User talk:David Eppstein/2021a

Logical fallacies
If some famous professor writes a draft paper and submits it to a journal in 2015, by 2020 it should either have been accepted, or the content doesn't belong in Wikipedia. What's the current status of this paper? Jehochman Talk 21:30, 31 December 2020 (UTC)
 * You obviously have no experience with the tedious and drawn-out process of getting journal versions of theoretical computer science papers refereed and published. Even in simple cases it often takes over a year, and I can point to cases like Baker's technique where it took over ten. In the meantime the result was published, in what counts as proper publication for the field, in STOC in 2016, and he had a follow-up paper published in STOC 2019. These are peer-reviewed publications; the standard for such conferences is that all papers get at least three in-depth reviews, and most submissions are rejected. Many theoretical computer scientists would have left it at that point rather than even making the effort to get an additional journal publication out of it. There are papers where the lack of a journal version has been problematic and has led to controversy over whether the results should be accepted as true; this is not one of them. Maybe if it is still not published another five years from now people might begin to wonder, but at this point the publication delay appears to be merely par for the field. As for "what is the current status of this paper": not transparent to us. Usually even the authors can only get answers to this question by repeated and severe nagging of the editors handling the submission. —David Eppstein (talk) 21:34, 31 December 2020 (UTC)


 * I lost you at "You obviously." Try to be more polite if you want to convince somebody of something. If there's some long drawn out process, that needs to be explained in the article.  Otherwise it looks weird for some five year old, unaccepted paper to be there.  No doubt there's advanced math and set theory and it's hard to verify the proof. Jehochman Talk 04:29, 1 January 2021 (UTC)
 * No doubt. But when you're ignorant of the basic publication norms of a field, come blustering in thinking that this sort of delay must have some great meaning that it doesn't have, and based on that incorrect assumption start removing information that is of great significance to the article, a little condescension seems in order. And no, the article on graph isomorphism is not the place to explain typical review time-periods in theoretical computer science journal publication, even if published sources on that topic could be unearthed. —David Eppstein (talk) 05:39, 1 January 2021 (UTC)


 * David, could you send me an email? I’d like to forward you a message from one of the foremost living theory of computation authorities who I consulted.  The finding has not been proven or accepted yet.  Wikipedia needs to frame this carefully. Jehochman Talk 01:23, 2 January 2021 (UTC)
 * That's...an infelicitous way of phrasing it, one that misses a lot of subtlety. The proof has been written and made available, so saying it "has not been proven" as an absolute statement is inaccurate. The proof has been publicly stated to be accepted, by Harald Helfgott, who made a careful study of it and was the one who found the bug in the earlier version, so saying it "has not been accepted" as an absolute statement is equally inaccurate. It has also been published, in a preliminary (and pre-bugfix) but peer-reviewed form, in STOC 2016, and in still-preliminary but post-bugfix form in the proceedings of the ICM, so saying it has not been peer-reviewed or that it has not been published would also be inaccurate. It has not been published in full detail in an archival journal form, and until it has it should be considered as a preliminary announcement rather than as an accepted result. We can certainly say something of that nature in the article. But it is widely expected to be valid (different from accepted as being valid) and ripping it out entirely as you did as being "unverified speculation" is nonsensical, wrongheaded, mistaken, and detrimental to the encyclopedia. As for forwarding to me emails from respected authorities: what was it you were saying in the first sentence of your first comment above about appeals to authority? —David Eppstein (talk) 02:33, 2 January 2021 (UTC)
 * Since you seem to like authority I thought you'd enjoy hearing what mine had to say. You'd be surprised how often a professor will respond to a friendly message asking them for advice how to improve Wikipedia.  Anyhow, I think the announcement can stay in the article, so long as we present it correctly.  I'll go make an edit, and if you want to revise, please do. Jehochman Talk 18:25, 2 January 2021 (UTC)

Happy new year and a question about John Conway
I see you worked on Ronald Graham and wondered if you would be interested in helping me to improve John Horton Conway to GA status. I started thinking about this because Graham, Conway, and Dyson were just mini-profiled in The New Yorker. Have a good 2021 either way. HouseOfChange (talk) 16:45, 1 January 2021 (UTC)
 * I can make minor edits if necessary on Conway's article and would be happy to see it reach GA status but it's not really one of my priorities right now. So I wish you well on your project but I don't think you can count on me for a lot of effort on it. If there's some specific part of the article that you think I have expertise in and need help with, though, please let me know. —David Eppstein (talk) 17:46, 1 January 2021 (UTC)

Your GA nomination of De quinque corporibus regularibus
Hi there, I'm pleased to inform you that I've begun reviewing the article De quinque corporibus regularibus you nominated for GA-status according to the criteria. This process may take up to 7 days. Feel free to contact me with any questions or comments you might have during this period. Message delivered by Legobot, on behalf of Ján Kepler -- Ján Kepler (talk) 11:01, 2 January 2021 (UTC)

Ken Parsons, Professor
Re your comment about revising the article to be more neutral and encyclopedic; is there a particular section (or sections) that you think should be revised? Guidance appreciated. EEParsons (talk) 21:59, 3 January 2021 (UTC)
 * (This is re new article Ken Parsons, Professor.) Your user name suggests that you have a conflict of interest and should not be touching the article at all. But for starters, is it really the case that his full name is "Ken Parsons Professor Professor Ken Parsons"? Sort of like "Bond James Bond" only even more repetitive? (see MOS:DR: usually we omit titles such as Dr. or Prof. altogether.) As for my "like resume" tag: that was directed at the apparently indiscriminate listing of all positions held and of publications (not described as selected); some of the important differences between an encyclopedia article and a cv are that we try to tell a story about the life rather than a bare listing, we generally omit the unimportant details such as editorial board memberships and consulting gigs, and we base everything on what we can find in reliable published sources rather than using our own knowledge of the subject. —David Eppstein (talk) 22:19, 3 January 2021 (UTC)

A barnstar for you!

