Misplaced Pages

talk:WikiProject Mathematics: Difference between revisions - Misplaced Pages

Article snapshot taken from[REDACTED] with creative commons attribution-sharealike license. Give it a read and then ask your questions in the chat. We can research this topic together.
Browse history interactively← Previous editNext edit →Content deleted Content addedVisualWikitext
Revision as of 11:36, 9 February 2014 editJohnBlackburne (talk | contribs)Extended confirmed users, Pending changes reviewers, Rollbackers30,799 edits →Help needed with Correlation coefficient and Transition function: re Correlation coefficient← Previous edit Revision as of 11:42, 9 February 2014 edit undoDeltahedron (talk | contribs)Pending changes reviewers7,714 edits →Changes of Math 2.0: Perhaps it would help to create a page www.mediawiki.org/Extension:Math/FeedbackNext edit →
Line 324: Line 324:
:::::By the way, it would help us to help you if we knew where you want this sort of problem reported -- is there a centralised location for collecting comments and questions -- surely this subsection isn't it? Is there a list of known problems that don't need to be reported again? Is there a list of changes or proposed changes that you particularly want tested? Is there a list of things you've changed but which you're quite sure can't possibly affect the user experience? Is there a roadmap for future development and a schedule of proposed work? Where is all this information collated and made visible to other editors? ] (]) 09:41, 9 February 2014 (UTC) :::::By the way, it would help us to help you if we knew where you want this sort of problem reported -- is there a centralised location for collecting comments and questions -- surely this subsection isn't it? Is there a list of known problems that don't need to be reported again? Is there a list of changes or proposed changes that you particularly want tested? Is there a list of things you've changed but which you're quite sure can't possibly affect the user experience? Is there a roadmap for future development and a schedule of proposed work? Where is all this information collated and made visible to other editors? ] (]) 09:41, 9 February 2014 (UTC)
::@Deltahedron: I'm open for proposal how to organize feedback and mange the todos. Yesterday I went throguh bugzilla there were 85 bug reports. Now there are 71. (Some of them were really old and already fixed.) I updated the bug List at . From what I see there are two option for the organization of the project. Either to user that roadmap page and maybe some subpages or to use Bugzilla. I saw some green and red buttons on Misplaced Pages buttons where user can give feedback. Maybe they could be useful to get a quick overview, which features are most wanted. I'll clean the math-2 test server and remove the experimental features that are connected to my PhD thesis about Formulasearch and are not supposed to be used at Misplaced Pages in the short term. PS: My prototype use case, to find the ] by searching for <math>\langle \frac{1}{x} \rangle \ge \frac{1}{\langle x \rangle}</math>, is not reachable in the near feature, due to a lack of semantic information in formulae. --] (]) 10:18, 9 February 2014 (UTC) ::@Deltahedron: I'm open for proposal how to organize feedback and mange the todos. Yesterday I went throguh bugzilla there were 85 bug reports. Now there are 71. (Some of them were really old and already fixed.) I updated the bug List at . From what I see there are two option for the organization of the project. Either to user that roadmap page and maybe some subpages or to use Bugzilla. I saw some green and red buttons on Misplaced Pages buttons where user can give feedback. Maybe they could be useful to get a quick overview, which features are most wanted. I'll clean the math-2 test server and remove the experimental features that are connected to my PhD thesis about Formulasearch and are not supposed to be used at Misplaced Pages in the short term. PS: My prototype use case, to find the ] by searching for <math>\langle \frac{1}{x} \rangle \ge \frac{1}{\langle x \rangle}</math>, is not reachable in the near feature, due to a lack of semantic information in formulae. --] (]) 10:18, 9 February 2014 (UTC)
:::Perhaps it would help to create a page and advertising it here, be appropriate for comments and discussion on such things as the roadmap at www.mediawiki.org/Extension:Math/Roadmap and experiments and experiences at math-test2.instance-proxy.wmflabs.org. Then advertise it widely, with a permanent link at places like ] and with a mention in the current debates at ]. ] (]) 11:42, 9 February 2014 (UTC)


Physikerwelt, I just want to thank you for your work. I tried to do some work many years ago and I did not get much feedback from the developers then (it was all run less professionally then). Don't feel too bad about the align issue and don't give up. -- ] (]) 17:14, 8 February 2014 (UTC) Physikerwelt, I just want to thank you for your work. I tried to do some work many years ago and I did not get much feedback from the developers then (it was all run less professionally then). Don't feel too bad about the align issue and don't give up. -- ] (]) 17:14, 8 February 2014 (UTC)

Revision as of 11:42, 9 February 2014

This is a discussion page for
WikiProject Mathematics
This page is devoted to discussions of issues relating to mathematics articles on Misplaced Pages. Related discussion pages include:
3
Please add new topics at the bottom of the page and sign your posts.
? view · edit Frequently asked questions

To view an explanation to the answer, click on the link to the right of the question.

Are Misplaced Pages's mathematics articles targeted at professional mathematicians? No, we target our articles at an appropriate audience. Usually this is an interested layman. However, this is not always possible. Some advanced topics require substantial mathematical background to understand. This is no different from other specialized fields such as law and medical science. If you believe that an article is too advanced, please leave a detailed comment on the article's talk page. If you understand the article and believe you can make it simpler, you are also welcome to improve it, in the framework of the BOLD, revert, discuss cycle. Why is it so difficult to learn mathematics from Misplaced Pages articles? Misplaced Pages is an encyclopedia, not a textbook. Misplaced Pages articles are not supposed to be pedagogic treatments of their topics. Readers who are interested in learning a subject should consult a textbook listed in the article's references. If the article does not have references, ask for some on the article's talk page or at Misplaced Pages:Reference desk/Mathematics. Misplaced Pages's sister projects Wikibooks which hosts textbooks, and Wikiversity which hosts collaborative learning projects, may be additional resources to consider.
See also: Using Misplaced Pages for mathematics self-study Why are Misplaced Pages mathematics articles so abstract? Abstraction is a fundamental part of mathematics. Even the concept of a number is an abstraction. Comprehensive articles may be forced to use abstract language because that language is the only language available to give a correct and thorough description of their topic. Because of this, some parts of some articles may not be accessible to readers without a lot of mathematical background. If you believe that an article is overly abstract, then please leave a detailed comment on the talk page. If you can provide a more down-to-earth exposition, then you are welcome to add that to the article. Why don't Misplaced Pages's mathematics articles define or link all of the terms they use? Sometimes editors leave out definitions or links that they believe will distract the reader. If you believe that a mathematics article would be more clear with an additional definition or link, please add to the article. If you are not able to do so yourself, ask for assistance on the article's talk page. Why don't many mathematics articles start with a definition? We try to make mathematics articles as accessible to the largest likely audience as possible. In order to achieve this, often an intuitive explanation of something precedes a rigorous definition. The first few paragraphs of an article (called the lead) are supposed to provide an accessible summary of the article appropriate to the target audience. Depending on the target audience, it may or may not be appropriate to include any formal details in the lead, and these are often put into a dedicated section of the article. If you believe that the article would benefit from having more formal details in the lead, please add them or discuss the matter on the article's talk page. Why don't mathematics articles include lists of prerequisites? A well-written article should establish its context well enough that it does not need a separate list of prerequisites. Furthermore, directly addressing the reader breaks Misplaced Pages's encyclopedic tone. If you are unable to determine an article's context and prerequisites, please ask for help on the talk page. Why are Misplaced Pages's mathematics articles so hard to read? We strive to make our articles comprehensive, technically correct and easy to read. Sometimes it is difficult to achieve all three. If you have trouble understanding an article, please post a specific question on the article's talk page. Why don't math pages rely more on helpful YouTube videos and media coverage of mathematical issues? Mathematical content of YouTube videos is often unreliable (though some may be useful for pedagogical purposes rather than as references). Media reports are typically sensationalistic. This is why they are generally avoided.

