition of Le Siecle because it has the strikingly similar 3x3 sub-squares marked out. Le poulet noir 00:17, 5 June 2006 (UTC)
OK, I think we're now fairly close to common ground. :) --angusj 05:09, 5 June 2006 (UTC)




[edit] Tournaments Section Disappeared!

Someone deleted the Tournaments section without even notifying anyone on this discussion. Can someone please restore it. I remember that there was the Sudoku Champion mentioned and the website: http://www.unitedstatessudoku.com/ mentioned.

Yes, it looks like a known vandal removed that. Glad you reinserted it. Just wondering, could you provide an explanation for the paragraphs you've just removed. I'm reading the discussion above and I get that people misattribute sudoku's invention but the article wasn't doing that. Latin squares is less relevant, Nikoli is very relevant and Da Vinci Code parodies, though not integral, are examples of the game's popularity.--Lo2u 19:55, 3 July 2006 (UTC)
I vaguely seem to remember that the http://www.unitedstatessudoku.com/ competition was cancelled due to lack of interest. Also their website is very primative, there's no indication of any organisational structure and their forum has only 3 posts (all by the one person). I really think its premature to give them any exposure here. --angusj 06:33, 4 July 2006 (UTC)

Perhaps you already know, but speaking of tournaments and multiplayer sudoku playing, eSUDOKU [1] is offering an online multiplay sudoku game where up to 4 players can compete against each other. The beauty of it is that it is still at no cost. A bit funny as there is no advertising involved.

[edit] Sudokube

There is a new rubiks cube/sudoku variant called the sudokube. Do we already have an article on that?--Kylemagne

Stillbourne Indeed that would be very intersting!

[edit] References

I believe the References section needs a major clean up, having become a defacto External Links section. It should only contain references to items mentioned within the main body of the article. Is anyone interested in fixing this?

I'll second the motion. Proposal: remove all the "mathematics" references (they should be attached to the mathematics of sudoku wikipedia page instead) and all the "Java" references (nothing currently in article on Java implementation that I noticed). Any objections? MDWthing

[edit] Wikipedia entry does not square with MSNBC news story

This Wikipedia entry does not seem to square with this MSNBC story on Sudoku.

Sudoku generates big numbers for its promoter Logic puzzle becomes million-dollar enterprise for retired judge CNBC Updated: 11:57 a.m. ET July 14, 2006 Bill Griffeth

[removing text of news article -- violates copyright and conversation is old - in future provide links to news stories under discussion]

Michael D. Wolok 15:11, 15 July 2006 (UTC)

What doesn't square? You're comparing a news article to an encyclopedia article. --KSnortum 00:37, 19 July 2006 (UTC)
I agree with Ksnortum. MSNBC has merely reported the story from the perspective of the craze in Britain and America, not the whole history of the puzzle. Le poulet noir 14:52, 19 July 2006 (UTC)

[edit] The French Connection redux

This sentence, I feel, is incredibly misleading: Le Siècle, a French daily, produced a 9x9 grid with 3x3 sub-squares as early as 1892, but used double-digit numbers rather than the familiar 1-9. A person reading this would naturally assume that it was Sudoku with different symbology - double-digit numbers rather than single digit. However, the reality of this is very different. Reading over the newspaper photocopy, it became very clear the rules of that puzzle are to find numbers that add up to a particular value -- something akin to a magic square variant, not a Sudoku variant. This is an important distinction because Sudoku is a logic puzzle while this puzzle is clearly a mathematics puzzle. Other than its strikingly similiar grid, they have absolutely nothing in common. I think having this in the article clearly confuses the history of Sudoku and should be removed altogether. -- ShinmaWa(talk) 04:41, 2 September 2006 (UTC)

I entirely agree and have removed the relevant sentence. --angusj 07:52, 2 September 2006 (UTC)
I disagree. What this shows is the evolution of the puzzle in the 19th century. There was evidently a craze for number puzzles in the French press at the time and there were two puzzles that were very similar to the modern Sudoku. Bear in mind that no one had tried to make magic squares into a puzzle by removing numbers before. Le Siecle's puzzle was the first to have a 9x9 square with 3x3 subsquares. It shows a link between the magic square and the logic puzzle in the same way that a fossil record of homo erectus shows the link between apes and humans. La France's puzzle, three years later, shows another leap in the evolution to what was very nearly a modern Sudoku. It lacks only the 3x3 subsquares (although if you do the puzzle you will find that each subsquare does indeed contain the numbers 1-9). I'll revert this in a few days unless someone can give a reason that the History section should not show the evolution of Sudoku and its antecedents. Le poulet noir 16:42, 3 September 2006 (UTC)
I understand what you are saying, but can you show that Sudoku (a logic puzzle) evolved from a mathematics puzzle? Other than the grid, which I'll admit is very similiar, there's truly no connection. Sudoku requires no math skills at all -- it can be played with letters or shapes or colors, while magic squares is entirely an exercise in algebra. They are truly apples and oranges. To say that just because the grid is the same that they share a common history sounds like Original Research to me. This argument is akin to saying that Chess and Draughts share a common history because they both are played on a 8x8 checkered board (they actually share no common history at all). The La France daily connection is hard to dispute. They have a similiar grid AND similiar rules. But, a magic squares to Sudoku connection? That one I have a hard time buying without source material from a reputable source that states that precisely. -- ShinmaWa(talk) 19:51, 3 September 2006 (UTC)
It is not original research. The discovery of both of the French puzzles and their connection to magic squares was in an article published in Pour La Science (http://www.pourlascience.com/complement/pdf/Suppl%20Ancetres%20Sudoku.pdf) (the French edition of Scientific American). It was reported subsequently by both The Times (http://www.timesonline.co.uk/article/0,,2-2208881.html) and Wayne Gould (http://www.sudoku.com/forums/viewtopic.php?t=4228), the puzzle setter. I don't think the chess and draughts analogy is very helpful. If chess and draughts had been invented in the same city within three years and were both variants on the same mathematical formula then it might have more currency.
I acknowledge that Sudoku is a logic puzzle and Le Siecle's puzzle involves arithmetic, but Le Siecle's puzzle is significant because it is the first time a magic square had numbers removed to make it a puzzle.
Sudokus are magic squares without diagonals. You may not need to prove it to complete the puzzles, but every row and column of a Sudoku adds up to 45. Is it just a coincidence that these apparently modern puzzles happen to be long-established mathematical constructs? You seem to argue that it is, but only by ignoring the evidence that connects them.
Paris in the late 19th century saw newspaper puzzle setters competing to find more entertaining puzzles for their readers. In 1892, Le Siecle creates a 9x9 magic square, divides it into 3x3 subsquares and removes numbers to make a puzzle. The sum of the rows, columns and subsquares is 369. Mathematically, the only difference between this and the modern Sudoku is the sum of the numbers. It is a quantitative, not qualitative difference.
This idea was simplified over the course of three years so that La France published a 9x9 puzzle in which the sum of rows and columns (and 3x3 boxes, although they are unmarked) is 45 - the same as a modern Sudoku.
You acknowledge that La France's grid is similar to Sudoku, even though the two are separated by 84 years and 4000 miles. Yet you are unwilling to acknowledge that La France's puzzle was influenced by other number puzzles published three years apart in the same city.
All of this is pointed out in the Pour La Science article. I cannot find the full version online, but the precis I gave the link to (above) is a fair summary. Le poulet noir 18:10, 5 September 2006 (UTC)
I do appreciate the links. Thank you. The synopsis from Pour la Science is weak at best, but I believe the original article that it summarizes is much more concrete. Out of all the references you provide, I believe the summary at sudoku.com [2] is probably the clearest and most accurate summary of Pour la Science's' findings. Here's my recommendation: The way it was worded before did not quite follow the findings of Pour la Science. After reading both the Pour la Science synopsis and Wayne Gould's analysis, I understand much better. I believe The Times mangled it somewhat and confused the issue. Let's resummarize Wayne Gould's summary and include that (and choose one graphic or the other, but not both, provide a link to the synopsis instead). I think we have consensus. Thanks. -- ShinmaWa(talk) 18:52, 5 September 2006 (UTC)
Consensus ahoy. Let's use the pic from La France, then. I'll do a rewrite of the first paragraph of the history section later today. Also, I think we should amend the second paragraph too. It states: "Garns added a third dimension to the traditional Roman practice of Latin Squares and presented the creation as a puzzle, providing a partially-completed grid and requiring the solver to fill in the rest." It is clear from the French newspapers that the idea of a partially completed grid was not his innovation, although he may have believed it was. Le poulet noir 15:27, 6 September 2006 (UTC)