 * Thanks! —David Eppstein (talk) 17:21, 4 January 2021 (UTC)

Constance van Eeden
You may (or may not) be correct, I have updated the Reference_desk/Language, We'll see what they say ! GrahamHardy (talk)
 * You were right ! GrahamHardy (talk) 21:31, 5 January 2021 (UTC)
 * Sometimes it happens. No problem double-checking, though. —David Eppstein (talk) 21:36, 5 January 2021 (UTC)

Legobot rollback
Did you mean to rollback Legobot at GA? Best, Barkeep49 (talk) 04:46, 6 January 2021 (UTC)
 * Uh, no? I hadn't even realized I did it. Must have been a mouse slip. Sorry about that, Legobot. Fortunately rollbacks there are completely ineffective — it just kept on doing it again. —David Eppstein (talk) 05:30, 6 January 2021 (UTC)

Your GA nomination of Curve of constant width
Hi there, I'm pleased to inform you that I've begun reviewing the article Curve of constant width you nominated for GA-status according to the criteria. This process may take up to 7 days. Feel free to contact me with any questions or comments you might have during this period. Message delivered by Legobot, on behalf of Ovinus -- Ovinus (talk) 22:00, 8 January 2021 (UTC)

Pending AFD nomination
Hey Dave, i need your help with this one. Articles for deletion/Deviprasad Dwivedi is lying untouched without any comments/votes for the 6th straight day. Although i know that this is fairly common and we should remain patient, my query is that if i can do something to increase the traffic there. This is helpful considering that we are basically decreasing the workload of potential voters by taking the discussion to them. Pesticide 1110  Lets wrestle! 17:09, 9 January 2021 (UTC)

Note
I hope my reply in Articles for deletion/Mo Moulton is not interpreted harshly, I did not intend it that way.  // Timothy ::  talk  08:30, 10 January 2021 (UTC)
 * It depends on whether you mean "harshly" as that I might think I was being insulted, or whether you mean that my opinion of you will be lowered by reading what you wrote. —David Eppstein (talk) 08:34, 10 January 2021 (UTC)
 * It was simple note expressing that my tone was not intended to be harsh, the kind most editors would reply to graciously, not with a backdoor insult. WP:CIVIL, "Differences of opinion are inevitable in a collaborative project. When discussing these differences some editors can seem unnecessarily harsh, while simply trying to be forthright."  // Timothy ::  talk  09:20, 10 January 2021 (UTC)

Sylvester-Gallai theorem reversion
Hello again.

The article needs a down-to-earth way of explaining what the complication is in constructive mathematics. Citing the particular weak form of excluded middle (which is what was there until my recent edits) is mumbo jumbo to most people who might land on the page. The most WP:COMMON-ly understood of the many implications of LLPO is the halting problem. Another possibility is to add some words about real number in that context being an infinite-precision object, or giving the example of a real number whose base 2 expansion is any given sequence, so that deciding if it is zero functions as a halting solver. If in computational geometry "real number" means something else and this is a non-issue (though the examples in Mandelkern's article are simple enough to have complexity-theoretic implications of the same flavor) it seems to me the solution is to add that information rather than reduce clarity about the constructive issues. 73.89.25.252 (talk) 20:41, 11 January 2021 (UTC)
 * Saying it's as hard a problem to compute as the halting problem, as you did, is flat-out incorrect, because that's a statement that is not qualified by assumptions of constructivity. Even in constructive mathematics, inputs with integer coordinates (a very standard assumption) are completely unproblematic. Your reference to the halting problem was unsourced and appears to be original research. And it appears to be making unwarranted assumptions about what an algorithm can do: even if we can't tell for some lines whether they are ordinary or not, one could easily run in parallel an algorithm that tests each line for being ordinary and returns the first one it finds. It would need a guarantee that its input is non-collinear (because that part really is hard to test constructively), but your statement was about algorithms for finding the ordinary line, not for testing whether one exists. And it might not be a function (the output of the algorithm would depend on the specification of the input and not just on its geometry) but so what? —David Eppstein (talk) 20:46, 11 January 2021 (UTC)
 * The S-G theorem is about plane geometry (i.e., real or perhaps real-closed coordinate field), not the more restricted input models in computational geometry, so if any statements in the article are false or misleading for computational geometry then the restricted models should be explained as an alternative context, not as overturning statements that are true for real or constructive-real inputs.
 * The computational geometry form of Mandelkern's results is that if coordinates are given in (e.g., rational) interval arithmetic, then it can happen that for no line in the configuration is there a proof that it goes through only two points. The constructive real model is exactly that but with the ability to compute any coordinate to whatever precision is desired.  Maybe the explanation in the article could be given in these terms without discussing LLPO at all. 73.89.25.252 (talk) 21:20, 11 January 2021 (UTC)
 * Your statement that constructive mathematics should be viewed as the default method for understanding geometric algorithms, counter to everything done by every computational geometry researcher ever, is noted but dismissed as WP:FRINGE. And your continued failure to provide published sources for your assertions is also noted. As for "it can happen that for no line in the configuration is there a proof that it goes through only two points": yes, but only for inputs that are hard to distinguish from being totally collinear. Otherwise, the ordinary line exists (classically) and has nonzero distance from all other points, so one can simply continue computing until reaching a level of precision sufficient to distinguish that distance. So the difficulty of computing has nothing to do with actually finding ordinary lines, but is really only about determining collinearity of point sets, something that is only tangentially related to this article. —David Eppstein (talk) 21:30, 11 January 2021 (UTC)
 * constructive mathematics should be viewed as the default method for understanding geometric algorithms -- where did I say that? (You call it "(my) statement"!)
 * In a section of the article devoted to the status of the S-G theorem in constructive mathematics, cited solely to an article by Mark Mandelkern (apparently a constructivist from his articles) about a Brouwerian counterexample to the classical statement of S-G, then, uh, very obviously the context for whatever is explained there is... constructive mathematics, and its relation to classical mathematics and classical S-G. None of which are computational geometry (your area of expertise).  Certainly I agree that adding material on S-G in computational geometry would be a big improvement to the article.  But the edits in question and this here discussion thereof do not directly concern computational geometry.  You are raising objections based on a computational geometry setting that is not currently in the article; to repeat, I think it would improve the article to add such material.  But if the constructive business is to stay in the article, then how to present that is a matter of its own unconnected to comp.geom. 73.89.25.252 (talk) 21:45, 11 January 2021 (UTC)
 * Certainly I agree that adding material on S-G in computational geometry would be a big improvement to the article: Uh, it's already there. You did read the article before making additions to it, right? —David Eppstein (talk) 21:52, 11 January 2021 (UTC)
 * To be excessively precise, as I think it was obvious (from words such as "section of the article devoted to the status of the S-G theorem in constructive mathematics", and all edits under discussion being in that section), I have been talking about the paragraph on constructive S-G, in the subsection on Axiomatics, in the section Proofs.   That paragraph, section and subsection say nothing about the computational geometry of S-G, in particular the types of input for which an algorithm is or is not guaranteed to succeed -- an issue also not discussed anywhere else in the article, such as the sections on algorithmics.  The addition of material on that issue, such as your remarks here about collinearity testing, to that section of the article would be useful but would probably belong in an additional paragraph on CG distinct from the context of constructive mathematics. 73.89.25.252 (talk) 22:08, 11 January 2021 (UTC)