This is the talk page for discussing WikiProject Mathematics and anything related to its purposes and tasks.
Archives: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73Auto-archiving period: 15 days 

Misplaced Pages:Misplaced Pages Signpost/WikiProject used

Archiving icon
WikiProject Mathematics archives ()
Earlier years

This page has archives. Sections older than 15 days may be automatically archived by Lowercase sigmabot III.
Edit this box

Distance from a point to a line and Perpendicular distance

These two articles have been around for a while and except for a few sentences in the lead of the second, their content is pretty much the same (or could easily be made so). It seems to me that a merge is in order. The question is which way should the merge go? I am not happy with either possibility, which is why I am bringing the subject up here. The distance from a point to a line article consists of several different proofs of the same 2-dimensional formula. This article was considered for deletion not too long ago, and several of you argued to keep it (but it doesn't look like anyone has seriously worked on it). I fixed a couple of proofs that had gotten garbled, but to improve the page I think we need to generalize the topic rather than provide more and better proofs of this simple formula. The Perpendicular distance article provides yet another proof of the 2-dimensional formula, and then states and proves a generalization to a formula for the distance between a point and a flat. This article has no references and no diagrams. Neither article contains a calculus based proof (or any other type of proof) that the shortest distance is along a perpendicular (but it is clearly assumed in each article). I think that at least the 2-dimensional proof would not be too far off the level of these articles. I also think that "perpendicular distance" is just a terrible title. My ideal solution would be to merge both articles into a new article, Distance from a point to a flat, but that title would not resonate well with the level of the readers of this article. I am looking for comments, suggestions, etc. Thanks. Bill Cherowitzo (talk) 18:28, 25 January 2014 (UTC)

I converted perpendicular distance into a set-index article listing four other articles (including the point-line distance one) on specific types of perpendicular distance. —David Eppstein (talk) 21:38, 25 January 2014 (UTC)

Thanks, I can work within that framework, but it still leaves open the issue of the weakness of the point-line distance article. There must be something we can do besides pile on proof after proof of the same result. Bill Cherowitzo (talk) 18:55, 26 January 2014 (UTC)

Topology of uniform convergence

Topology of uniform convergence until minutes ago redirected to Polar topology. Someone created a new article titled Topologies of Uniform Convergence, with all those capital letters and the plural, and I moved it to Topology of uniform convergence, deleting the redirect in the process. So now we have

So:

  • Should these get merged?
I'm not sure. The article Topology of uniform convergence deals with the case where the dual pair is (X, X') (i.e. with Y = X' in the polar topology article). So depending on what kind of presentation the reader is looking for, they might want one page but not the other.Mgkrupa (talk) 17:57, 30 January 2014 (UTC)
Also, thank you for your improvements on the article. Mgkrupa (talk) 18:14, 30 January 2014 (UTC)
  • What work should be done on the new article, which is longer than the old one? I am not comfortable with the intro paragraph, since the first sentence doesn't even mention the term topology of uniform convergence. Michael Hardy (talk) 04:51, 27 January 2014 (UTC)
Polar topology is a better title, since "topology of uniform convergence" usually means something else. Sławomir Biały (talk) 12:11, 27 January 2014 (UTC)
I think that Polar topology is not the better title since the article topology of uniform convergence describes topologies, such as on the spaces L(X, Y), that are NOT polar topologies. These topologies are all known as topologies of uniform convergence. Also, you said that ""topology of uniform convergence" usually means something else", what do you mean by that? Mgkrupa (talk) 15:08, 28 January 2014 (UTC)
See the disambiguation page uniform topology. Sławomir Biały (talk) 16:52, 28 January 2014 (UTC)
Also, you're right that these don't seem to be called polar topologies in the literature. The standard term seems to be just the S {\displaystyle {\mathfrak {S}}} -topology (see Bourbaki, Espaces vectoriels topologiques, III.§3). Sławomir Biały (talk) 00:41, 29 January 2014 (UTC)
From the page Uniform topology I see that one of the links is for Polar topologies, which this article covers, and the other two are for uniform convergence of real-valued (which this article also covers) or metric space-valued functions (which it doesn't cover) and finally for uniform spaces. In terms of sending people to the most general notion of uniform convergence the article "Topology of uniform convergence" should then just be a link to uniform spaces, or a dis-ambiguity link (although I think that this would just confuse anyone who doesn't already know what uniform convergence means and is simply trying to learn the basic notions). What if we were to call this article "Topology of uniform convergence of vector-valued functions"? A long name but also the most accurate.Mgkrupa (talk) 17:45, 30 January 2014 (UTC)

Pollock's conjecture(s)

I'm proposing a merge of the "tetrahedral" and "octahedral" conjectures into a single article. Any comments? (I only noticed this because one of them was added as a "See also" to Fermat polygonal number theorem, where I'm sure it should go.) — Arthur Rubin (talk) 21:20, 27 January 2014 (UTC)

In general I favor separate articles for separate topics, but in this case the topics are so similar, and what little there is to be said about them also so similar, that I agree that a merge makes sense. —David Eppstein (talk) 21:26, 27 January 2014 (UTC)
Working copy at User:Arthur Rubin/Pollock's conjectures. The only difference is that the Mathworld article "Pollock's Conjecture" points only to the tetrahedral conjecture, while "Octahedral Number" has mentions the octahedral conjecture. — Arthur Rubin (talk) 22:25, 28 January 2014 (UTC)

spelling of "l'Hopital's rule"

With this edit, a user has recently changed the text from explaining the difference as an "alternative spelling" to a "misspelling." This switches the tone from prescriptive to proscriptive, and so it deserves some attention. The previous version seems safer, if valid. Anyone know enough about French or math history to be able to verify it is an alternative spelling? Thanks Rschwieb (talk) 14:09, 28 January 2014 (UTC)

From Guillaume de l'Hôpital#Notes: "In the 17th and 18th centuries, the name was commonly spelled "l'Hospital", and he himself spelled his name that way. However, French spellings have been altered: the silent 's' has been removed and replaced with the circumflex over the preceding vowel. The former spelling is still used in English where there is no circumflex." Misplaced Pages is not a reliable source, but it does match my knowledge about French spelling, and makes sense. No such user (talk) 14:45, 28 January 2014 (UTC)
I agree. In French, the circumflex is (almost) always used for replacing a silent "s". The spelling "l'Hospital" is attested by the titles of two references of fr:Guillaume François Antoine, marquis de L'Hôpital and two external links of the same article. In fact, it appears that the correct spelling is "de l'Hospital", and that the other spellings are only common misspellings. Omitting the "de" is similar as talking about "Gaulle" instead of de Gaulle. However, "l'Hôpital rule" is a translation of the French "règle de l'Hôpital", and "règle de de l'Hôpital" is not so euphonic. D.Lazard (talk) 15:36, 28 January 2014 (UTC)
I don't think the spelling with a circumflex is an error. Some editions of his book appeared with this spelling. Tkuvho (talk) 15:49, 28 January 2014 (UTC)
Authors aren't known for misspelling their own names, so presumably this proves beyond any doubt that L'Hôpital's rule was in fact due to Bernoulli! Sławomir Biały (talk) 16:23, 28 January 2014 (UTC)
To editor Tkuvho: This depends if one accepts the modernization of the spelling for old people names. Thus "l'Hôpital" could be correct for the rule but not for the mathematician. Similarly, we have Vieta's formulas and François Viète (here, it is not modernization, but translation in another language). D.Lazard (talk) 16:50, 28 January 2014 (UTC)
Fontenelle's Éloge de M. le marquis de L'hopital (cited at the French page you linked) must have been one of the first to make the mistake of spelling it without an "s". Tkuvho (talk) 17:11, 28 January 2014 (UTC)

