Talk:Cantor's first set theory article/GA1

RESULT

Article failed 29 January 2014; major issues not addressed

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

GA Review[edit]

Article (edit | visual edit | history) · Article talk (edit | history) · Watch

Reviewer: Spinningspark (talk · contribs) 21:31, 3 December 2014 (UTC)[reply]

Hi Michael, I just saw this fascinating subject in passing, so I 'm going to review it, even though I don't usually review maths articles. I'm going to be busy tomorrow so might not be able to do a full review for a few days, but one thing jumps out at me straight away so I'll mention that now.

Is there really a controversy over the constructiveness of the proof, or is it merely two groups of mathematicians talking at cross purposes not understanding that the other is talking about a different proof? That would seem to be the case, but is far from clear in the lead. It strikes me that undue emphasis is being given to this when it amounts only to a mere misunderstanding of the other. Unless of course there really has been a decades long dispute with neither side ever realising that they were not talking about the same proof.

One minor point, MOS:HEAD says that headings should not contain questions. Although that is not actually something covered by the GA criteria. SpinningSpark 21:31, 3 December 2014 (UTC)[reply]

Review[edit]

I have already raised the issue of whether the so-called controversies discussed in the article are genuinely controversial points. Looking through a few previews in gbooks, it is obvious that Cantor's work was indeed controversial, but more on the question of whether it was valid to study transfinite sets at all rather than the issues raised in our article. I may be completely wrong about this (not my subject) but that's the way it comes across to me. If so, there is a big hole in the article coverage which is a failure of GA criterion 3a.

Lead

See my initial comment on the nature of the controversy
Wikilinking part of the bolded part of the lead sentence is discouraged by WP:LEAD.
Leaving two unanswered questions is not a sufficient summary of the article content on them
There is a lot in the article about the history of the idea (whole section on development) but absent from the lead.

The article

"Note how Cantor's second theorem..." There is a WP:WTW issue here.
"...Cantor's second theorem separates the constructive content of his work from the proof by contradiction..." I think it needs stating explicitly that the second theorem is the constructive content (if indeed that is the case)
Transcendental number should be wikilinked. It is linked later in the article, but that is not the first mention.
The paragraph beginning "The first half of this remark..." is uncited. At the very least it needs a cite for "Cantor...probably" which is making a theory of mind statement about Cantor's motives.

The proofs

"interior of the interval" requires a definition (in running text, by gloss, or by wikilink) as it has a specific technical meaning here.
"Since at most one xn can belong to the interior of [aN, bN], any number belonging to the interior of [aN, bN] besides xn is not contained in the given sequence." There seems to be an implication here that one can prove that there is a number other than xn in the interior, or am I missing something? In any case, this part of the proof does not seem to have been brought to a conclusion.
"Cantor observes that the sequence of real algebraic numbers falls into the first case..." There is an ambiguity here over which case is being discussed. There is a first case and second case of finite or infinite intervals, and then the second case has a first case and second case.
"...thus indicating how his proof handles this particular sequence". Not particulary clear what this is saying.

Is Cantor's proof of the existence of transcendentals constructive or non-constructive?

Headings should not contain questions per MOS
See my initial comment on the nature of the controversy
One has to do a great deal of reading between the lines, or going back to earlier in the article to get the basics of which proof is being discussed here: which proof is the subject of this article, whether the 1874 proof is a synonym for the subject of this article, whether the subject of this article is a constructive proof or not, and whether mathematicians cited are discussing the subject of this page or not.
"... or it uses his diagonal method." If this is referring to Cantor's diagonal argument it should be wikilinked. The page is already wikilinked, but later in the article.
"The constructive nature of Cantor's work is easily demonstrated by using his two methods to construct irrational numbers." Apparently contradicting "one proof is constructive while the other is non-constructive".
Why are we suddenly discussing irrationals here? The dispute in question is over the constructibility of transcendentals, not irrationals.

Why does Cantor's article emphasize the countability of the algebraic numbers?

Question in heading
"This has led to a controversy." This is uncited and seems to be an overstatement of what I can see in the article. Dauben says it was influenced by Kronecker and Ferreirós says it was influenced by both Kronecker and Weierstrass. Hardly a controversy, a slight difference in emphasis maybe.

See also

Links already in the body of the article should not be repeated in see also.

Images

Likenesses of Cantor and other major mathematicians in this story are available. Why not use them in this article?

SpinningSpark 16:04, 6 December 2014 (UTC)[reply]

Hello. Thank you for working on this. I've done a few edits today and I'll be back. Michael Hardy (talk) 00:10, 10 December 2014 (UTC)[reply]

The review says:

"There seems to be an implication here that one can prove that there is a number other than xn in the interior, or am I missing something?".