Math problem...
Hi, it's true math isn't my strong suite, and I was trusting in the accuracy of the article to craft my description. Also, I did check the revision history to see how long it had been since anyone edited the lead, and I figured that surely any mistakes would have been corrected since November, so I decided to proceed with my faith in the accuracy to guide me in lieu of my knowledge of mathematics. I thought I did enough due diligence, but I guess not enough people are keeping up with these math articles.

Thanks for catching the error and fixing it, but boo on you for being so eager to point out personal weaker areas with demeaning public remarks such as "ridiculous short description added in equal ignorance". While that might not be untrue, it was an honest mistake not really deserving of humiliation, and it's just not kind.

That being said, you might look at another one I wasn't 100% sure about: Möbius_function I think it's actually ok since the lead doesn't appear to have been edited in a long time, and that is where I always get my material for descriptions, if I have knowledge about the article or not. Thanks again for your help. Huggums537 (talk) 06:55, 18 January 2021 (UTC)
 * That one looks mostly accurate enough, although I think this is really just number theory without the combinatorics. However, too much of the editorialization ("important") has been carried over into the shortdesc, which really only needs the factual part (it's a multiplicative function, and its context is number theory). —David Eppstein (talk) 06:59, 18 January 2021 (UTC)
 * Thanks, I shortened the description to reduce the "editorialization". Huggums537 (talk) 07:08, 18 January 2021 (UTC)

David Eppstein, I think I'm getting conflicting advice from different editors, where you have advised me in this edit summary to be less "inane", and Trovatore has advised in this edit summary that simply adding "Mathematical concept" is sufficient enough context for distinguishing that title from other titles that might pop up in search results with the same name. (This coincides with the goals of the short description project as well.)

David, you seem to posit here that it is not titles with similar names that need to be distinguished, but rather titles that have similar topics that need to be distinguished, and I have to tell you that the people over at Wikipedia talk:Short description have been insisting it is only the titles, but I think I partially agree with you here by saying it should be both the titles and the topics. I believe it would improve the descriptions to distinguish topics as well.

I will admit I have no idea what I was thinking when I introduced "Axiom" into the description, if I had it to do over again, I would simply make the Probability space description say, "Model of random process". However, that is from the perspective of a layman, and you two are the experts here.

I'm doing exceptionally well just to be able to edit on this site with my highest academic achievement being that I completed 8th grade Jr. high school, then did some freshman high school before going to get my GED. I have studied some college texts just for my own personal amusement, but no formal training. Huggums537 (talk) 22:07, 26 January 2021 (UTC)
 * Your shortdesc for additive function failed to distinguish it from articles with similar titles, not just similar topics, including function and (more specific to number theory) multiplicative function and Partition function (number theory) among others. As for your education, I am certainly not one to require credentials — it impresses me when people without them reach the same level of accomplishment, because to me that shows that they were able to do it the hard way. But I think the Wittgenstein quote "Wovon man nicht sprechen kann, darüber muss man schweigen" may be relevant. —David Eppstein (talk) 22:13, 26 January 2021 (UTC)
 * My understanding of short descriptions is that they are mainly intended to give broad context to mobile users (to take a recent example, a mobile user who sees "probability space" in a list of search results shouldn't have to wonder if it might be space between Mars and Jupiter). I don't think we need to make distinctions much finer than that.  Short descriptions are supposed to have a "soft limit" of 40 characters, and while it's fine to go over that when it's really necessary, I think we should try to hold the line on that whenever reasonably possible. --Trovatore (talk) 23:11, 26 January 2021 (UTC)
 * It's the first thing shown to all readers of our articles, both mobile and non, after the title but before even the initial sentence of the article. As the first thing they see of a topic, it's likely the take-away message of the article, what they will remember the topic as being about. It's also used in mobile to list search results. As such it needs to be readable to a broad audience, short and to the point, and capable of telling searchers which of multiple similar results is the one they're looking for. Something vague but correct (like, "it's a mathematical concept" or "it's a Wikipedia article") isn't helpful, but neither is copying all of the technical detail from the start of the article. It's important that people writing short descriptions have a clear understanding of both the topic they're writing the description for, and of other topics that might appear with it in search results: so they can write something that accurately describes the topic and that distinguishes it from the others. That's why I think it's a mistake for editors to take on writing short descriptions of technical topics as a gnome-like task, something like minor grammar corrections that can be done en masse to thousands of articles just by quickly reading the first sentence and trying to summarize it (or punting and writing "concept in [category name]" when even summarizing a sentence is too much work). Sometimes, if that first sentence has already been written well, abbreviating it into a short description works. Often, it doesn't, and the editor writing the short description needs to be capable of telling the difference. —David Eppstein (talk) 23:23, 26 January 2021 (UTC)
 * Hmm? Web users don't see it at all, unless they enable it in CSS. --Trovatore (talk) 23:38, 26 January 2021 (UTC)
 * It's certainly visible to me, and I don't recall doing anything special to enable it. But maybe I did and I just don't remember. —David Eppstein (talk) 23:39, 26 January 2021 (UTC)
 * In any case I agree that "Wikipedia article" is useless, but I think "mathematical concept" is fine for most math articles named after a kind of object. It isn't supposed to be a definition.
 * I don't object to more specific short descriptions as long as (1) they don't go over 40 chars and (2) they don't confuse math-naive readers, who mostly just need to know that it's a math article, which may not be obvious. --Trovatore (talk) 23:50, 26 January 2021 (UTC)
 * Why 40, specifically? When I try the mobile search I get over 50 characters per shortdescription line and many are two lines. Also, (3) when math-savvy readers are trying to search for an article on mobile, all they see are the titles and shortdescs, so it needs to be informative to them as well. —David Eppstein (talk) 00:00, 27 January 2021 (UTC)
 * 40 is the "soft limit". Unless that has been changed.  I haven't really kept up with the shortdesc project.  I think we specifically want them to stay on one line whenever possible.  That's just for UX reasons.
 * I don't think short descriptions were ever really meant to cater to experts. Just not really what they're for.  I'm pro-power-user myself, but I don't think we need to make every feature a power feature, and this one is a good example of one that shouldn't be, unless it can be done without making it less useful to non-experts. --Trovatore (talk) 00:08, 27 January 2021 (UTC)
 * So you want to make people searching for mathematics articles using mobile guess which article they want based only on titles and inane "article in mathematics" shortdescs, rather than even trying to make mobile search useful for the people who want to find those articles? Have you ever actually tried using mobile? —David Eppstein (talk) 00:12, 27 January 2021 (UTC)
 * David, I think Trovatore has only tried explaining how the guidelines at the short description project seem to have wanted us to do it so far, and I think I tend to agree with your arguments that the way they seem to want us to do it needs improvement so we can write more informative descriptions. However, the rebuttal argument I have seen at the project that is against us follows something along the lines that the more informative description is not needed to distinguish the titles because the titles are always together with the descriptions, and the titles already do the job of distinguishing themselves anyway, e.g. additive function, multiplicative function, and Partition function (number theory). While this may be true, I still think the informative descriptions are of more benefit to the reader, so it's a pointless truth to focus on in my opinion. Please note the descriptions do not show up on my mobile in the articles, just in the search results as they were intended to. [ I'm positive I had to enable them in my settings for my PC too since they don't show up in any of the level 1-3 vital articles that for sure have descriptions when I view incognito/private mode.] Also, I don't want to leave the wrong impression that I'm some uneducated buffoon who quit learning after high school, since I did continue on to acquire both formal and informal education or training for skillsets (trades) outside of the academic realm (university), and I can do these technical descriptions. I've done a bunch of them with relatively few errors considering how many I have completed, although I will admit a couple of the errors were rather large blunders, but I can be more careful with that. Huggums537 (talk) 07:36, 27 January 2021 (UTC)