[edit] Addition of link to Solving Techniques

I suggest we could put a link or more information related to basic and advanced solving techniques used by SuDokers.

Eg: http://www.sudoku.com/index.htm

Also, I did not go through the Rules before putting that Compulsive Sudoku Syndrome bit. If anyone feels it is not appropriate, please post here and remove. - coolmuths

Its already in the article (reference #3) -- ShinmaWa(talk) 01:31, 4 September 2006 (UTC)

Thanks, but the link to solving techniques is not present. Also no mention of compulsive sudoku syndrome anywhere. I would think it is an interesting piece of information for Sudokers. - coolmuths

If you are looking for a basic but important information on how to get started about solving Sudoku, you can see some Sudoku Hints in the next link:

http://www.e-sudoku.com/onlinegames/sudoku/hints/index.htm [Source: eSUDOKU, October 2nd 2006]

[edit] Prime WP:FARC candidate

This article certainly does not exemplify the high standards of featured articles on Wikipedia. Stubby paragraphs abound, certainly not an FA characteristic. However, most offensively, the article does not cite a single source until halfway through (The first ref or external link does not occur until the Construction section). Even then, some of the more technical sections, such as how computer solutions are completed, do not have any references whatsoever. This article is not a good representation of the work that is featured article quality on Wikipedia. If referencing and other issues are not taken care of, this article will be moved to WP:FAR then WP:FARC immediately. — Scm83x hook 'em 08:34, 5 September 2006 (UTC)

One of the major problems of referencing in this article is the propensity for it to generate links to very poor quality sites. There are just too many kids out there with sudoku websites just itching for attention. For example, adding a couple of references to solving techniques immediately opens the flood gates to all and sundry to add or change the links to their own sites. It's been almost impossible to gather consensus as to which sites stand out from the multitude. Having said that, the article would definitely benefit from a major pruning. That at least would remove much of the unreferenced material. --angusj 13:47, 5 September 2006 (UTC)

While I take exception to slightly aggressive and threatening tone of your message, Scm83x, I tend to agree with your assessment of the article. It is currently not up to the caliber of FA. We would, of course, welcome constructive criticism with concrete points of concern (for exzample, what do you consider a stubby paragraph?). Could you perhaps lend a hand in returning this article to its former glory? -- ShinmaWa(talk) 17:37, 5 September 2006 (UTC)

One of the best ways to make people move is to tell them that you are going to act. WRT Angusj's comments, have the authors of this article considered using books as references instead of relying on the internet? Stubby "paragraphs" include:
  • The attraction of the puzzle is that the rules are simple, yet the line of reasoning required to solve the puzzle may be complex. The level of difficulty can be selected to suit the audience. The puzzles are often available free from published sources and may be custom-made using software.
  • Later in 2005, the BBC launched SUDO-Q, a game show combining Sudoku (albeit only the 4×4 and 6×6 variants) with a general knowledge quiz.
  • During February 7th's episode of The Daily Show, correspondent Jason Jones suggested that to ease the conflict over the Jyllands-Posten Muhammed caricatures, newspapers should be stripped down to only featuring Sudoku puzzles.
These are not statements strong enough to warrant their own paragraphs. In addition, this article had a very poor traffic on its FAC and did not really look all that good when it was promoted in the first place. Take a look at WP:WIAFA, especially with regard to 1a and 1c, and you will see that this article is not even up to WP:GA standards. — Scm83x hook 'em 22:19, 5 September 2006 (UTC)
Thank you for your feedback. Feel free to click the [edit] buttons and help us fix what you don't like. -- ShinmaWa(talk) 01:20, 6 September 2006 (UTC)

I agree that this article needs a makeover to ensure that references are thorough and it reads less like it has been written by committee, which of course it has. That said, I think guardians of this page, notably Angusj, have done a great job in preventing it from overflowing with links. Le poulet noir 17:29, 26 October 2006 (UTC)

[edit] Sudoku Variants taking over

Having gained some control over links to standard Sudoku websites, there seems to have been a recent explosion in links to Sudoku Variants in the references section and the Sudoku Variants section itself is starting to dominate the whole article. I feel strongly that either that Sudoku Variants be moved to a separate Wiki entry or there's a drastic pruning of this section so it doesn't dominate this article. Do others agree? --angusj 09:05, 14 September 2006 (UTC)

The Variants section itself seems of reasonable length, but the inclusion of a bullet list within it spooks me. Those things invariably seem to balloon out of control, and the bullet format itself seems to invite that. It's seems practically designed as a way of claiming notability for a topic that you don't then actually write anything about. -Stellmach 14:48, 14 September 2006 (UTC)
My take - if there is an inclination to expand upon a section of an article, working against a sense that the section might be getting too large in proportion to the overall article, then it might be a good time to break out that section into its own article. --Mwalimu59 18:11, 14 September 2006 (UTC)

Suggestion: Blindfold Sudoku http://www.nakedscience.com/memory/blindfold%20sudoku.htm

[edit] Terms Question

The article contains a reference to Nanpure and to "Nampure". I think one's just a typo? - 12:59, 17 September 2006 (UTC)

Since it's a Japanese term, these are two valid transliterations of the same thing, depending on whether you're transliterating literally or phonetically. The "n" character in Japanese takes on an "m"-like sound when it comes before a voiced or semi-voiced consonant such as "p" or "b". Best to pick one and stick with it. I think the "m" is more conventional, and besides, it makes more sense as the abbreviation for "number place" (which in Japanese would sound something like "nahmbu purase"). -Stellmach 13:30, 19 September 2006 (UTC)

I changed it at the top to "Nanpure", since that's the spelling on that seemed more common on a search. Not the most authoratative research method, but... MDWthing

The Manual of Style indicates that "Nanpure" is what should be used on Wikipedia, too, unless "Nampure" is more commonly used. Since it isn't, "Nanpure" is correct. (^_^) ···日本穣? · Talk to Nihonjoe 08:00, 4 October 2006 (UTC)

[edit] sudoku and copyright

Please include some comment on copyright restrictions with respect to sudoku. Many of my friends plan to poublish their nine-squares. I am sure that replications of previously published or posted squares will be extensive. Does Dell have a copyright on the concept or only on the name? Which name? Any other potential copyright problems?