Thanks for all of the knowledgeable and prompt help. Rschwieb (talk) 15:26, 29 January 2014 (UTC)

AfC submission (Deep learning)

Misplaced Pages talk:Articles for creation/Deep learning. FoCuSandLeArN (talk) 13:41, 29 January 2014 (UTC)

There is already a deep learning article, apparently identical to the submission. Ozob (talk) 14:19, 29 January 2014 (UTC)
I (manually, as the script won't install until at least I close the browser) declined the submission, for that reason. The editor who requested the AfC submission replaced the mainspace article with his modified version. The question of whether this is a matter for WP:MATH is unclear, but I think this section should be considered closed. — Arthur Rubin (talk) 16:51, 29 January 2014 (UTC)

Mathematics and art

Mathematics and art asserts as factual many discredited fringe theories e.g. about the use of the golden ratio in Greek and Egyptian aesthetics. I tried a modest {{dubious}} tag, only to have the edit reverted and a (bad) source for these claims added. Any suggestions for a process that can get this cleaned up? —David Eppstein (talk) 01:27, 30 January 2014 (UTC)

The dubious tag is understandable--that's a pretty ambiguous sentence. It looks like Chiswick tried to be helpful by adding two references, which you don't like. He probably thought he answered your call for references and (along with the refs in the, e.g., Parthenon section) removed the dubious tag. I've worked with him on a symmetry article and he seems a reasonable fellow--why not discuss it with him on the article's talk page? --Mark viking (talk) 01:53, 30 January 2014 (UTC)
There are more citations in Mathematics_and_art#Parthenon. The only statement questioning the use of the golden ration does not have a citation yet. RockMagnetist (talk) 03:17, 30 January 2014 (UTC)
The various subsections of Golden_ratio#Applications_and_observations contain a bunch of useful references for someone interested in rewriting the article more skeptically. --JBL (talk) 04:10, 30 January 2014 (UTC)

Intersection

This has recently been created, with the disambiguation page previously at this location moved to Intersection (disambiguation). The problem is I don't think the mathematical term is the primary topic: if anything the common usage of the word is when talking about road intersections (or listening to your sat-nav read them off to you), but probably being a common English word used in many ways and fields its best without a primary topic, as before.

At the same time the new article at Intersection doesn't seem to be on a distinct topic: it's mostly on geometric intersection, but includes some set theory and possibly other areas, leading to a very confusing introduction, some even more confusing links, and little else. The problem is the different mathematical uses have little in common, except for being different interpretations of the word in different fields. But that does not make for a good article topic.--JohnBlackburnedeeds 02:21, 30 January 2014 (UTC)

A road intersection is a 2-D intersection. Intersections in (Euclidean) geometry, algebraic geometry, and set theory derived from one common idea. What namely “does not make for a good article topic”, this WP:CONCEPTDAB? Okay, make the intersection (geometry) article first, and then we’ll discuss the merit of the “most general article”. I think, most people here will agree that intersection (Euclidean geometry) shall not consider differentially-geometric aspects. Imagine one wants to know what is an intersection of a line and a quadric in a projective plane. These are generally curves, but the article on curves hardly considers the question of intersections. These are submanifolds, but the reader hasn’t necessarily know this word. If s/he come to my stub, s/he will obtain some minimal idea how to approach the problem. If s/he come to the original dab, it will be a puzzle. Where to go: intersection (Euclidean geometry) (formerly “intersection point”)? It misses the line at infinity. Intersection (set theory)? S/he will not learn anything new about the problem in question. Intersection theory? The reader will get a huge charge of general nonsense. Whereas for a reader who looks for something about Euclidean geometry there is little difference between “my article” and a conventional dab page.
“Even more confusing links” – which namely? “Little else” – yes, the house is just started. I think one should say about dimensions, as well as relationship between intersection of submanifolds and the orientation structure (what is expressed with the intersection number for dimintersection = 0, but in more general case). Incnis Mrsi (talk) 08:59, 31 January 2014 (UTC)
I agree with Incnis Mrsi that the situation was confusing. However, IMO, his solution is not the best one. In fact, in mathematics, there is only one concept of intersection, the set theoretical one. I agree that, in incidence geometry, lines and planes are not considered as sets, but they are intuitively. And, in incidence geometry, one may consider the set of the points that are incident to a line; this allows to consider the intersection of two lines as a set theoretical intersection. Therefore I suggest the following
D.Lazard (talk) 11:08, 31 January 2014 (UTC)
I do not agree with Daniel that “there is only one concept of intersection, the set-theoretical one”. Set-theoretically there is no difference between an intersection proper and a tangential point (note another red link), but geometrically and algebraically there is. Also look at the picture please: set-theoretically there is a set of two points, geometrically there are simply two intersections, algebraic-geometrically there is a 0-manifold consisting of two components with different signs, provided the line, the circle, and the plane all are oriented. Why not explain it in one article, indeed? Also, what is now intersection (Euclidean geometry) is not a suitable candidate for converting to intersection (geometry). It is a typical enumerative article like the current revision of rotation (mathematics) (see talk:Rotation group for further development) and, unfortunately, many other Misplaced Pages articles, not mathematical only, that need to be conceptual instead of enumerative. Incnis Mrsi (talk) 15:07, 31 January 2014 (UTC)
I certainly edited Intersection (mathematics) (edit | talk | history | links | watch | logs) to redirect it to intersection (I even remember how I entered {{R from extra disambig}}). Apparently there was a browser glitch and the edit wasn’t saved. BTW, the fr:Intersection (mathématiques) article is IMHO too poor to borrow something really usable from it. Incnis Mrsi (talk) 16:28, 31 January 2014 (UTC)
@Incnis Mrsi: When you write the the circle and the line of the picture have "two intersections" you are correct with respect to the non-mathematical meaning of "intersection" (road intersection), but you are not mathematically correct: the common mathematical terminology is that the circle and the line have two points of intersection, or two intersection points, or also that their intersection consists in two points. All these formulations are standard ways to talk about set intersections. Also it is clear that when the sets that are intersected have an extra structure, their intersection has also an extra structure. If I follow you argument, we should have an article for the intersection of vector spaces (the set intersection of two subspaces is a subspace), for ideals, and so on. Here, we have that the intersection of two algebraic varieties is an algebraic set, which, on your example, is not irreducible and has two components consisting of isolated points. Similarly a tangential point, aka tangent point, is a point of the (set) intersection that has a specific extra structure (the two curves or varieties or manifolds are tangent at that point). I do not see in your post any argument against merging Intersection and Intersection (set theory) into a single article called Intersection (mathematics). However, I see many reasons to use such a merger to improve both articles. D.Lazard (talk) 17:34, 31 January 2014 (UTC)
A good point about an additional structure. The only disagreement is about the precedence: I do not think the set theory should be qualified as the only formalism for intersection in mathematics. It is the standard one, certainly, but not an unique, and definitely not historically an original one (see below). Incnis Mrsi (talk) 19:28, 31 January 2014 (UTC)
(edit conflict)I agree with most of that: I think Intersection (geometry) is better than Intersection (Euclidean geometry); if that means adding some non-Euclidian geometry then do so, it's a good and natural extension of Euclidian geometry. I obviously agree with moving the DAB page back to Intersection. But I don't think it or the set theoretic use of intersection is the main mathematical use, I'd say that the geometric usage is as important. Looking at Intersection (Euclidean geometry) and Intersection (set theory) both are significant areas. So I'd change Intersection (mathematics) to redirect to the DAB page, i.e. Intersection. Apart from that any see-also links and {{main}} type links can be added as makes sense to clarify connections and help readers find articles.--JohnBlackburnedeeds 17:47, 31 January 2014 (UTC)
John, do not ignore my question, please. Which of my links are confusing? Of course my solution is not the best possible, but what namely did I do wrongly? Incnis Mrsi (talk) 19:28, 31 January 2014 (UTC)
I mentioned projective plane only as an easily accessible geometry that isn’t Euclidean. There are numerous other alternatives to the Euclidean space: pseudo-Euclidean spaces, for example, or just the affine geometry where words “circle” and “perpendicular” are meaningless. Incnis Mrsi (talk) 22:41, 31 January 2014 (UTC)