Disambiguation and Wikidata questions
Hi User:David Eppstein. I am working on a page on Marilyn Burns, a prominent math educator who won the AWM Hay Award in 1997. There are already pages for Marilyn Burns and Marilyn Burns (politician). Is it kosher to create at page for Marilyn Burns (mathematics educator) or should I use Marilyn Burns (mathematician)? I am just learning how to input data on Wikidate. I know how to add "Statements" to sections that already exist. How to I add a section "Identifiers"? I added a VIAF ID to Marilyn Burns' page on Wikidata but that didn't activate a section on "Identifiers". Thanksfor your help. Mvitulli (talk) 18:07, 21 January 2021 (UTC)
 * Either title looks ok to me. The difficulty with Burns is making an article that doesn't look like an ad for her consulting business. As for the identifier section on Wikidata: just add statements listing the identifiers (VIAF, MGP, etc) and Wikidata will automatically sort them into a section for you. If you didn't see it when you added it, you probably just need to reload your browser window. —David Eppstein (talk) 18:16, 21 January 2021 (UTC)
 * User:David Eppstein I just published the page for Marilyn Burns (mathematics educator). See what you think.  Even though I added identifiers to Wikidata first, nothing shows up in authority control.  I tried dummy edits.  I'm stumped.  Mvitulli (talk) 18:58, 21 January 2021 (UTC)
 * For authority control, usually you have to wait a few minutes and/or purge the page (Special:Purge). Dummy edits are just a placebo to help you wait. I'm seeing the authority control now. As for the new article, the WP:PEACOCK language like "highly regarded", the Amazon sales links for her books (both removed now), and promotional blurbs for her company sourced only to her company web site (all removed now) are exactly the sort of thing I was warning against in making the article look too promotional (and therefore likely to come under scrutiny by deletionists opposed to promotional material). —David Eppstein (talk) 19:57, 21 January 2021 (UTC)

211.116.85.179
Hi David Eppstein, I have reduced the block duration of 211.116.85.179 to 2 years, especially as it is a hardblock. Best regards, ~ ToBeFree (talk) 03:52, 24 January 2021 (UTC)
 * It was a block for a single article that the IP editor has been making disruptive edits to for months, as their only edits, and that they started right up again on as soon as their previous (full rather than article-specific) three-month block expired. If it were a full block, I would have chosen a year, but I think for an article-specific block on an article they have only ever been problematic on, indef is appropriate. I hope you have watchlisted the article and plan on being around 2 years later to handle the issue when they come back again. —David Eppstein (talk) 04:37, 24 January 2021 (UTC)

Your GA nomination of Curve of constant width
The article Curve of constant width you nominated as a good article has passed ; see Talk:Curve of constant width for comments about the article. Well done! If the article has not already appeared on the main page as a "Did you know" item, or as a bold link under "In the News" or in the "On This Day" prose section, you can nominate it within the next seven days to appear in DYK. Bolded names with dates listed at the bottom of the "On This Day" column do not affect DYK eligibility. Message delivered by Legobot, on behalf of Ovinus -- Ovinus (talk) 08:42, 28 January 2021 (UTC)

Your GA nomination of Ronald Graham
Hi there, I'm pleased to inform you that I've begun reviewing the article Ronald Graham you nominated for GA-status according to the criteria. This process may take up to 7 days. Feel free to contact me with any questions or comments you might have during this period. Message delivered by Legobot, on behalf of Ovinus -- Ovinus (talk) 09:20, 28 January 2021 (UTC)

Your GA nomination of Ronald Graham
The article Ronald Graham you nominated as a good article has passed ; see Talk:Ronald Graham for comments about the article. Well done! If the article has not already appeared on the main page as a "Did you know" item, or as a bold link under "In the News" or in the "On This Day" prose section, you can nominate it within the next seven days to appear in DYK. Bolded names with dates listed at the bottom of the "On This Day" column do not affect DYK eligibility. Message delivered by Legobot, on behalf of Ovinus -- Ovinus (talk) 22:22, 29 January 2021 (UTC)

A barnstar for you!