This is an encyclopedia, not an online legal service. --angusj 14:27, 23 September 2006 (UTC)

[edit] definition of Sudoku in question

In the first sentence, Sudoku is refered to as the condensed version of the phrase of "suuji wa dokushin ni kagiru". When I refered this phrase to my father, (who was raised in Japan for the first twelve years of his life) he laughed at the phrase, saying that it is a rough translation of "numbers are for bachelors". Can someone please correct that portion of the article? Sr13 06:49, 4 October 2006 (UTC)

Well, considering the phrase is a bit of wordplay, it's not surprising he laughed at it. I clarified the translation to "digits must occur only once", which is more clear in English than the previous translation. Also, that part of the English article is almost an exact translation of the Japanese article, and it's likely the Japanese know what they're talking about. If your father was raised in Japan for the first twelve years of his life, it's likely there are phrases that have come into common usage of which he is unaware, especially since it's likely been at least 20 years or so since he lived in Japan (just guessing that you're in your teens or early 20's at the most given the comments on your user page). A lot has changed since the 1980s. ···日本穣? · Talk to Nihonjoe 07:58, 4 October 2006 (UTC)
My father works in the tourism business. I'm sure he is "updated" because how else could you speak with the younger audiences (20's, 30's) with them understanding you without knowing the updated language and grammar? Are you sure that the Japanese is translated correctly? Sr13 08:20, 4 October 2006 (UTC)
Yes, it's translated correctly. I consulted a couple phrase dictionaries and they both said it was correct. It's an idiom. ···日本穣? · Talk to Nihonjoe 08:38, 4 October 2006 (UTC)

So if it's an idiom, you cannot necessarily call the phrase a translation...you may be able to call this phrase an idiom of the word. Sr13 01:44, 5 October 2006 (UTC)

I've provided six sources stating the meaning of the phrase. ···日本穣? · Talk to Nihonjoe 07:11, 6 October 2006 (UTC)

And, if you really want to get into a discussion about the meaning of the phrase, here you go:

数字 (sūji, meaning numeral or number
は (subject particle, meaning that "sūji" is the subject of the phrase)
独身に (meaning "the state of being a bachelor", but it can also mean "single" or "the state of being single")
限る (meaning "to confine" or "to restrict"

Taken all together, it basically translates as "a number restricted to being single." As the Japanese are very fond of shortening phrases into something easier to say, they did it here, creating "sūdoku" (or more commonly, "sudoku"). If you enter the phrase on WWWJDIC, you'll receive the following:

数字は独身に限る 【すうじはどくしんにかぎる】 (exp) digits must occur only once (contracted to "sudoku"); ED

Does that help clarify it for you? ···日本穣? · Talk to Nihonjoe 07:24, 6 October 2006 (UTC)

Thanks for clarification. Sr13 00:02, 30 November 2006 (UTC)

[edit] Rapid-style definition and/or citation

Are there citations for "rapid-style"? (In "Computer Solutions", under "Solution Methods") Does rapid refer to efficiency of execution, or ease of writing? Wouldn't human-style techniques (set inference) often constitute optimizations for trial-and-error exhaustive search? A human could approach the problem by making lots of grids and starting a new copy every time a "hypothetical" is deemed necessary. True, a computer is better equipped for doing that rapidly and accurately. But you ought to gain in efficiency every time you eliminate the need for a particular exhaustive search. Of course, enumerating and evaluating all possible subsets of all regions to look for constraints on value (of a particular cell), or constraints on position (of a particular value) will (or, at least, could), when done by a computer, require a lot of iteration. Humans are better at noticing the patterns of constrained values and grouping them so as to imply further constraints. I'm just starting. I finally did what the local paper (David Green?--Oh yes, a search using the Google-brand search engine suggests he's "Mr. Sudoku", although he isn't mentioned in this article) called 5-stars (22 numbers given). After filling out obvious inferences, I still never needed any complex "what if" trial-and-error; I did it all with set inference (needed custom grids, though). Perhaps that week's puzzle was unusually easy in that respect, though. Is there, or should there be a separate item for computer techniques? (In addition to Mathematics of Sudoku ). However, IIUC, explaining in detail techniques for people themselves to use would be considered unencyclopedic. ("wikipedia is not a How-to Guide")--SportWagon 18:56, 27 October 2006 (UTC)

Thought about it a bit. The hypotheses (hypothetical solutions) will likely more rapidly generated than are ones which use more sophisticated (but time-consuming) techniques so as to be more likely to be correct. So each hypothesis will be more rapidly-generated. Whether that results in a overall win depends on the complexity of the problem.--SportWagon 17:49, 30 October 2006 (UTC)
But I go and read the article, and it doesn't really make that clear.--SportWagon 17:51, 30 October 2006 (UTC)

That said, the article already includes "Solution methods", doesn't it. But it doesn't mention set analysis. I'll review what the Math article says again, and see if something brief should be set here mentioning set analysis/inference.--SportWagon 19:03, 27 October 2006 (UTC)

[edit] Killer Sudoku

I've added a link to the 'killer sudoku' article. These are a particularly tricky (for me at least) type of sudoku that often start with no answers, but where each 'grouped' series is the sum of the numbers held within the lines. As with normal sudoku the numbers cannot be reperated on a row/column, within the 9-squares but also cannot be repeated within the same 'group'.

[edit] Sudoku in Japanese (数独 and/or カズオ)

The article on Sudoku spells Sudoku in Japanese as 数独 At least one other place, the Go! Sudoku page, spells it as カズオ I see that there are several Japanese writing systems, but are both really right? My guess is that that only 数独 is actually correct in Japanese. My guess is that the original Japanese word was in Kanji (数独) and that somebody converted it to Katakana (カズオ) as though it was an imported loan word. 24.35.112.221 01:20, 31 October 2006 (UTC)

カズオ is read Kazuo, rather than Sudoku. 58.185.72.54 05:27, 1 November 2006 (UTC)
The correct Hiragana for Sudoku is probably すうどく and the correct Katakana would therefore be スウドク, but I don't know which would actually be used (or neither, if it's always in Kanji). Kind of trivial, but I'd like to know which is correct, if anyone with more knowledge could assist on this. Miketsu (talk) 05:11, 31 May 2008 (UTC)
Incidentally, the Go! Sudoku page has been edited to clarify that Kazuo (カズオ), is the Japanese trade name for the Go! Sudoku game, because "Sudoku" is trademarked. Miketsu (talk) 05:20, 31 May 2008 (UTC)

[edit] What's the big deal about solving these things. It's easy.

I took a look at this craze with a calm head.

The first reaction was to believe the comments and do things the way everybody else did. This consumed time and brain cells which would be better consumed by alcohol.