The primary usage of the word "intersect" in my opinion is to cut, in the geometrical sense. This is consistent with the Latin origin of the term, and the current English vernacular (as in the intersection of two streets). The set theoretic meaning if the term was not even introduced until the 20th century (or possibly the late 19th century, the OED puts it at 1909). I'm not sure what this implies for an article on the topic. Sławomir Biały (talk) 17:40, 31 January 2014 (UTC)

A bit of support from an unexpected source. But I think we have two distinct (although related) questions: the fate of the new stub (as well as of intersection (mathematics) that I failed to redirect because of a glitch), and the primary topic of the title Intersection (edit | talk | history | links | watch | logs). I do not have a highly competent opinion about the latter and IMHO it should be discussed at talk:Intersection (disambiguation) because it can possibly escalate to an RfC or otherwise attract an attention of non-mathematical disambiguators. Incnis Mrsi (talk) 19:28, 31 January 2014 (UTC)
If an RfC's needed it could be created here; RfCs can be started anywhere, including user pages. But I don't think one's needed; we're not yet reached that point. As for the fate of the current Intersection there's no need for it: Intersection (Euclidean geometry) and Intersection (set theory) are two distinct topics. If there's need to connect them that can be done using appropriate text and links in both articles, and both should be linked from the disambiguation page. But they're not both sub-topics of a more general "mathematical intersection" topic.--JohnBlackburnedeeds 19:58, 31 January 2014 (UTC)
John, you catch from the discussion only things convenient to you, and dodge my question about your own claim the second time. Nobody tries to refute that “Intersection (Euclidean geometry)” and “Intersection (set theory)” are distinct. The discussion deals with following problems:
  1. There is no article “intersection (geometry)” (virtually all participants).
  2. The “new intersection” topic can be merged with “intersection (set theory)” to “intersection (mathematics)” (D.Lazard).
  3. The “intersection (Euclidean geometry)” has an inappropriate content and structure to be simply expanded to “intersection (geometry)” (Incnis Mrsi).
  4. “Intersection (geometry)” is historically the primary meaning, and the set-theoretical interpretation of intersection was popularized only in 20th century (Sławomir Biały).
You did not comment on any of these 4 points in a reasonable way, only reiterate some of things you said from the beginning: one should not bother to improve or save anything of it, Incnis Mrsi wrote this crap. Incnis Mrsi (talk) 22:41, 31 January 2014 (UTC)
Since there is discussion over what "intersection" truly means, mathematically, I would like to point out that most intersections can be interpreted as fiber products in an appropriate category. This is true of sets, vector spaces, ideals, schemes, and so on. But the primary meaning of "intersection" is not "a kind of fiber product", and the primary article on intersections should not introduce it as such. Ozob (talk) 02:00, 1 February 2014 (UTC)
Just a hint: Please have a look to the German WIKI on intersection = Schnitt. --Ag2gaeh (talk) 10:57, 1 February 2014 (UTC)
The German is less directly analogous, since "Schnitt" is already part of the German vernacular and literally means "cut" (as a noun, at least, as in "ein Schnitt von Rindfleisch"—"a cut of beef"), whereas in English intersection is an old loan word from Latin that would have been used exclusively in a scientific setting before its adoption into the language. This is reflected in the relative rarity of the word "intersection" in modern day-to-day English (with the exception of referring to day-to-day things that are actually geometric intersections, like streets). A better comparison for de:Schnitt would be the English disambiguation page cut. Sławomir Biały (talk) 13:41, 1 February 2014 (UTC)
Sorry, I only had the mathematical part of the German page in mind and did not think about the "intersection" of a knife and a beef. --Ag2gaeh (talk) 10:45, 2 February 2014 (UTC)

Misplaced Pages talk:Articles for creation/Phan Sequence

Hello, mathematicians! Last chance to read this old Afc draft before it disappears for lack of reliable sources. —Anne Delong (talk) 15:11, 30 January 2014 (UTC)

There is no chance that reliable sources exist for this sequence. --JBL (talk) 16:49, 30 January 2014 (UTC)
Yes, I thought so. Thanks. —Anne Delong (talk) 21:41, 30 January 2014 (UTC)

Misplaced Pages talk:Articles for creation/Stability of Boolean networks

Dear math experts: Here's another old abandoned Afc submission that's about to be deleted. Is this a notable topic, and should the article be kept? —Anne Delong (talk) 18:10, 30 January 2014 (UTC)

Delete it. Ozob (talk) 02:02, 31 January 2014 (UTC)
The stability of a Boolean network is one of the most important properties of the dynamics of such a network. The article gives a poorly described criterion for stability in NK models with no significant intro or background. Our Boolean network article could use more material on dynamics, but this article is unfortunately not it. I would recommend passing on this one. --Mark viking (talk) 03:02, 31 January 2014 (UTC)
If the AfC is kept, then the material can be used to supplement the existing articles. Is this reason enough to keep it for a few months?
Sorry I wasn't clear--I don't think the prose is salvageable and refs are pretty easy to find on this topic, so I don't see the utility in saving any of it. Thanks, --Mark viking (talk) 20:04, 31 January 2014 (UTC)
Thanks for taking the time to check this out. —Anne Delong (talk) 03:55, 3 February 2014 (UTC)

AfC submission

Misplaced Pages talk:Articles for creation/K-trivial sets. Regards, FoCuSandLeArN (talk) 00:03, 4 February 2014 (UTC)

Also Misplaced Pages talk:Articles for creation/One-Shot Deviation Principle. Thank you, FoCuSandLeArN (talk) 01:08, 6 February 2014 (UTC)

EOM links

Spinningspark (talk · contribs · deleted contribs · logs · filter log · block user · block log) has been reverting the addition of EOM links en masse. I don't really have strong opinions about this, but I generally find the EOM to be a rather useful supplement to our own treatment of mathematical topics. At least some of the removed links are of a high quality (actually the first I noticed at Korn's inequality.) What does the project think about these edits? Should they be reverted? Sławomir Biały (talk) 16:31, 4 February 2014 (UTC)