 * You're welcome! —David Eppstein (talk) 19:50, 2 February 2021 (UTC)

Creating a template
Hi User:David Eppstein. I just create the Template:AWM Presidents modeled on Template:AMS Presidents. The template doesn't display the same way that Template:AMS Presidents does. Do you know how to fix this? Thanks for your help. Mvitulli (talk) 23:49, 2 February 2021 (UTC)
 * Fixed. The problem was the spaces in front of the stars. The stars need to be the very first character in each line of text. —David Eppstein (talk) 00:01, 3 February 2021 (UTC)
 * WOW! Thanks.  Mvitulli (talk) 00:04, 3 February 2021 (UTC)

Ricky Allman AFD
Could you please review Ricky Allman AFD one more time. I have added new information, such as museum exhibitions and Harvard Business Review citation. I feel these may change your mind.Webmaster862 (talk) 05:25, 3 February 2021 (UTC)

Grötzsch graph
Regarding your reversion of my edit in Grötzsch graph: If you start with the null graph (which I assume is the graph with no vertices) and repeatedly apply the Mycielskian, you get graphs with 1, 3, 7, 15, ... vertices. You never get the Grötzsch graph, which has 11 vertices. (Applying the Mycielskian gives twice the number of vertices plus one.) Or did I misunderstand something? /Pontus (talk) 18:16, 3 February 2021 (UTC)
 * You start with a graph with one vertex and zero edges. Maybe we should say one-vertex graph instead of null graph to be less ambiguous. —David Eppstein (talk) 18:18, 3 February 2021 (UTC)
 * But then the following graphs will have 3, 7, 15, 31, ... vertices. /Pontus (talk) 18:21, 3 February 2021 (UTC)
 * Also, in the article on Mycielskian it says that the initial graph is the one-edge graph. /Pontus (talk) 18:28, 3 February 2021 (UTC)
 * Ok, you're right. Mycielskian always has an odd number of vertices so to get two vertices you need the one-edge graph as a start. It's definitely incorrect to call this the null graph. I'll restore your edit. —David Eppstein (talk) 19:06, 3 February 2021 (UTC)
 * Ok, thanks. Keep up the good work! /Pontus (talk) 19:21, 3 February 2021 (UTC)

Nikitha Grero
Hi. I don't get that. How you cite that he is non notable?? As a Sri Lankan and a native person, he is well adjudged in our country due to his leadership qualities and popularizing the international school system in Sri Lanka. I have added a hardcopy of his notability through a newspaper. So, my concern is "HE is notable" and valuble asset in our country. So help me to keep the article. If you need soft article references, I will send them soon to Wikipedia. There are many non notable things in Wikipedia, but never ever get deleted and never concerned by Admins. Thank You. Gihan Jayaweera (talk) 21:41, 5 February 2021 (UTC)

Proposed deletion of Timothy Budd


The article Timothy Budd has been proposed for deletion&#32;because of the following concern: "NN academic"

While all constructive contributions to Wikipedia are appreciated, pages may be deleted for any of several reasons.

You may prevent the proposed deletion by removing the notice, but please explain why in your edit summary or on the article's talk page.

Please consider improving the page to address the issues raised. Removing will stop the proposed deletion process, but other deletion processes exist. In particular, the speedy deletion process can result in deletion without discussion, and articles for deletion allows discussion to reach consensus for deletion. Apologies for the template. Bearian (talk) 19:14, 5 February 2021 (UTC)
 * If I had wanted to prevent deletion of this article, I would have addressed and removed the notability tag, instead of merely tweaking it so that it pointed to the correct notability guideline. —David Eppstein (talk) 19:50, 5 February 2021 (UTC)

Template:Did you know nominations/Kristoffer Domeij
Thanks for the clearance. It needs a tick before it can become airborne. 7&amp;6=thirteen (☎) 20:49, 6 February 2021 (UTC)

DYK for Ronald Graham
Cwmhiraeth (talk) 12:02, 7 February 2021 (UTC)

Bed Hangings Article Question
Hello @David Eppstein, I am contacting you about a set of changes you made and still have concerns about on the Bed hangings article that I've done a lot of work on. I would like to know what I should try to do to fix the undated ones that you indicated are still broken. This was the first time I used the sfn style (at the suggestion of an administrator), and I clearly have a lot to learn. If there is any guidance you could give me, I would be most appreciative. Thank you. TrudiJ (talk) 18:00, 9 February 2021 (UTC)
 * I don't see any broken links currently. I make these errors visible to me using the script at User:Ucucha/HarvErrors. —David Eppstein (talk) 20:19, 11 February 2021 (UTC)

Song of the South
Your recent editing history at Song of the South shows that you are currently engaged in an edit war; that means that you are repeatedly changing content back to how you think it should be, when you have seen that other editors disagree. To resolve the content dispute, please do not revert or change the edits of others when you are reverted. Instead of reverting, please use the talk page to work toward making a version that represents consensus among editors. The best practice at this stage is to discuss, not edit-war. See the bold, revert, discuss cycle for how this is done. If discussions reach an impasse, you can then post a request for help at a relevant noticeboard or seek dispute resolution. In some cases, you may wish to request temporary page protection.

Being involved in an edit war can result in you being blocked from editing&mdash;especially if you violate the three-revert rule, which states that an editor must not perform more than three reverts on a single page within a 24-hour period. Undoing another editor's work—whether in whole or in part, whether involving the same or different material each time—counts as a revert. Also keep in mind that while violating the three-revert rule often leads to a block, you can still be blocked for edit warring&mdash;even if you do not violate the three-revert rule&mdash;should your behavior indicate that you intend to continue reverting repeatedly. KristofferR (talk) 01:39, 11 February 2021 (UTC)
 * I think you know better than to template the regulars, and I'm sure you also know that a bold edit, once contested through reversion, should be discussed before restoration of the contested material to the article. Propose your addition at Talk:Song of the South and obtain a consensus for its inclusion. BD2412  T 01:54, 11 February 2021 (UTC)

Draft:Judith Eisen
Hi User:David Eppstein. There is already a page for this person, Judith S. Eisen, that I created in November. The page has been reviewed. I will notify the writers of the draft on their talk page. Mvitulli (talk) 21:04, 11 February 2021 (UTC)
 * Ok, I redirected the draft to the article. —David Eppstein (talk) 21:12, 11 February 2021 (UTC)
 * User:David Eppstein BTW Judith Eisen was contacted by Tasmin S. Kelly who offered to rewrite her draft article and get it approved. I see that Kelly offers to fix pages for money.
 * Well that's sleazy. If you're in contact with Eisen you might tell her that we tend to frown on paid editing, and that taking up such offers is likely to cause her article to fall under suspicion. —David Eppstein (talk) 21:21, 11 February 2021 (UTC)
 * I think she realizes this and that's why she notified me that she was contacted. She did not reply to Kelly. Mvitulli (talk) 21:25, 11 February 2021 (UTC)

A barnstar for you!