My second reaction was to try this stuff. It turned out to be easy. Everybody was talking about methodology. Do this then that and you will get close enough to guess and confirm... Being of a math background I was replulsed by the childish focus on candidation elimination. Why not candidation elimination with unique candidate detection. This made each puzzle a no brainer. Both lines gave a solution with no guessing or complicated pairs...

If you want to get to this point then first consider each sudoku 9 interrelated puzzles. The first puzzle solves for 1, the second puzzle solves for 2, etc... If you solve for all possible 1-9 entries then the puzzle is complete. Since a unique solution is dictated by the puzzle then this will generate a unique solution every time. If a guess must be made then the puzzle has no unique solution.

And better yet, no matter how big the puzzle the solution can be found this way (deterministic) since the fact that enouch entries to guarantee a single solution will guarantee that all single entry puzzles can be solved.

The math isn't so easy and is dificult to describe here. But the results are easy to express. I solved six sudoku puzzles in less than one hour. I solved 2 hexadoku puzzles in about two hours. With the xls spreadsheets that I wrote based on these rules I finished six sudoku in 30 minutes and 2 hexadoku in 10 minutes (including the time to enter the initial puzzle).

Compete if you will but this lame puzzle isn't really something to compete with.

Al --Al 22:34, 7 November 2006

  • You need to try one after consuming some alcohol. Wahkeenah 23:51, 7 November 2006 (UTC)
  • His rambling almost sounds like he had. Yes, I find you can use what I call "set inference", but I can't do harder ones as fast as OP (Al) claims. I alternate between bouts of "set inference" to eliminate candidates, and elimination of possible positions. Right now my biggest problem is I sometimes make transcription errors which cause a lot of incorrect followups. If I don't get more reliable at doing them, I'll give up. But if I get reliable, then perhaps future ones will be a waste of time? So then I should stop, anyway. I'm interested to hear someone did make a spreadsheet to help him with it. I would guess it does many of the set operations to eliminate potential candidates. My intuition also says you never truly need to do "what if"; if you do, you're overlooking an existing restrictive inference. The puzzles sit on the outer limit of what lots of people can comprehend, like Rubik's Cube, but Sudoku does not have the tactile sensations of a Rubik's Cube. www.sudoku-grok.com eliminates some of the bookkeepping and allows me to do puzzles faster than on paper, and even frees my mind to find a few less-obvious inference techniques.--SportWagon 19:17, 8 November 2006 (UTC)
  • The article mentions "(p,q) and (p,q)" and "(p,q,r) (p,q,r) (p,q,r)" for a set of three, but doesn't mention other variants such as "(p,q) (p,r) and (q,r)" which also remove candidates p,q,r from all other cells in the region. I.e. it doesn't generalize to set inference and cardinality in general. But if we put too much Math (set theory) into the article a) many people won't understand it, and b) someone will come along and describe how they manage to do it entirely with intuition without all that theory. (Maybe).--SportWagon 19:17, 8 November 2006 (UTC)
    I also has the intuition that set inference techniques could solve these puzzles. I started working it out, but never got to the point of creating a spreadsheet (although I was headed in that direction.) I for one would be very interested to see an explanation of this solving technique added to Mathematics of Sudoku. —Doug Bell talk•contrib 19:58, 8 November 2006 (UTC)
    Yes, I noticed that such was entirely missing from Mathematics of Sudoku and wondered if it was deliberate. That article deals more with "enumeration", and not at all with solving techniques. I already (27 Oct 2006) added a suggestion to Talk:Mathematics of Sudoku, but received no responses. I guess no-one's actively maintaining that article anymore? I thought one or both of these articles was in danger of being "too long", but can't find evidence of that any more? Oh, and there's always the "no original work" bugaboo. I.e. wikipedia rules require we find the work second-hand, rather than work it out ourselves.--SportWagon 20:56, 8 November 2006 (UTC)
  • I am sorry if my critic hurts anybody's feelings, but:

1. What does Al say about speedy solutions ? I don't believe a single word of it (Why don't you come out of the shadows Al and give us a userpage to argue in ?) I've been addicted to sudoku for some months now and only try to solve the advanced versions, without algorithms and spreadsheets, which can take me an hour, or maybe more.

2. I think the article description on how to solve the puzzle is too confusing, but I could not understand all of it. My old boss (Ivor Ireland, God bless his soul) used to say 'You have to be a Philadelphia lawyer to understand this! ...and the writer must be one of them. --- Adding to what I said about solving the puzzle under 1. I do not use any subscripts or dots, and write in ballpen, till I arrive at a 'what if' case.

3. I agree with the notion of Sportwagon that the 'what if' situation should not arrive too often. For this situation I have an efficient way, and with trial and error you will solve any sudoku problem, but you can argue that using this loses some of the logic element, which is what the puzzle is all about. I admit it is not easy to write down a concise but simple description on 'how to solve it?' I might make an attempt, but it will take some time. I'll be back.LouisBB 17:46, 29 November 2006 (UTC)


FWIW, I use pencil, but only occasionally the eraser (except when I blow it). I do the King Features 7 weekly puzzles, which are really easy on Monday and Tuesday, getting more difficult through the week. I usually end up marking a few contingencies (a lot, later in the week), and have not yet found a case where the set properties of the contingencies were not sufficient to make the necessary inferences. (But an issue here is that, to some people, some of those might be "what-if"). For the difficult puzzles, especially if I blow it, I resort to some "cheat sheets" which have extra space for each cell, including a markable grid of candidates. The Saturday and Sunday puzzles, and sometimes even Thursday or Friday, often require a lot of contingencies marked before I can even make any inference (occasionally that happens after an initial round of easy inferences).--SportWagon 18:12, 29 November 2006 (UTC)
Al again. Please look at the header on www.harkcom.us and you will find a spreadsheet which proves my statement. All sudoku, if uniquely solvable can be solved simple by this method. Sorry to rain on the parade. The spreadsheet deterministically solves 4x4 (hexadoku) puzzles. I've solved 12 puzzles in less than an hour for 3x3 puzzles using books I bought in airports. And if any body is interested I write operating systems for microcontrollers. I would like to find others who do so. So far I haven't found anyone. —Preceding unsigned comment added by 84.160.125.2 (talk • contribs)

Al is absolutely right. His technique is known as Autism. 87.187.26.173 (talk) 00:09, 25 May 2008 (UTC) S. Heinrich

[edit] Not solved?