I reverted them en masse because they had been inserted en masse. I am not particularly objecting to the site, I don't really know anything about the site one way or the other. However, the IP that has been inserting the links has been doing nothing else besides inserting them. That is spamming, regardless of the quality of the links. No justification was offered for inserting them in the edit summary, and I noted that many had been added as references without adding anything to the text. Since references are supposed to be there to support the text then that in itself looks spammy. As to their quality as external links, if editors think they are adding something that is not covered in the article then I'm fine with that, but it would be preferable that, if the material pointed to is suitable for Misplaced Pages, it should be added to Misplaced Pages and use the site as a ref instead of an EL. There are sometimes reasons material cannot be added to Misplaced Pages and an EL is justified on those grounds, but I'm not seeing anything here that fits in that category. SpinningSpark 16:59, 4 February 2014 (UTC)
To be sure, I don't question your reasons for removing the links, but I do think that in many cases these links are a valuable supplement to the article and that seems worth some consideration. Sławomir Biały (talk) 17:24, 4 February 2014 (UTC)
I'm fine with you just reverting me on the ones you think are useful. SpinningSpark 18:02, 4 February 2014 (UTC)
Sorry, I was the one adding all the links. My name is Camillo De Lellis and I am one of the editors of the Encyclopedia of Math, you can see my page there at User:Camillo.delellis. I noticed that several entries of the EoM are already regularly linked here at wikipedia, but the vast majority refers to entries which are rather old versions. Since I and other people have been renewing several of the pages there, I thought it was a good idea to check where they could be appropriately linked in[REDACTED] and I started going through mine ("En masse" might be subjective... note that from Christmas till now I anyway did not manage to link all the pages I renewed at EoM :-)). I spend a couple of hours per week with the entries at the EoM: this pretty much exhausts my alloted time for wikis and so I am not an active contributor of wikipedia. I was linking only pages which I believed appropriate and it seems to me that the pages I linked are more useful and appropriate than most of the EoM pages which are linked by other contributors (I am pretty choosy myself :-))). Some of the EoM pages do contain more material than wikipedia's corresponding pages, as for instance Function of bounded variation. Some, however, do not and I would consider EoM as a general reference work. In any case if what I was doing is considered spamming I stop right away. Sorry again, Camillo. — Preceding unsigned comment added by 188.154.99.222 (talk) 09:01, 8 February 2014 (UTC)

I'm not a fan of addition/revert em masse either way. However disagree with an overly literal application of particular line in WP:EL here, that is only adding it if it really offers information not being contained in the article yet. I rather treat it (and to degree links to MathWorld, MacTuror and PlanetMath as well) as "standard" link as long as the EL is rather small and empty. In a way similar to linking the IMDB in movie related articles. The reason for that being twofold. For short article EOM, MathWorld and MacTutor can also be considered as "general" sources/sources outside of footnotes. In such that cases a placing under references might be more appropriate, but since they are links some editors place them under EL. The other reason is simply, that I consider it as beneficial to readers to offer links to alternative encyclopedic representations of math content.--Kmhkmh (talk) 15:39, 7 February 2014 (UTC)

Actually, WP:EL is much more stringent than you portray. It does not allow EL for information that is "not contained in the article yet". Rather, it is for information that would not be in the article if it were developed to FA standard: a much more stringent requirement. Also, I believe that the community consensus on ELs is that the default position is not to include them; that is, there should be a definite good reason before they are added rather than add them automatically. However, I agree that it is sensible to consider the state of the development of the article. Good ELs in a stub or short start class article can benefit both readers and editors seeking to develop it. Well developed articles on the other hand should be a lot more choosy about including ELs. SpinningSpark 00:35, 8 February 2014 (UTC)
Yes, I'm aware of that, it is so strict to the point, that it is almost nonsensical if applied literally (always). I'd also dispute the community consensus with regard to the literal reading somewhat (after following the page for years). It is more that it has been established via "tradition" and some editors prefer an overly strict formulation (as a sharp knife in dispute and block link farms and undesirable links) but in doubt relax it in practice. In fact somewhat similar to your approach. However looking at the discussion archives and behaviour of reasonable editors in practice, I don't really see a consensus. In fact imho this guideline only works because the strict literal reading of the line you refer too is largely ignored in practice and people similar follow the nutshell description or common sense.
Anyway having said all I more or less agree with you distinction between smaller and larger article. However even for a large article an external link with less information can be beneficial as it potentially conveys important or some information faster than a rather large article and it might be beneficial for readers to the simply see the "authoritative" encyclopedic take on the subject. That's why I don't mind to link the EOM for instance as long as external link section is not already filled with better or more suitable links. But that's just me.--Kmhkmh (talk) 08:41, 8 February 2014 (UTC)
I would prefer not to focus on the policy for external links in general but to focus the discussion on EOM. If the consensus is that addition of links to the EOM is helpful for readers of mathematics articles, so that the links improve the encyclopaedia, then we should Ignore All Rules that prevent us from making that improvement. Deltahedron (talk) 09:03, 8 February 2014 (UTC)
I agree, above I tried to make a case for WP:IAR and explain why many editors do not follow a literal reading of the policy, that is apply WP:IAR. Personally I'd consider an EOM-Link in most cases as beneficial.--Kmhkmh (talk) 14:36, 8 February 2014 (UTC)

"Octagonal"

The usage of Octagonal (edit | talk | history | protect | delete | links | watch | logs | views) is under discussion, see talk:Octagonal -- 70.50.148.248 (talk) 07:48, 5 February 2014 (UTC)

Cyclic sieving

I've created a new, painfully stubby, article titled cyclic sieving. So work on it. Or, in other words, have fun. Michael Hardy (talk) 05:36, 6 February 2014 (UTC)

I think it would benefit greatly from one or two examples. JRSpriggs (talk) 06:19, 6 February 2014 (UTC)

So far only two articles link to it, so that's another thing to work on. Michael Hardy (talk) 20:07, 6 February 2014 (UTC)

Problem with multiline equations

The Spherical trigonometry article has several bits of LaTeX that Failed to parse, for example the displayed equation in the Spherical trigonometry#Polar triangles section. My guess is that the alignat and align environments are not being handled. I went to the Wikimedia page Help:Displaying a formula and found that the align and alignat environments seem to be supported, but also generate a Failed to parse error on the page. A bug report at the wikimedia bugzilla suggests that these environments might be supported in MathJax, but not texvc. Is this a bug that others see? Thanks, --Mark viking (talk) 23:46, 6 February 2014 (UTC)

align should work with texvc. I just looked at a place I recently edited a formula with align, Triple product#Using geometric algebra, which I'm sure was working then, and it's now broken. The parser is complaining about 'aligned' not align, so is doing some sort of substitution, but I don't know if this is something it normally does. But I'm sure this was working two weeks ago: I was only editing that section, there's only one formula, and I would have used preview when editing it. I've tested MathJax before but haven't had it enabled for several months.--JohnBlackburnedeeds 00:01, 7 February 2014 (UTC)

I've asked about this issue at Misplaced Pages:Village pump (technical)#Math aligned environments failing to parse. Maybe an answer will magically appear there... Melchoir (talk) 02:08, 7 February 2014 (UTC)

Thanks John, for verifying the problem, and thanks Melchoir, for submitting to the technical village pump. --Mark viking (talk) 03:29, 7 February 2014 (UTC)
There is a similar problem at Noether's theorem‎. Unfortunately, some people are trying to fix the problem by editing the source, not realizing that it is the software's interpretation of the source which is at fault. JRSpriggs (talk) 07:57, 7 February 2014 (UTC)

I ran into the problem as well, all multline formulas using \begin{align} ... \end{align} seem to be affected (see here Help:Displaying_a_formula, scroll through the page). However some special symbols seem to have a rendering issue as well (see de:Hilfe:TeX, scroll through the page)--Kmhkmh (talk) 11:38, 7 February 2014 (UTC)