 * You're welcome! —David Eppstein (talk) 00:47, 17 February 2021 (UTC)

Feedback on citation metrics
Hi, would you be able to give me some feedback on my NPROF C1 citation analysis method? I've gotten some pushback lately but most of the criticism seems to apply just as much to any other citation evaluation approach. I developed my coauthors (and coauthors-of-coauthors) heuristic because I felt a lot of AfDs were decided on individual editors' overly-generalized perceptions of what "highly-cited" means, despite wide variation between fields. My intent is to create a much more personalized and in-depth comparative assessment of someone's contributions to their field (for C1 citation analysis only), so I'd welcome comments on how to improve it. Thanks, JoelleJay (talk) 22:33, 19 February 2021 (UTC)
 * I'm not sure. It's definitely a good idea to calibrate impressions of citation record strength based on the typical numbers for the same field. On the other hand, these numbers are not really very meaningful at the best of times and getting into the weeds trying to extract the	#9400D3 last drop of information from them is probably a mistake. But if you want to keep putting the effort in, I won't complain. —David Eppstein (talk) 07:25, 20 February 2021 (UTC)
 * Hmm, ok. Why do you feel the citation record isn't meaningful for C1 assessment? I frequently choose to do analyses when AfD discussions are centered around citations due to other criteria not being met, in which case it seems the fairest method would be to do field calibrations. In the situations where other criteria are in play I usually just leave the analysis as a comment rather than !vote. I figure since I have access to Scopus through my university and it's much more reliable than GS for most science disciplines, the data would be more accurate. JoelleJay (talk) 17:23, 23 February 2021 (UTC)
 * It's really the only way for an outsider to an area to do C1 assessment, but papers can be highly cited for a lot of reasons, not all of which are that they are important research contributions. And the emphasis on citation counts and impact factors as a method of evaluation within academics over the last few years has meant that those numbers have become distorted, as people do work aiming for better numbers rather than aiming for better work; see Goodhart's law. —David Eppstein (talk) 17:28, 23 February 2021 (UTC)

MOS:COLOR
Hi David, Thanks for educating me in colour contrast guidelines. If I use DarkViolet	#9400D3 for the purple colour, will it be fine? Thanks, cm&#610;&#671;ee&#9094;&#964;a&#671;&#954; 11:32, 20 February 2021 (UTC)
 * Snook's Colour Contrast Check (linked from MOS:COLOR) says yes: Dark violet on a white background passes WCAG 2 AA. (It doesn't pass at the AAA level but we don't require that.) I don't think it's capable of checking whether the three colors you picked look different from each other to color-blind people, though; there might be a different checker somewhere for that. —David Eppstein (talk) 18:42, 20 February 2021 (UTC)
 * Good to know that AA is the level we're targeting. I'm aware of different forms of colour-blindness and distinguishing red-yellow-green covers over 99.97% of the sighted population. I thus try to use red-purple-blue-grey schemes where understanding solely depends on colour. cm&#610;&#671;ee&#9094;&#964;a&#671;&#954; 02:13, 21 February 2021 (UTC)

Christina Kendziorski revert
My apologies. My edit to Christina Kendziorski to add an approximate birth date was in error. I intended to make it c.1970, on the basis that her Bachelor degree was achieved (per the article w/reference in 1992), and a person generally achieves that degree around the age of 22, hence 1992 less 22 is 1970. I failed to notice my typo of c.1980, which of course is totally wrong. Many thanks for catching my error. Would it be appropriate, and in accordance with MOS:CIRCA, to edit her approximate birth at c.1980? Truthanado (talk) 01:22, 21 February 2021 (UTC)
 * No. We don't guess on biographies of living people; we need published sources for all claims there. See WP:BLP. Additionally, in cases where the subject has not made their birthdate widely known, it is common to respect their privacy and not include it, even when less-widely-known but published sources can be found; see WP:BLPPRIVACY. —David Eppstein (talk) 01:58, 21 February 2021 (UTC)

dr eppstein i have math question
dr higham hasn't gotten back to me (*sorta* surprising), so i thought mb i'd ask you.

i know if we have an n x n symmetric matrix it will have n real eigenvalues, and that the sum of the trace is the sum of the resulting eigenvalues.

but i want to say something a little stronger:

if i have such a matrix whose diagonal values are strictly positive, then can i say anything about the resulting eigenvalues also being strictly positive? 198.53.159.44 (talk) 20:04, 21 February 2021 (UTC)
 * Not my forte, sorry. And if you're looking to add this somewhere on Wikipedia, you'll need sources, not just another editor's opinion. —David Eppstein (talk) 20:05, 21 February 2021 (UTC)
 * naw, not looking to add it. just wanna know if one of the brains on here could prove this. it seems like it COULD be true, but my intuition suggests that it's not so simple. i shall bug some other people! 198.53.159.44 (talk) 20:57, 21 February 2021 (UTC)
 * It is not true; this is already easy to see in the 2-by-2 case, where you can choose the off-diagonal entries to be arbitrarily large and force the determinant to be negative. --JBL (talk) 21:18, 21 February 2021 (UTC)

Was wondering if you could give me your opinion on the Notability of someone
Can you take a look at Bulusu Jaganadha Sastry and let me know what you think? I'm not super familiar with the academic notability criteria, but I recently nominated the article on his brother for deletion and it doesn't look like his case for notability is much better. It should be noted that the author of the articles was a relative, who also wrote a since-deleted article about himself and an article about his family tree (Bouloussou family, in which he added himself). Best, GPL93 (talk) 22:06, 24 February 2021 (UTC)
 * Doesn't appear to pass WP:PROF, but that's not unusual because many academics from that time period wouldn't — it's aimed more at modern publication practices. So I think you'll have to go by WP:GNG instead, but the difficulty there is that if sources from that time exist, they may be more likely to be in print than online. I have no particular expertise in finding print sources from early to mid 20th-century India. —David Eppstein (talk) 23:33, 24 February 2021 (UTC)