A quick search seems to indicate the problem has not been solved, but there were a number of interesting items related to this:

  1. This claim: "Sudoku, by the way, IS math. It's also an NP-Complete problem, which basically means that a simple way to solve them will probably never be found (because if it IS found, it would actually revolutionize much more than solving Sudokus, and solve one of the Millenium Problems)" [3]
  2. Sudoku is a 6-dimensional problem
  3. an SQL solver based (at least partially) on set theory
  4. (PDF paper) Nonrepetitive Paths and Cycles in Graphs with Application to Sudoku
  5. Good explanation of solving discussions employ set theory to explain known solving techniques
  6. (PDF) an enumeration of known solving techniques expressed as set theory equations
  7. an interesting discussion on the subject

—Doug Bell talk•contrib 21:52, 8 November 2006 (UTC)


1. is rubbish. Alright, an efficient solution for N×N cannot be found because the complexity and time can be demonstrated to be more than a polynomial function of N. But N=9 is a pretty small N, and solutions can be found using exhaustive search, with elementary pruning. That is, just because an algorithm is exponential doesn't mean it is not practical for small values of N. The 4point-webdesign.de PDF looks complete, but I'd need to sit down with it for a long time to begin really understanding it. A related problem would be, "can it be proven that some (set inference) techniques can be omitted from a solver's toolkit, and still allow them to solve without trial-and-error?" (Some such proofs would be trivial if some techniques can be demonstrated to be necessary implications of other techniques).--SportWagon 23:46, 8 November 2006 (UTC)

There is a difference between solving for all solutions (0,1 or many) for an arbitrary partially-filled grid, and solving a puzzle known to have one, and only one, solution. When there are multiple solutions, you will get to a point where restrictions do not force your choice. You will need to enumerate all remaining non-restricted possibilities. (Observing, of course, restrictions created by your arbitrary choices). So we get back to the question of whether it can be proven than a set-inference based algorithm (working with sets to limit candidate values for positions, as well as candidate positions for values) will ever get to a point where the puzzle is not complete, but no further inferences can be made.--SportWagon 23:55, 8 November 2006 (UTC)

[edit] Source for three basic conflicts

http://www.setbb.com/phpbb/viewtopic.php?t=1169&mforum=sudoku

Now I understand what the contributor was saying. He was describing how you determine a bad candidate in a "what-if" method. Based on the above phpbb discussion, I probably will delete the entire section after all.--SportWagon 20:05, 10 November 2006 (UTC)

Alright, someone else can apply the axe. But I don't think the 3-part description will be clear to the average reader. "fish" especially confuses me, but perhaps only because of the odd name. The method described at the url, and implied here, is an odd combination of "what-if" and "candidate-elimination", suitable for computer solving. Perhaps the section should be moved to the Computer solutions section. Maybe?--SportWagon 20:13, 10 November 2006 (UTC)

I never made it clear, I suppose, that my assumption is the poster at the above BBS is the same as the original contributor of the "three conflicts". (If not, it's minor plagiarism). I.e. I was not indicating that the link was an authorative source; just that editors should look at if for extra information.--SportWagon 18:02, 29 November 2006 (UTC)

[edit] electronic handheld versions?

Does anyone know of a decent electronic handheld version of Sudko?

Would it be worth having a list of products, with desired features etc, e.g. electronic handheld, board games, books on theory ... probably asking for trouble I guess !! --Quantockgoblin 23:23, 4 December 2006 (UTC)

I own a handheld version of sudoku, and I have to say it's not bad. The interface is a little difficult to work with, but all the puzzles are sound (if not symmetrical), and the system claims to contain many billions of puzzles. I have not completed nearly enough of them to know if this is true, and the game keeps no score, but the game might be worth mentioning. I will try to find more information on it. Bio 21:13, 9 January 2007 (UTC)

I love the handheld version I found called: Pocket Sudoku. Author is Emil Andersson. Neverlostever 21:34, 13 July 2007 (UTC)

[edit] compact algorithm to solve jigsaw sudokus

I started writing the section on the basic algorithm for jigsaw sudokus, then discovered that it might not even be appropriate for this page. The Perl code for the algorithm is 86 lines long and may be found here: [sudoku solver]. Would it be appropriate to add this link to the external links, which I guess is strongly discouraged? Or should we include a suitably formatted version of the Perl code on the page? Or is this level of detail simply not appropriate for Wikipedia? Thanks. -Zahlentheorie 15:47, 13 December 2006 (UTC)

[edit] Is there a maximum to the number of sudoku puzzles that can be made?

i mean its only a 9x9 grid, the options have to run out at some point right? Bliks 01:30, 3 January 2007 (UTC)

Sure. Moreover, there's "only" 5,472,730,538 essentially-different completed Sudoku grids. — Matt Crypto 01:46, 3 January 2007 (UTC)
Thanks, I was just curious about that one. Should that be in the article? Bliks 22:18, 3 January 2007 (UTC)
It's in "Mathematics of Sudoku". — Matt Crypto 23:11, 3 January 2007 (UTC)

In the mathematics of sudoku section there is a really large number describing the number of sudoku problems. Should this number be reduced to scientific notation (it contains 22 digits)? Bio 21:15, 9 January 2007 (UTC)

No, because people are interested in knowing the exact number. McKay 03:07, 10 January 2007 (UTC)
Ok, thanks, it will reamain as it is. Bio 18:15, 10 January 2007 (UTC)

[edit] Need a citation

Concerning the total number of 9x9 solutions, we write "Both results have been confirmed by independent authors." Can anyone identify an example of such an independent verification? Thanks. McKay 07:06, 3 January 2007 (UTC)

[edit] I don't see any links to playable online sudokus

Surely at least one link (at the bottom of tha page) to an online sudoku, would be useful? I find this flash version to be pretty good: http://www.sudokuplayer.net As it is a fairly accessible version (made in flash) and it generates random puzzles every time. It also includes difficulty settings to appeal to a large age range. No I am not associated with this one. I just play a lot of online sudokus and I think this is the best, if you want to put a link on the wiki. —The preceding unsigned comment was added by 84.9.45.31 (talk) 12:46, 14 January 2007 (UTC).

I agree, a link should be added. Bio 18:52, 1 March 2007 (UTC)
This has been discussed (and tried) many times in the past and it doesn't work. It simply starts a free-for-all where every tom, dick & harry adds their own favorite links (and often I suspect to their own personal websites). --angusj 09:04, 3 March 2007 (UTC)

[edit] Howard Garns

Ed Pegg Jr: The entry for Howard Garns has been removed from Wikipedia. Information about his life was just starting to become known, so I strongly disagree with removing his entry. As I noted in the Howard Garns discussion (before his name was changed to a redirect), the March 2006 issue of Indianapolis Monthly has an extensive article on Howard Garns.

[edit] Popularity in the media

I propose that this section be removed altogether. --angusj 08:37, 26 January 2007 (UTC)

It seems like a helpful section. It shows how popular Sukodu is growing, expanding beyond paper and pen(cil) into electronics. In today's society, young children are more likely to be drawn to electronics than written work, so this section seems to be appropriate. I say we keep it. Bio 19:28, 13 February 2007 (UTC)
Fair points except that the section is entirely Anglo-centric. --angusj 21:23, 14 February 2007 (UTC)

[edit] Excel spreadsheet to solve SUDOKU 9x9

I've read about solvers and algorithms, and some people have expressed interest in an excel solver. I do have developed a very simple excel solver that can handle 9x9 SUDOKUS. It is 100% automatic when the puzzle does not need trial-n-error. I've found that 90% of puzzles are straight forward. The tool allows to analize alternative solutions when guessing is needed, so one can easily solve even sudokus with unique solution that require guessing. I am also using that tool to develope a ranking scheme that is pretty much objective and accurate to describe expected dificulty to solve the puzzle on hand. Any one interested in having a copy, raking a particualr sudoku against the ones I have or get a copy of the most dificult one I have found so far just let me know (is email publication allowed here?). --Alvaro Ledesma 22:58, 4 April 2007 (UTC)

[edit] Evidence of notability for Brain Age over all other Sudoku games

Quality - According to Game Rankings, Brain Age is the highest-rated game featuring Sudoku on consoles and handhelds. The only games that beat it are not only only on mobile or PC, but also feature only one or two reviews compared to 62 reviews for Brain Age.