That is correct: there must be such a number since there are infinitely many numbers in the interval. But it is not clear what you're suggesting should be done about it, as far as editing the article is concenrred. Michael Hardy (talk) 23:55, 19 December 2014 (UTC)[reply]

That is what needs saying, since there are infinitely many numbers in any given finite interval there must be a number other than xn. The implication is there, but the article fails to explicitly say this is why it is proved. I don't think that step is going to be obvious to all readers. It is not even obvious that one is still left with a finite interval. (I am not disputing anything here of course, just looking at it from the perspective of someone completely unfamiliar with the material). SpinningSpark 13:05, 20 December 2014 (UTC)[reply]

The fact that there are infinitely many points in every open interval is known to students in secondary-school math courses, so it seems a bit like explaining what a question is in an article that quotes Hamlet saying "That is the question." However, I suppose there's not much harm in adding that. Michael Hardy (talk) 02:14, 24 December 2014 (UTC)[reply]

I see that you have rephrased the subheadings so that they are no longer questions and tinkered with the corresponding phrasing in the lead. I am afraid this is not really getting to the heart of the matter. I think some structural changes to the article need to be made to take the emphasis off this alleged dispute/disagreement. The disagreement does not seem to amount to a whole pile of beans. If it does, some sources saying so are needed. Even more, fundamentally from a GA perspective (criterion 3b), the discussion of this dispute is part of a tendency for the article to go off at a tangent to discuss Cantor's other proof(s). The non-constructive proof is the diagonal argument, no? which is not the subject of this article. I have already commented on how easily the reader can become confused over which proof is being discussed. The diagonal argument should be discussed only inasmuch as it is needed to describe this method, or in passing to say Cantor went on to use other methods. SpinningSpark 13:05, 20 December 2014 (UTC)[reply]

To me to the article seems to fall short of GA quality for structural reasons. In some ways it reads more like an essay than an article, and as such is difficult to follow as WP articles aren't essays, i.e. posing questions and answering them. For a mathematical article it contains relatively little mathematics, with too much relegated to footnotes, which themselves are far too long; they should contain only references, with footnotes kept to a minimum and clearly separated. And it has far too many quotes; quotes should be integrated into the article so the article not sources make the point, other than a quote of the proof I can see none that are needed. A reader should be able to read the article and extract the content without referencing footnotes, without the flow being interrupted by quotations.

Some more particular points. The proof is given very discursively, with 'Cantor' used throughout. It's his proof with his name but in a mathematical proof there's no need to describe it as if he's doing it, just give the mathematics. It would also be clearer with proper math, i.e. <math>, formatting. The two 'answering the questions' sections seem very unclear; 'Constructive or non-constructive nature of Cantor's proof of the existence of transcendentals' seems to be mixing the proof up with the diagonal method from the very first quote. 'Why Cantor's article emphasizes the countability of the algebraic numbers' never clearly answers that point. It also mentions a "controversy" but does not say where it's from; such a controversy definitely needs a source or sources.--JohnBlackburne^words_deeds 14:18, 20 December 2014 (UTC)[reply]

I couldn't agree more that the article needs a fundamental restructuring to make it acceptable for GA. Now that another editor has felt the need to make that comment, combined with the nominator's seeming reluctance to do anything drastic, I'm inclined to fail this now and allow it to be improved in slow time. It does not seem productive to work on the minutia when a new review from scratch would be needed after rework, but I'll wait to hear from the nominator first.

On the notes, I had noticed this myself, but did not comment as referencing formatting is explicitly not included in the GA criteria. It is an issue however. It makes it very difficult to distinguish what text is actually referenced and what merely has a note attached. Seperating the two things with grouped references would be very helpful, along with incorporating more of it into the text body.

I don't entirely agree that the lack of a formal proof is problematic. Wikipedia is not a textbook and is aimed at a more general audience. That's not to say that a formal proof would not be beneficial, but I can't see any GA criterion that is being run afoul of here. SpinningSpark 15:56, 20 December 2014 (UTC)[reply]

It does give the proof. The problem is it's so buried in text that it's hard to follow. Reducing it to just the formal mathematical proof, which is in there, would be much clearer. The attribution 'Cantor' is clear from elsewhere, while discussion should also be separate, or at least clearly distinguished. I don't disagree with WP:NOTTEXTBOOK but where a proof is short and relevant it's often given. Here the proof is the topic of the article and as such is central, while other sections such as on whether it's constructive depend on knowing how the proof works.--JohnBlackburne^words_deeds 16:24, 20 December 2014 (UTC)[reply]

To all involved, with no progress being made since December 20, I will be closing the review in 48 hours if no progress is made.--Dom497 (talk) 03:29, 14 January 2015 (UTC)[reply]

You're hilarious. Seriously.--Dom497 (talk) 14:31, 22 January 2015 (UTC)[reply]

I already made the position clear to you on your talk page, a discussion that you chose to delete with the edit summary "you're about to piss me off" rather than actually bottom out the issue with discussion. So let me be quite explicit about this: every time there is a threat to, or an actual attempt to, close this review out of process (read [[WP:GAN/I if you don't know what the process is) I will extend the review by at least a week just in case an editor who could address the issues was discouraged from doing so by the imminent closure. SpinningSpark 15:09, 22 January 2015 (UTC)[reply]

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.