As far as I can see, it seems to work again for me. -- Jitse Niesen (talk) 12:05, 7 February 2014 (UTC)
I'm having the same problem and unfortunately it's leading to people (mostly from anonymous IPs) removing important formatting. The issue with aligned formulas should have critical priority: it needs to be fixed as soon as possible. The servers were offline at some point yesterday, and I assume that this was related. Sławomir Biały (talk) 12:25, 7 February 2014 (UTC)
This remains broken for me. Seconding the urgency of this; the longer this takes, the more reverting we will have to do later. --Rhombus (talk) 13:07, 7 February 2014 (UTC)
I discovered that it has to do with the maths preferences; MathJax works fine (at least for me) while PNG gives problems. It is still not fixed; sorry for falsely suggesting it is working again. The relevant bug seems to be bugzilla:60997 bugzilla:61012. -- Jitse Niesen (talk) 13:27, 7 February 2014 (UTC)

Problem still present. Isn't there a way to post a warning about this issue? Just to avoid people editing all sorts of math articles, escalating problems. This section in this talk page isn't all that easily found, even when looked for:) This is a serious issue after all. I think it should be posted somewhere central that it's known and is a server-side problem. --Loudandras (talk) 19:21, 7 February 2014 (UTC)

Personally, I think all articles in the List of mathematics articles should be semi protected, and an edit notice added saying that Misplaced Pages's implementation of LaTeX is broken, until it is fixed. I doubt that I would be able to convince WP:RFPP that this is an appropriate course of action, but even after one day I think this is going to require related changes patrolling of the List of mathematics articles by someone with tool server access (which afaik means Jitse). More than a day will be an unmitigated disaster. Sławomir Biały (talk) 19:30, 7 February 2014 (UTC)
I'm afraid a bunch of physics-related articles will be affected as well (Maxwell's equations for instance have already been 'fixed'). Not that removing the 'align' environments causes disaster, but these formulae would still look better in the original version. This could really get messy. --Loudandras (talk) 22:37, 7 February 2014 (UTC)

I've been finding the server very slow to respond today when I edit articles with mathematical equations in them, to the point where it sometimes times out without saving my changes (that is, it is a back end issue rather than a front end issue). Perhaps this is related? —David Eppstein (talk) 01:08, 8 February 2014 (UTC)

I was getting timeout/too many people accessing the page errors just looking at Noether's theorem and Help:Displaying a formula. Maybe the latter was unusually busy with the problems but not the former. And anyway WP handles pages becoming suddenly very busy for all sorts of reasons without problems usually. So it seems very likely it was the rendering problems on those pages, causing it somehow to timeout.--JohnBlackburnedeeds 04:02, 8 February 2014 (UTC)

Wikid77 (talk · contribs) and I have reverted each other a couple of times at Integral and Spherical trigonometry regarding this issue. Other editors may want to express an opinion. Ozob (talk) 06:08, 8 February 2014 (UTC)

As I expressed at the VP I back you completely on this: the articles aren't broken so no need to 'fix' them.--JohnBlackburnedeeds 06:29, 8 February 2014 (UTC)
Ultimately we need to do the right thing for the readers! That includes today's readers and future readers, after the software bug is fixed (and we don't know when that will happen).
It would be a reasonable compromise to "fix" high-traffic articles and list the edits on this talk page. Then, after the bug is fixed, it will be easy to revert the edits. Melchoir (talk) 07:06, 8 February 2014 (UTC)
Yes, but please wait a little bit if someone succeeds in turning amsmath support back on in texvc. Btw., the better replacement of align is {array}{rl}, {array}{rlrl} etc.--LutzL (talk) 11:48, 8 February 2014 (UTC)

I wrote a hotfix for that. It's waiting to be reviewed. After the problem should diaper. --Physikerwelt (talk) 11:48, 8 February 2014 (UTC)

Thank you! Ozob (talk) 15:30, 8 February 2014 (UTC)
Thanks for addressing this bug! Is there some way we can help write an automated test so that for instance breakages like this are caught in Jenkins before being deployed? --Mark viking (talk) 17:43, 8 February 2014 (UTC)
Sorry again for breaking the rendering. It was really only my fault. Prior to the change, the user input was checked (i.e. whitelisting to a secure subset of all possible TeX commands) for the PNG rendering mode only and not for MathJax and Source rendering modes. This caused a potentially security problem, i.e. given the fact that we wanted to render MathJax at the server side. I opened a bug in May 2013 and wrote and fixed it with the in the first attempt to release Math2.0 in October 2013. Since nobody wanted to review this, I broke down the change (that changed a lot of things (see list below) to several small commits. While breaking down the commits I made the mistake to put the required change for the texvc renderer to another commit than the security checking for MathJax. Since still nobody from WMF side wanted to do code review for that, I recruited new code reviewers for the math extension. In the end the change that does the security check for MathJax was merged and the changes to the PNG rendering mode were not merged. As a result the PNG rendering mode did the security checking twice. Unfortunately the output of the checked output is insecure, e.g. align is transformed to aligned which is not whitelisted. So that's the whole story.
For the testing, I think it would help if someone would vote for the bug concerning the test coverage. --Physikerwelt (talk) 21:49, 8 February 2014 (UTC)
The fix has been deployed now, and looking at a few pages it does seem fixed, though I had to purge one page to see it working.--JohnBlackburnedeeds 21:21, 8 February 2014 (UTC)
For those who don't know, you can purge the server cache with ?action=purge. Melchoir (talk) 21:26, 8 February 2014 (UTC)
You can also enable it in a menu in your gadgets preferences, 'Appearance' section, perhaps temporarily if you're checking and cleaning these up.--JohnBlackburnedeeds 21:55, 8 February 2014 (UTC)

Displayed equations are centered?

I recently edited Talk:Transitive relation, and found that displayed equations are centered. I use MathJax, so my view may not be typical, but there could be a problem here.

a R b ∧ b R c → a R c {\displaystyle aRb\land bRc\rightarrow aRc}

Yep, it's a problem here, also. — Arthur Rubin (talk) 17:07, 7 February 2014 (UTC)

Looks like a problem with the code used to generate the MathJax code examining the code we have
<div class="MathJax_Display" role="textbox" aria-readonly="true" style="text-align: center;">
  <span class="MathJax" id="MathJax-Element-2-Frame" style=""> ...
  </span>
</div>
I've created Template:Bug for this issue.--Salix alba (talk): 19:35, 7 February 2014 (UTC)

How should it look like? I had a long discussion about that with one of the MathJax developers. As a result, the goal is to introduce a Displaystyle feature --Physikerwelt (talk) 11:55, 8 February 2014 (UTC)

It should respect the style defined in the wikitext. If you look at the example above its
:<math>a R b \and b R c \rightarrow a R c</math>
the maths tag is preceded by a : and WP:MARKUP#Indent text which says "Each colon at the start of a line causes the line to be indented by three more character positions." Editors will expect this rule to be applied, if the intention had been to center the equation then the {{center}} template or equivalent to be used.
There is a problem with the wikitext and you could say the above markup is a workaround for lack of proper maths support for display maths. --Salix alba (talk): 13:17, 8 February 2014 (UTC)
If you disable JavaScript in your browser, you can see that the code used to pass to MathJax is:
<dl>
<dd><span class="tex" dir="ltr">$ aRb\land bRc\rightarrow aRc $</span></dd>
</dl>
The centering decision is done within MathJax. To control the decision of MathJax the displaystyle attribute was proposed.
I think we should not start to change MathJax itself, because if we do so it will be hard to update to a new version. If there is a agreement to use the displaystyle attribute, and someone is willing to test that, we can merge that feature within a few days.--Physikerwelt (talk) 13:42, 8 February 2014 (UTC)
But things seem to have changed very recently. Possibly in the same update which caused the Texvc problem. Equations have been left aligned with MathJax until very recently, I've a screen shot from Oct 13 which shows this. --Salix alba (talk): 14:35, 8 February 2014 (UTC)
MathJax was updated from version 2.2 to version 2.3. Probably this caused the change. --Physikerwelt (talk) 15:28, 8 February 2014 (UTC)

How to display math with the next release

Hi, I feel sorry about the problems with the current version of the Math extension.