Scholia relevance?
Dear Prof. David Eppstein, I'd like to ask about your revert of the Scholia link on Vojt%C4%9Bch Jarn%C3%ADk, "Scholia is merely a Wikidata-scraper...". Does this mean that Scholia is generally considered irrelevant? (for the sake of adding that link to Wikipedia I added the most important scholarly articles of Vojtěch Jarník to Wikidata). Paul Erdős has the Scholia link too, that could serve as an example, at least for me. Thank you for any comment on the topic. --PKalnai (talk) 08:41, 28 February 2021 (UTC)

GA nomination of List of University of Illinois Presidents
Hi David Eppstein,

I am removing the nomination of this article, as I have learned that lists cannot be promoted to WP:GA. I will be nominating it for WP:FL instead. Thanks. Thrakkx (talk) 15:23, 28 February 2021 (UTC)

$$...$$ markup and images
Can you copy Wikipedia content created by $$...$$ markup and paste it as digits of text? My preferences are already set as you indicate they should be and, regardless of browser, that markup is rendered as an image and pastes as an image. Thanks — Q uantling (talk &#124; contribs) 21:50, 2 March 2021 (UTC)
 * Yes. I can. As I have repeatedly indicated in my edit summaries. Using Chrome. —David Eppstein (talk) 22:14, 2 March 2021 (UTC)
 * Thank you for taking the trouble to respond to my questions. The formula
 * $$\zeta^{\prime}(0) = -\frac{1}{2}\ln(2\pi)\approx -0.918938533\ldots$$
 * is rendered as an image in Chrome for me despite that my settings are as you suggested. Do you see this formula as an image?  If so, how do you copy just the digits -0.918... into your clipboard?  — Q uantling (talk &#124; contribs) 01:39, 4 March 2021 (UTC)
 * It looks like a formula. I don't know what you mean by "an image": everything on my screen is an image. That's what screens do. My browser doesn't tell me how the things it displays are coded, it just displays them for me. I can select the formula by clicking and dragging my mouse across it; only the whole formula can be selected, not pieces of that. Using the keyboard commands to copy the selection and paste into another window or program produces the text . —David Eppstein (talk) 01:55, 4 March 2021 (UTC)
 * An image is a JPEG, PNG, TIFF, GIF, SVG, etc. My article edits were specifically motivated to allow the user to easily copy the digits; copying the image itself or its ALT text is not the same experience.  Now that I have confirmed that there is nothing wrong with my browser nor my ability to use it, I will move this discussion to the article's talk page.  There we can discuss the relative priority of prettiness vs. copy-ability. — Q uantling (talk &#124; contribs) 02:24, 4 March 2021 (UTC)
 * Ok, so you admit that all along you have actually been able to copy the text of the mathematical equations, you merely have a psychological block against doing so because you think it's an image and you are too lazy to pick out the parts you want. That is a bad reason for degrading the reading experience of all other editors. If you repeat your edits making mathematical articles worse I will revert them and if necessary escalate the issue. —David Eppstein (talk) 02:42, 4 March 2021 (UTC)
 * Thank you for your reply. I have posted to the article's talk page.  — Q uantling (talk &#124; contribs) 17:42, 6 March 2021 (UTC)

Roald Dahl
Hi,

I just wanted to let you know I was not trying to make any sort of statement about Dahl's antisemitism by undo-ing your edit. I'm actually the one who added the entire antisemitism/racism/criticisms section to his page in October? So I apologize if it came off incorrectly and a cleaned up version has been added to the lede so it should all be okay now.

Apathyash (talk) 16:42, 4 March 2021 (UTC)

Your GA nomination of De quinque corporibus regularibus
The article De quinque corporibus regularibus you nominated as a good article has passed ; see Talk:De quinque corporibus regularibus for comments about the article. Well done! If the article has not already appeared on the main page as a "Did you know" item, or as a bold link under "In the News" or in the "On This Day" prose section, you can nominate it within the next seven days to appear in DYK. Bolded names with dates listed at the bottom of the "On This Day" column do not affect DYK eligibility. Message delivered by Legobot, on behalf of Ján Kepler -- Ján Kepler (talk) 07:02, 5 March 2021 (UTC)

reference to secondary rather than primary source on Sarah B. Hart page
Hi David - thanks for your work on Hart's page. When I saw it, I noticed that her J Hum Maths paper on "Ahab's arithmetic" was referenced not directly, but through a Spanish blog posting. As this is a secondary source, in a language other than English, I replaced it with a reference to her paper - which, I thought, would directly establish that she'd written in this area. J Hum Maths is peer-reviewed and has a respectable editorial board; thus, if Hart has persuaded her referees and editors that she is writing about Moby Dick, this convinces me more than an un-refereed blog post does.

I've just seen that you've reverted that, noting "We cannot use a publication BY Hart as a reference ABOUT what Hart has done. That's why this reference goes to a review of her work, not to the work itself". Can you point me to the wiki policy you have in mind here? (I've been guided by that at https://en.wikipedia.org/wiki/Wikipedia:Identifying_and_using_primary_sources, e.g. "'Secondary' is not, and should not be, a bit of jargon used by Wikipedians to mean 'good' or 'reliable' or 'usable'" and "'Primary' is not, and should not be, a bit of jargon used by Wikipedians to mean 'bad' or 'unreliable' or 'unusable'".

I've looked at some other mathematicians' pages - admittedly not a random or extensive sample - and note that, for example, Wiles' page both cites his 1995 Annals special issue, and then a Scientific American article four years later noting that the Annals proof had survived scrutiny. In Hart's case, there's no claim that she's solved a particular problem, so no need for a secondary source that could comment on the community's consensus in a way that she couldn't.

What would you think of a 'Wiley' solution, citing both her actual paper as well as noting that it had been picked up by the popularisation literature?

Thanks!