Sales - 6.5 million - something like 1.3 million in the US, 2.1 million in Europe, 3.1 million in Japan, and >500,000 in other territories, significantly more than any other Sudoku game for consoles or handhelds.

Inspiration - 2005 and earlier, no console or handheld featured a Sudoku game. In fact, there were only like four Sudoku games released in 2005. However, in 2006, when Brain Age became popular, there was an explosion of Sudoku games, especially for handhelds. - A Link to the Past (talk) 23:14, 4 April 2007 (UTC)

OK, I removed the reference to this application simply because it sets a precedent, thus opening the door to everyone else to add references to other applications. This has been an major ongoing problem for this article - adjudicating which applications are 'worthy' and which aren't, and it got totally out of hand. The only viable solution has been to avoid reference to any specific applications (with the exception being Gould's original). --angusj 23:27, 4 April 2007 (UTC)
But that is not a good reason to leave content out. Just because listing the by-far most popular Sudoku video game ever made may cause people to list another version doesn't mean that the most-popular Sudoku video game ever made shouldn't be mentioned. - A Link to the Past (talk) 23:33, 4 April 2007 (UTC)
I'm afraid it seems to be the only workable solution. (I presume you aren't aware of what was happening here about 12 months ago - what was perceived as the 'best' or 'most popular' was totally subjective and abused, and it became a free-for-all mess.) --angusj 03:38, 5 April 2007 (UTC)
Maintain the article. Don't censor it. - A Link to the Past (talk) 03:40, 5 April 2007 (UTC)
I'm afraid I'm going to have to side with ALttP here. This is a popular article, people will fix it when something bad comes in. If there is good reason to have Brain Age in it, then have it. It is better to have to edit bad things out than not be able to put good things in. Sharpevil 19:10, 5 April 2007 (UTC)
I like how you phrase that - "I'm afraid". Does siding with me frighten you?
p - A Link to the Past (talk) 19
46, 5 April 2007 (UTC)
I can't say I've contributed to this article, but I came over here as Link to the Past invited discussion from the CVG Talk page. I do believe that Brain Age is one of the more popular items to feature Sukoku, thanks to the success of the DS, and as such deserves a mention, though briefly. How does this sentence sound: "Versions for video game handhelds such as the PlayStation Portable and Nintendo DS have also been released, notably its inclusion in the popular Brain Age software." It wasn't Brain Age that popularized Sudoku, but its worth a mention if we are addressing Sudoku in videogames. -Digiwrld1 23:05, 5 April 2007 (UTC)

I don't see how Brain Training is worth mentioning over any other Sudoku game. The Sudoku element of the game was just tacked on for the English language release because Sudoku-spacker-fever was in full flow in the West. Brain Training is not primarily a Sudoku game, that's just a bonus round, I doubt you'll find that the "quality" of the Sudoku puzzles were rated over fully fledge Sudoku games, nor is there the range of puzzles, nor random generation. Sure, the overall quality of the game is superior to Sudoku number crunchers, but the Sudoku element itself is not. "Inspiration" is misleading, being that Sudoku on handhelds wasn't Brain Training's idea, and it's pretty certain that all those 2006 titles would have come out anyway. Just searching on Gamespot for Sudoku games brings up quite a lot of ones released before Brain Training. - hahnchen 01:37, 6 April 2007 (UTC)

I looked at GS' list, and it does not contradict what I say - I stated that the only Sudoku games before Brain Age were PC and mobile phone Sudoku games. The only time a Sudoku game was released for any console or handheld video game console was after Brain Age. Additionally, looking at reviews, while Sudoku was not present in the Japanese version, it is one of the more well-liked aspects of Brain Age, according to both IGN and GameSpot. I believe a combination of the highest quality video game version of Sudoku created so far combined with the best-selling Sudoku game (and don't say that it's not a Sudoku game - Sudoku is a heavily advertised feature of Brain Age; it's practically a part of the title almost). Regardless, looking at all other titles, I do not see random generation either. Looking at every video game Sudoku game, I do not see any other Sudoku game even doing near 1 million copies. Brain Age is, in particular, the only video game featuring Sudoku to have sold 1 million copies (or for that matter, 2 million, 3 million, 4 million, 5 million, or 6 million). It is consistently selling at a high rank in Germany, the United States, UK, France, Spain, and Australia. Anyone purchasing Brain Age knows that it comes with Sudoku. It is a significant feature of Brain Age. I'm really at a loss for words at the very idea that Brain Age's inclusion of Sudoku is comparable in notability to any other Sudoku video game in existence. Sudoku is often attributed as a primary feature of Brain Age, not a minor side feature. - A Link to the Past (talk) 03:53, 6 April 2007 (UTC)
No, Braing Training was not designed from the ground up to be a Sudoku game. A few hundred puzzles were included in the Western release to ride upon the wave of Sudoku madness. Have you played Brain Training? You actually choose whether you want to play Sudoku or Brain Training at the beginning, it's not actually part of the main game in that sense. The reason it was heavily advertised was that 2006 was the year of Sudoku, Brain Training didn't make it so, it joined in. I don't see how you believe that Brain Training "inspired" other developers either, Go! Sudoku was released in 2005, Sudoku Fever was released in March 2006, Dr. Sudoku was April 2006 (a few days after Brain Training). Of course, this is from a European perspective, 2006 every lacky was onto the pervasive Sudoku by the time Brain Training came out, maybe it introduced Americans to the pathetic number game, I don't know. - hahnchen 10:16, 6 April 2007 (UTC)
Here's the problem - when Brain Age/Training is assessed in reviews, Sudoku is mostly treated as no less than just-as-important. You may consider it a side-mode, but sites like IGN and GameSpot consider it as one of the high-points of the game. - A Link to the Past (talk) 18:49, 6 April 2007 (UTC)
In all due respect, I see nothing to even imply that any Sudoku-exclusive video game is more well-known than the Sudoku mode of Brain Age. The fact that this lesser mode is so much more well-known than any existing Sudoku video game is notable. It's not like it's irrelevant trivia. - A Link to the Past (talk) 22:06, 9 April 2007 (UTC)
As I've said repeatably above, it isn't about which program is the most well known or whatever for any given media. From frustrating experience, it's evident that we'll never reach consensus over which programs should be worthy of mention here and we end up in either endless argument or tolerating an ever expanding lists of 'worthy' programs. The only workable alternative has been to avoid mentioning specific programs. --angusj 22:58, 9 April 2007 (UTC)
Yes, but that is not a good argument to censor Wikipedia. Brain Age is an award winning game, a multi-million selling title, and the Sudoku aspect of it is leaps and bounds more well-known than any other Sudoku video game. I'm sorry that you do not wish to maintain the article, but that's not a good reason to leave content out. I get the feeling that you are more leaving it out because you don't wish to maintain the article, not that it is unmaintainable. If someone adds another game, remove it. Tell the user to open discussions. You may as well protect the article if you don't want people to edit it. A slippery slope is NOT a good argument - under your logic, creating an article on Mario would lead to Luigi, and then Princess Peach, and then Bowser, and then Donkey Kong, and then Wario, and then Yoshi, and on and on hypothetically. I don't see Brain Age's Sudoku to even be comparable to Sudoku-exclusive games (that is, it is much more well-known and much greater, having more reviews than any other video game product to ever feature Sudoku) or that Brain Age is anywhere near any other games, failing a need to keep it off of this article because Sudoku video games of very little importance in comparison will be added if a highly popular and well-covered award-winning highly successful game is mentioned. - A Link to the Past (talk) 23:11, 9 April 2007 (UTC)