My goal was to move from the current version to a completely new version of the Math-extension aka Math 2.0 in one step. Unfortunately, this does work with the code review system used at MediaWiki, since it was a lot code and nobody was willing to do a code review for that. So I started to integrate the changes via continuous integration. That means all features from the list below are going to be integrated step by step. I turned out that this is not a good idea as well.

Since the new features interact with each other, it happened that parts were merged whereas other parts are still waiting for the code review. For example in the case of the broken align environments the adjustments to the PNG mode were in another feature as the general security improvement that was merged. A hotfix for that is waiting to get merged to the Misplaced Pages live version https://gerrit.wikimedia.org/r/#/c/112057/ I really want to get feedback for the planned changes and asked over and again on the Mediawiki mailing list. But I did not get much feedback. So I’d be more than happy if someone would like to help testing the changes before they break things in production environments. See http://www.formulasearchengine.com/review for a guide that I wrote to attract people for that task.

The discussions above show that I might find someone in this portal… If not… sorry for the spamming.

Changes of Math 2.0

  • MathML support with SVG fallback (pending)
  • Distinction between inline and display-style
  • MathJax now available independent of the rendering mode (pending)
  • Debug functionality (pending)
  • independent of the file-system (pending)
  • backwards compatible, without dependencies to it (pending)
  • removes dependencies to math specific core functions
    • getMathOptions() (merged)
    • armourMath (pending)
  • Asynchronous generation of Math images (to be discussed)
  • upgrade to MathJax 2.3, use the unpacked version and remove unusued files (merged)
  • use MediaWiki's resource loader for Javascript files (merged)
  • improve the way MathJax is configured (merged)
  • make math fonts available as Web fonts for native MathML (pending)
  • The MathML/SVG output is used as a preview while MathJax is processing (pending)
  • new table (mathoid) enables quick migration from Math1.0 to Math 2.0 (no schema change for math table required) (pending)
  • SVG generation is done by using MathJax on the server-side. (pending)
  • Link to idividual equations

See a live demo http://math-test2.instance-proxy.wmflabs.org/Fourier_series.

It would help us to help you to be a little clearer about what you are trying to do (as an overview), why you are trying to do it, how you are doing it, what the results should look like, when it is proposed to happen and what you want our help with. For example, the paragraphs PS31,...,PS61 make no sense to me. Deltahedron (talk) 12:27, 8 February 2014 (UTC)
Oh, and it would be nice to know how it happened that the very first that people at this project, such as myself, heard about all this was when it broke mathematics rendering. Earlier involvement would have been helpful. Deltahedron (talk) 12:29, 8 February 2014 (UTC)

This isn't really my area expertise, but the bug makes me wonder whether there is any regression testing done before such software or configuration changes go live. I mean just a regression test against a single project side such as Help:Displaying_a_formula presumably would have revealed problems as the current one.--Kmhkmh (talk) 12:42, 8 February 2014 (UTC)

That's a good idea. I'll add the test (I had already implemented that before.). The hard thing will be to get a code review for the test. --Physikerwelt (talk) 13:09, 8 February 2014 (UTC)
I created a bug about the type of the tests. --Physikerwelt (talk) 14:09, 8 February 2014 (UTC)
When I started to work on the Math extension of MediaWiki (about one year ago) the connection to the page Wikipedia_talk:WikiProject_Mathematics mathematics was not obvious to me. I just found that page today, while googling for a bug report at bugzilla. And I really appreciate the feedback.

I'm basically looking for some people who want to discuss about how the display of math can be improved in the future and to test the implemented features. I feel sorry about the mistakes I made in the past, but I think it's better to look forward rather than to complain about the past. I'll update the Roadmap that I started more than a year ago. And come back to this project page again after completing this task. --Physikerwelt (talk) 13:05, 8 February 2014 (UTC)

Its good to see the move towards making MathJax as the default. I'm quite happy to do some testing, and contribute to BugZilla, but only as an end user and not at the git/code review level (just because I know very little about it). If test version is released on one of the test wikis I'll happily look at that.
That's great to hear. We have to find a way how this testing can be done. The current worflow is that after the code-review the change gets merged to all live versions of Misplaced Pages. Maybe we can set up a common test environment (like described in the http://www.formulasearchengine.com/review). To validate the functional correctness of the change before doing the formal code review. (Manny aspects of code review are related to whitespaces and typos in comments and commit messages.). I'll think about that.--Physikerwelt (talk) 15:27, 8 February 2014 (UTC)
I've tried using http://math-test2.instance-proxy.wmflabs.org/Fourier_series but I can't log in so can only see it in SVG. I tried following the the log in instructions on the main page and 'There is no user by the name "MathJaX".'.--Salix alba (talk): 13:49, 8 February 2014 (UTC)
Thank you. You found the first error. The unser name is MathJax. I corrected that on the main page. --Physikerwelt (talk) 15:27, 8 February 2014 (UTC)
Cool, I can login now. The first obvious error is failing to parse some equations with 'Entity nbsp' not defined. Seem spaces inside \text( and ) get translated to   somewhere along the line.
I'm not really sure how things are different to what is on en.--Salix alba (talk): 16:19, 8 February 2014 (UTC)
I tried without logging in and search doesn't work: I got An error has occurred while searching: Error fetching URL: couldn't connect to host
At my first random page, Lundquist_number the subscripts and superscripts don't work: they just give spaces. Deltahedron (talk) 17:10, 8 February 2014 (UTC)
The same problem with sub/superscripts when I log in as MathJax. Typing into the box provokes a series of errors in the console at the foot of the page but the search then works. Rendering of letters in italic (text) and roman (display) seems inconsistent. Deltahedron (talk) 07:56, 9 February 2014 (UTC)
By the way, it would help us to help you if we knew where you want this sort of problem reported -- is there a centralised location for collecting comments and questions -- surely this subsection isn't it? Is there a list of known problems that don't need to be reported again? Is there a list of changes or proposed changes that you particularly want tested? Is there a list of things you've changed but which you're quite sure can't possibly affect the user experience? Is there a roadmap for future development and a schedule of proposed work? Where is all this information collated and made visible to other editors? Deltahedron (talk) 09:41, 9 February 2014 (UTC)
@Deltahedron: I'm open for proposal how to organize feedback and mange the todos. Yesterday I went throguh bugzilla there were 85 bug reports. Now there are 71. (Some of them were really old and already fixed.) I updated the bug List at the Roadmap. From what I see there are two option for the organization of the project. Either to user that roadmap page and maybe some subpages or to use Bugzilla. I saw some green and red buttons on Misplaced Pages buttons where user can give feedback. Maybe they could be useful to get a quick overview, which features are most wanted. I'll clean the math-2 test server and remove the experimental features that are connected to my PhD thesis about Formulasearch and are not supposed to be used at Misplaced Pages in the short term. PS: My prototype use case, to find the Jensen's_inequality by searching for ⟨ 1 x ⟩ ≥ 1 ⟨ x ⟩ {\displaystyle \langle {\frac {1}{x}}\rangle \geq {\frac {1}{\langle x\rangle }}} , is not reachable in the near feature, due to a lack of semantic information in formulae. --Physikerwelt (talk) 10:18, 9 February 2014 (UTC)
Perhaps it would help to create a page www.mediawiki.org/Extension:Math/Feedback and advertising it here, be appropriate for comments and discussion on such things as the roadmap at www.mediawiki.org/Extension:Math/Roadmap and experiments and experiences at math-test2.instance-proxy.wmflabs.org. Then advertise it widely, with a permanent link at places like WP:WPM and with a mention in the current debates at Misplaced Pages:Village pump (technical). Deltahedron (talk) 11:42, 9 February 2014 (UTC)