I've removed your link again. For more info see discussions for this article (in the archives) around July 2005. --angusj 23:54, 9 April 2007 (UTC)

Wow, what a wonderful revert. Could you bother to back it up with policy to show that if an action could theoretically lead to lowered quality of an article, it should not be allowed? Let's keep Wikipedia away from arbitrary and illogical censorship, please. You may want a worse Wikipedia, but I sure don't. - A Link to the Past (talk) 00:23, 10 April 2007 (UTC)
Also, not a single discussion on the Sudoku talk page talks about Brain Age or even video games, archived or otherwise. - A Link to the Past (talk) 00:26, 10 April 2007 (UTC)
Oh, no, I see - it was a discussion about the section and didn't even hint at video games being involved. Since I assume you knew about the discussion, it doesn't make me feel satisfied knowing someone misled me into thinking that there actually was a group of people who decide to not include certain video games, when it seems that the only thing discussed was a big laundry list of newspapers featuring Sudoku. - A Link to the Past (talk) 00:32, 10 April 2007 (UTC)

[edit] One External Link???

I cannot believe that there is only one ext link for sudoku. I found some of the strategy sections very hard to follow (with not nearly enough examples) that I wanted to read some outside sources, yet there are none. Not to mention some places that offer free game generators. I find the arguement about "every Tom, Dick & Harry" to be one of those condescending editor attitudes that far too many take. It's the old "well, if you aren't going to get along then I'm goint to take ALL the toys away" bullshit parents pull when they simple don't want to deal with a situation.

I suggest you DO let every Tom, Dick & Harry post links and you weed out them out if the section starts getting unwieldy. Editors can check them out and if it goes over a certain number, like say 5 - 10, then any new links should be debated and replace older ones if there is a consensus if they are judged better. You know, the way wikipedia is supposed to function. RoyBatty42 23:52, 7 May 2007 (UTC)

For the way Wikipedia is supposed to function regarding external links, see Wikipedia:External links. Free game generators are NOT in any way, shape or form encyclopedic. Wikipedia is not a web directory. It takes all of two seconds for you to go to Google and search for Sudoku. DreamGuy 18:40, 8 May 2007 (UTC)
And, as far as that goes, the ONE external link you complain about is a link to LOTS of links. All the Sudoku sites you could ever want there. So even whatever small complaint you might have had about the way things are done here doesn't even work. DreamGuy 18:57, 8 May 2007 (UTC)
External links do not have to be encyclopedic. They should contain content that aren't encyclopedic, actually. If they were encyclopedic, the info should be incorporated into the article and the link should be cited as a reference. --59.61.95.18 07:59, 23 July 2007 (UTC)
They have to serve an encyclopedic purpose. Please actual go read the policy. DreamGuy 22:14, 23 July 2007 (UTC)

[edit] Jim Wright's Review Comments

Under Strategy I suggest delete the last 3 words from this sentence "For fastest results, the numerals are scanned in order of their frequency, in sequential order." --Jimwrightbe 08:28, 21 May 2007 (UTC)

[edit] Exceptionally Difficult Sudokus

I feel that an encyclopedic article about Sudoku would be woefully deficient if it did not have some description of how difficult they can be. From discussion above, some people would think that they are all easy if one just learns the technique. In fact, some are near impossible for humans to solve without computer assistance. There even seems to be a class of Sudokus which are designed to challenge programmers. Many Sudoku handheld toys and programs cannot even solve all Sudokus, unless the program reverts to a long exhaustive search or trial and error.

I did place this section at the very end of the article, but just before "construction", and "variants". These hard puzzles are in fact sudoku puzzles (not variations) so any later placement would not flow well.

For some references to see where people "play' with these really hard puzzles (such as Top 1465 #77) check these links: http://www.setbb.com/phpbb/viewtopic.php?t=103&start=150&sid=a675c04ac6e72a7584fbdc2d24253285&mforum=sudoku http://www.setbb.com/phpbb/viewtopic.php?mforum=sudoku&p=7760 http://magictour.free.fr/top1465

AMcDermot 21:29, 27 May 2007 (UTC)

This level of detail seems trivial. And it's unsourced original research, which is a big non no in an encyclopedia. DreamGuy 13:27, 28 May 2007 (UTC)

[edit] subtle nationalism?

Am I the only one who sees an attempt at nationalism in the history section?

Let's get this straight:

Euler was pretty multinational, Swiss / Russian / German
The French Connection seems dubious and sounds like an attempt to claim priority
Garnes was probably unfamiliar with the French stuff, and with Euler, at least from reading between the lines on his article.
Nikoli certainly is Japanese, but
Gould is from NZ but he was a HK judge I think prior to the hand-over to PRC
I also think Gould was the one who made it symmetric. If this isn't correct I'm sorry
The Times is certainly British.

There's a lot of countries here. I think too much weight is on the French connection, which is the weakest link.

Thoughts? --203.117.92.2 08:32, 14 June 2007 (UTC)

You probably have a point here, but nationalistic bias can be one of the more tricky of POVs to sort through. DreamGuy 19:38, 16 June 2007 (UTC)

Guys, for sure nationalisms are key to content quality and localization of terms, idioms, etc. For me the one Sudoku network that has made it THAT clear is Sudoku Entertainment with 12 countries (local domains and local texts) a slight perhaps annoying detail is that the puzzle it self (in some countries like italy and netherlands) or in english, but the rest, chapeau! [www.sudokuENT.com]SudokuEnt.com this is their blog with articles and the list of webs. enought said

enough said, eh? Er, I think not. Can't quite understand what you are trying to say at all. No offense meant, but this sounds like something written by Mark V. Shaney. —Preceding unsigned comment added by 66.102.196.32 (talk) 12:18, 21 September 2007 (UTC)

[edit] The Hidden Logic of Sudoku reference

I thought this article interesting and useful until I got to the section: Generalised symmetries and extended Sudoku board

It seems to be an advertisement for the referenced book and/or the referenced program (SudoRules). It also starts using terminlogy not previously used in the article. (conjugacy links, bivalue cells, ...) Since I haven't read the book, nor do I understand all of what the summary is saying, I can't really edit the article.

I think the section should be dropped, and if not should be integrated better with the rest of the article. Below is the paragraph I think should be dropped.

In "The Hidden Logic of Sudoku"[13], a book based on a systematic logical formalisation of the game, all its generalised symmetries have been elicited, in particular between rows and numbers and between columns and numbers. A new resolution method has been developed, based on their systematic exploitation. As a result, an extended Sudoku board has been designed, which makes conjugacy links appear as bivalue cells and may ease the application of the method. Thus, Hidden subsets and fishy patterns (X-Wings, Swordfish and Jellyfish) all appear as Naked Pairs, Triplets or Quads. Within a general framework for dealing with chains, these symmetries are used to introduce new resolution rules, such as hidden xy-chains. This method has been implemented in the SudoRules solver, based on Artificial Intelligence techniques and simulating a human player

10 minutes later: forgot to sign the post, so am doing it now ... Neverlostever 21:35, 13 July 2007 (UTC)

[edit] The Hyper-Sudoku example has no unique solution

I found this:

5 7 9 1 8 2 4 6 3
8 6 4 9 3 1 2
1 3 2 4 6 8 9
9 1 8 2 3 4 6
6 5 3 4 9 8 7 1 2
4 2 7 3 6 1 5 8 9
3 9 1 6 5 4 2 7 8
2 4 5 8 7 9 6 3 1
7 8 6 2 1 3 9 5 4

and then what's left can be filled with 5s and 7s in two different ways.

Time to go and find (or construct) one that really works.... -- Smjg 11:59, 28 July 2007 (UTC)

It was originally added by some guy trying to spam his own personal website for these, so I guess it doesn't surprise me it's wrong. I don't even know if we need an example for this type, but certainly a wrong one is no good. It's sad that so many of the sites offering logic puzzles of various types are made by people who don't understand the rules for them. DreamGuy 20:04, 28 July 2007 (UTC)
Indeed. Worse still is that these broken puzzles sometimes find their way into commercial puzzle books.
Still, we can do better. It would help if, before people add examples of puzzles, whether self-created or cribbed from some other source, they would check for themselves that the puzzle is valid. Preferably using a computer program - it could indeed be that the person who added this one did check it by hand, complete with imperfections. -- Smjg 17:47, 29 July 2007 (UTC)
I added a new valid hypersudoku example. Crafted by myself, and no external references; uploaded to Wikimedia Commons. Hope it does not violate WP:NOR. No hurt feelings if it is not wanted, so feel free to take it away. Ideally it should have less than 17 clues (the puzzle example has 18), and preferably have some symmetry (this example has not). Oceanh 06:38, 31 July 2007 (UTC)

[edit] Image:Sudoku-by-L2G-20050714 solution.svg

Can someone modify the solution image Image:Sudoku-by-L2G-20050714 solution.svg so that answers are also circled in addition to being in red? That would be good for people with color blindness. See Wikipedia:Accessibility for more info. --Jtalledo (talk) 00:04, 6 August 2007 (UTC)

[edit] Why is Sudoku listed in italics?

Why don't put puzzle names in italics. Italics are only used for book titles, film titles, etc. and instances where the word is not being used for what it means but to highlight the word. Was this a recent change or did someone go through and do this? DreamGuy 18:21, 21 August 2007 (UTC)

It's better than having it bolded everywhere. Otisbum 22:37, 5 September 2007 (UTC)

[edit] Scanning

The Cross-hatching subsection of the Sudoku#Scanning section contains:

For fastest results, the numerals are scanned in order of their frequency, in sequential order. It is important to perform this process systematically, checking all of the digits 1–9.

I've bolded the part that I don't understand - help! - hydnjo talk 15:23, 12 October 2007 (UTC)

Thank you for pointing at this. The scanning order is irrelevant, and as it stands, there is also a conflicting "order of their frequency" versus "in sequential order". I removed the sentence. Oceanh 05:44, 21 October 2007 (UTC).

[edit] Colour Sudoku

I think "Colour Sudoku" [4] should be mentioned in the article, at least in the "External links" section. Giftlite (talk) 16:08, 18 December 2007 (UTC)

[edit] Sudoku in the Baltimore Sun (and other Sun papers?)

In the Sunday section, they publish a variation that they call "High Fives" that consists of five standard puzzles, one in the center and four that each share a corner 3x3 block (for example, the upper-left corner of the center block is also the lower-left block of one puzzle). I can post a scan of one, if it is felt that copyright would not be a problem (fair use?).

For their regular weekday puzzle, they have also gone to using a 1 to 4 numbering system and in general, much harder puzzles (4's almost always require guessing, which can be very frustrating). I'm not entirely sure that this rating is consistent though, short of using a computer solver, it is certainly possible that I miss non-guess avenues that would eventually lead to a solution.

One of the things that I like about the High Fives is that I have never found one that required guessing. Dfmclean (talk) 19:33, 11 February 2008 (UTC)

[edit] UK papers

All the stuff about how the British papers fought with each other to over Sudoku has been removed. Jooler (talk) 00:12, 28 March 2008 (UTC)

[edit] References?

I can see the numbers but no references at the bottom. Clicking on a 3 for example takes you nowhere. Penguinwithin (talk) 03:34, 8 April 2008 (UTC)

[edit] Hardest sudokus

I've rewritten the Exceptionally difficult sudokus (hardest sudokus) section on the algorithmics_of_sudoku page. Is it something that should be included here in some form ??!! SudokuBoss (talk) 09:14, 18 April 2008 (UTC)

How about just mentioning the Algorithmics_of_sudoku page? (in an x-ref).
Just my 0.02, Mike Schwartz (talk) 23:30, 9 May 2008 (UTC)

[edit] mis-placed "[edit]" links

The "[edit]" links for these 3 sections:

=== Marking up ===
==== Subscript notation ====
==== Dot notation ====

seem to be all shown together, (at the bottom of the "Dot notation" section) (in the article). On a hunch, I made some edits [to the article], but -- I did not know what I was doing, apparently -- and (hence) those edits of mine do not seem to have fixed anything! (feel free to revert those edits [to the article] if you see fit). I may give up - for now - trying to figure this out. If you can explain it, - fine! (and/or, just fix it). Thanks, Mike Schwartz (talk) 23:30, 9 May 2008 (UTC)

<nowiki>..</nowiki> tags added, (to this section on Talk page) Mike Schwartz (talk) 22:42, 28 May 2008 (UTC)

[edit] Set of Procedures?

The article might be given greater coherence if it was reorganized by listing a set of procedures for finding single options for entry. The scanning procedures mentioned do not appear to be complete. Acquiring the necessary skills is a part of the puzzle's learning experience. --Jbergquist (talk) 22:11, 11 May 2008 (UTC)

[edit] Sudopedia

There is a Sudoku wiki at sudopedia.org[5] - could that be added as a link as it has the most complete list of solving techniques, terms, guides, software, websites, etc. than any other site I have seen - plus its GNU, powered by MediaWiki and named "Sudopedia". I submitted it to dmoz.org a couple of weeks ago but it still hasn't been added. Note, this site was created by the owner of SudoCue.net[6] which as a free SudoCue program and the best solving guide I have ever seen. I was going to add all of the terms from this site to List of Sudoku terms and jargon, but I think it might be too much for a wikipedia article - will add discussion on this at that article too. As for the "External Links" section, I understand that it should be minimized, but I think the Sudopedia site could be an exeption. Do you Agree? If no objections in a week or two, I will use the bold method and add the site then see what happens. --Lefton4ya (talk) 19:16, 23 May 2008 (UTC)

[edit] External link?

I have recently made a forum about sudoku and it seems that there are little ways to get it across to people. I was wondering if there was any chance that I could put it into this page. I am sure that this will do verry little to this page but increase my members greatly. Here check the site out My Forum --Scad 19 (talk) 08:18, 31 May 2008 (UTC)