Physikerwelt, I just want to thank you for your work. I tried to do some work many years ago and I did not get much feedback from the developers then (it was all run less professionally then). Don't feel too bad about the align issue and don't give up. -- Jitse Niesen (talk) 17:14, 8 February 2014 (UTC)

  • I noticed the article Fermi–Pasta–Ulam problem has a messed up display, and when I looked at Help:Displaying a formula to see how to fix it, that is messed up too--I came here for help and see that others here have noticed the same thing. Is it possible to revert to the old version of the software for now, and debug the new code on a test server before deploying it to production? I believe the WMF does have some test servers available for this type of thing. Thanks. 70.36.142.114 (talk) 17:19, 8 February 2014 (UTC)
There is a fix for the bug that waits for review. I try to find someone who has the power to merge it.--Physikerwelt (talk) 18:23, 8 February 2014 (UTC)
@Physikerwelt: Is the following a valid summary of the current situation? As a security measure, and to get the same output regardless of method, you broke out the translation/normalization/macro-expansion part of texvc as texvccheck. The normalized output of texvccheck is then used as input for all the other output methods. This should have worked transparently except for the old misguided decision to name the multiline sub-environment "align" instead of "aligned". Misguided since if one considers the "math" tags as forming an outer math environment, everything inside "math" tags should be a sub-environment. And so the translation of texvccheck produces output "aligned" for input "align", which is correct for latex and mathjax, but not recognized by another texvc run (idempotence) to generate the png images, since texvc does not recognize "aligned" or "alignedat" as input environments. Is your fix making texvc accept them?--LutzL (talk) 18:44, 8 February 2014 (UTC)--Edit--LutzL (talk) 19:14, 8 February 2014 (UTC)
No to the last question, I see from the patch diffs that the second texvc run is given the original, untransformed tex input.--LutzL (talk) 19:20, 8 February 2014 (UTC)
That's correct. texvccheck is a subset of texvc... originally I wanted to call it texvc-light... but this name was not accepted. To explain it in a mathematical way. We call s {\displaystyle s} the securtity check and l {\displaystyle l} the latex conversion that texvccheck for input x {\displaystyle x} mans s ( x ) {\displaystyle s(x)} . Appling texvc to x {\displaystyle x} means l ( s ( x ) ) {\displaystyle l(s(x))} . The bug was that both, texvccheck and texvc were applied to x {\displaystyle x} and l ( s 2 ( x ) ) {\displaystyle l(s^{2}(x))} was displayed. s {\displaystyle s} is not idempotent which means that ∃ x : s 2 ( x ) ≠ s ( x ) {\displaystyle \exists x:s^{2}(x)\neq s(x)} . For example let's assumme that x 1 {\displaystyle x_{1}} contains a an align command than s ( x 1 ) ≠ s ( s ( x 1 ) ) {\displaystyle s(x_{1})\neq s(s(x_{1}))} since align is transformed to aligned which is not whitelisted. --Physikerwelt (talk) 22:14, 8 February 2014 (UTC)

The fix seems to have been deployed; I'm not seeing any problems. Ozob (talk) 22:34, 8 February 2014 (UTC)

I agree, the problems I was seeing with the align and alignat environments look to be fixed. Thanks again, Physikerwelt for getting on this and and getting the hotfix in! --Mark viking (talk) 23:06, 8 February 2014 (UTC)

Performance problems

Are the current performance problems also due to the upgrade? In several wikipedias previewing or editing articles with formulas takes very long at the moment. Articles with many formulas (such as List of mathematical symbols) even get a "504 Gateway Time-out" when you just try to preview them. Best wishes, --Quartl (talk) 06:32, 9 February 2014 (UTC)

Yes. I thought this was caused by the broken rasteriser but it still seems very slow despite that being fixed. I just tried loading Seven-dimensional cross product then unnecessarily purging it to force it to redo everything. Here (from the page source) is the timing information on the purge:
NewPP limit report
Parsed by mw1173
CPU time usage: 1.452 seconds
Real time usage: 26.816 seconds
Preprocessor visited node count: 3564/1000000
Preprocessor generated node count: 10682/1500000
Post‐expand include size: 50621/2048000 bytes
Template argument size: 3857/2048000 bytes
Highest expansion depth: 11/40
Expensive parser function count: 1/500
Lua time usage: 0.091s
Lua memory usage: 2.49 MB
26.8 seconds. And is isn't especially long or template heavy. The one thing it does have is a high density of formulae.--JohnBlackburnedeeds 06:45, 9 February 2014 (UTC)
I tried looking at List of mathematical symbols and that took a lot longer, though it did eventually load.
NewPP limit report
Parsed by mw1079
CPU time usage: 13.505 seconds
Real time usage: 308.243 seconds
Preprocessor visited node count: 11927/1000000
Preprocessor generated node count: 37075/1500000
Post‐expand include size: 217037/2048000 bytes
Template argument size: 90026/2048000 bytes
Highest expansion depth: 15/40
Expensive parser function count: 3/500
Lua time usage: 0.084s
Lua memory usage: 2.76 MB
This was only the first time accessing it. Reloading the page took a second. Obviously purging or editing would force it to redraw and so would take a similarly long time.--JohnBlackburnedeeds 07:05, 9 February 2014 (UTC)
I'm quite sure that this issue is connected to the new security feature. I followed the guideline, "you can not trust database entries" and check the user input text every time it's send to the rendering engine, even before the database lookup checks if the entry is already in the database. In local experiments this had almost no impact on the performance. Obviously the situation is different in production. I proposed to opt-out this new security feature via $wgMathDisableTexFilter = true on the bug report.--Physikerwelt (talk) 09:50, 9 February 2014 (UTC)
Out of interest, in what sense was this a "security feature"? Is there a specific security issue here or is this just belt-and-braces? Deltahedron (talk) 10:03, 9 February 2014 (UTC)
No there is no known urgend concrete threat, that can be caused by entering potentially dangerous input between <math> and </math>. But without this feature the possible input depends on the rendering mode, and we rely on the build in security of renderer i.e. MathJax or LaTeXML. Furthermore texvccheck provides a well defined set of allowed input, which could be discussed and modified in the future. However, I think that the drawbacks of slower editing are not compensated by the benefits of the well-defined and secure input independent of the rendering engine. --Physikerwelt (talk) 10:39, 9 February 2014 (UTC)

More on cyclic sieving

Several people have now contributed to Cyclic sieving. References and examples were added. I have now added a precise definition in a section labeled Definition.

There are still only three other articles that link to this new article. Michael Hardy (talk) 19:01, 8 February 2014 (UTC)

Help needed with Correlation coefficient and Transition function

Correlation coefficient and Transition function are currently disambiguation pages, and have been among the most-linked disambiguation pages for the past four months. It would be great if we could either get the incoming links to these pages fixed, or arrange them into freestanding articles. This will likely require some expert attention. Cheers! bd2412 T 04:16, 9 February 2014 (UTC)

I think Correlation coefficient would be better moved to Correlation coefficient (disambiguation), after which make Correlation coefficient into a redirect to Pearson product-moment correlation coefficient; that I think matches common usage especially outside of mathematics where it is widely used.--JohnBlackburnedeeds 11:35, 9 February 2014 (UTC)
Misplaced Pages talk:WikiProject Mathematics: Difference between revisions Add topic