Colin Rowat (talk) 21:06, 5 March 2021 (UTC)
 * The key guideline here is WP:BLPPRIMARY: A publication by the subject is clearly a primary source for the fact that the subject has published about that topic, regardless of the editorial approval it may have gone through. As the link says, "it may be acceptable to rely on it to augment the secondary source", but it should not replace the source. —David Eppstein (talk) 21:16, 5 March 2021 (UTC)


 * Thanks David. WP:BLPPRIMARY seemed to value secondary sources due to the concerns about the reliability of narrators' accounts of their own activities.  In this case, we're just claiming that Hart's written about Moby Dick, not that her work is deep, significant, or even correct.  Thus, the work itself suffices to establish this.  In any case, the New York Times has just published a profile on Hart, which is a better secondary source than the Spanish blog, if only because it's in English.  Thus, I've added the NYT reference, as well as the direct link, to save readers who might be intersted in seeing the original article a click or two. Colin Rowat (talk) 12:24, 7 March 2021 (UTC)

Tic-tac-toe
Dear David Eppstein,

Thank you for your review of my added figure to the tic-tac-toe page (https://en.wikipedia.org/wiki/Tic-tac-toe). You are totally right that the caption was not sufficiently self-explaining. I also agree that the figure is only readable in full-screen mode. However, I believe that the figure is very valuable in full screen mode, once its content and purpose is properly explained. Therefore, I made improvements to my proposal for the tic-tac-toe page, hopefully addressing your objections, and I would like to hear your opinion once more.

My intention for improvement of the tic-tac-toe page

The description of the optimal strategy is quite verbose for such a simple game. But it can not be made simpler. So, I was looking for a graphical representation of the strategy of the game. The existing fractal images are non-readable (even in full-screen mode, I need a magnifying glass to see the details), contain trivial moves (for which I do not need a cheat sheet), and are incomplete (optimal strategy for player x only starts in the corner, while there actually exist plenty of fun games when not starting in the corner). In my opinion, these figures should be removed. I found, I think, a nice alternative way of expressing the essential knowledge of the game graphically in a one-pager cheat sheet. I believe that this one-pager could help people explore the fun of the game much more, and is therefore a valuable add to the Wikipedia page.

My proposal for improvement of the page

- I added figure "Tic-tac-toe cheat sheet.svg" (improved title as compared to the previous version)

- Improved caption: "Cheat sheet listing all possible non-trivial mistakes. Here, a non-trivial mistake is a move that allows the opponent to force a win in more than one move. This cheat sheet can help you to prevent making a mistake, to choose those moves that allow your opponent to make mistakes, and to recognize mistakes by your opponent. Trivial mistakes (moves that allow your opponent to win in the next move) are not listed, because these are easily recognized without a cheat sheet."

- I removed the previous figures on the perfect strategy. They are not readable, even in full screen mode.

Thanks in advance for your review!

Wouter Peeters

I noticed that my contribution to the tic-tac-toe page was qualified as potential vandalism by a computer program. Looking back, I think I was a bit too enthusiastic about my own contribution, and I should have respected your initial judgement to remove my contribution. I have no intention of doing harm to Wikipedia, or to keep you occupied with vandalism. My apologies for my naïve behavior. Please feel free to remove this section from your talk page without a reply. Kind regards Wouter Peeters 19:29, 7 March 2021 (UTC) — Preceding unsigned comment added by Whpeeters (talk • contribs)

Your GA nomination of Borromean rings
Hi there, I'm pleased to inform you that I've begun reviewing the article Borromean rings you nominated for GA-status according to the criteria. This process may take up to 7 days. Feel free to contact me with any questions or comments you might have during this period. Message delivered by Legobot, on behalf of Tayi Arajakate -- Tayi Arajakate (talk) 16:40, 6 March 2021 (UTC)

MDPI
Hey David, I've noticed you are extremely passionate about the MDPI page we are editing together. You have stated that you have dealt with the company in the past and I am wondering if this might represent a WP:CONFLICT that makes it hard for you to edit the page objectively. I understand your concerns with this "borderline" publisher and I share your desire to accurately report their problematic practices but your contributions to the discussion and to the article appear quite one-sided and excessively heated. What do you think? This is not meant to be an accusation of any kind. I'm just trying to figure out a way to work more productively together. We are all volunteers here and nobody wants to waste their time in pointless discussions. Thanks for the work you do on Wikipedia. -- &#123;{u&#124; Gtoffoletto  &#125;}  talk 18:07, 8 March 2021 (UTC)
 * This is some high-quality trolling, this is. --JBL (talk) 18:20, 8 March 2021 (UTC)
 * [edit conflict] I have not dealt with them at all professionally (except for once turning down an invitation to be an editor for one of their journals) so I consider myself not to have any COI on them, any more than any other topic on Wikipedia where there is not complete agreement among all editors about what should be included. Most of my frustration with this article is with long-term abuse by editors who appear to be editing with a pro-MDPI conflict of interest, being very persistent at attempting to remove any criticism of the company from the article. You do not appear to have been editing the article long, but you appear to be falling into the same pattern of trying to whitewash their shortcomings. —David Eppstein (talk) 18:22, 8 March 2021 (UTC)
 * And I can assure you I have no intention of whitewashing anything at all so you can stop worrying about that :-) I'm glad we cleared this out and look forward to editing with you. -- &#123;{u&#124; Gtoffoletto  &#125;}  talk 18:56, 8 March 2021 (UTC)

Kuratowski's and Wagner's theorems
Hi Dr Eppstein,

As you seem to know much about mathematics, I wonder if I could check if I got the terminology and application of Kuratowski's and Wagner's theorems in this diagram.

Thanks, cm&#610;&#671;ee&#9094;&#964;a&#671;&#954; 01:09, 9 March 2021 (UTC)
 * The top one looks like a minor rather than a subdivision (because you have to contract the edge between the two black vertices to make it look like K5). So it is an example of Wagner's theorem but not of Kuratowski's theorem. Instead it is possible to get a K5 subdivision consisting of one vertex, its four neighbors, and paths of length two between each pair of neighbors. It's easy to prove that this graph is nonplanar without going through Kuratowski or Wagner, though: it has too many edges to be a bipartite planar graph (greater than $$2n-4$$). —David Eppstein (talk) 01:21, 9 March 2021 (UTC)
 * Thank you very much for the speedy replies. I've updated the captions accordingly. Though I understand there are other ways to prove it, I wanted an image to illustrate these two theorems.
 * Re "it is possible to get a K5 subdivision consisting of one vertex, its four neighbors, and paths of length two between each pair of neighbors", though, I tried to understand what you meant but couldn't find the subgraph. I read "paths of length two between each pair of neighbors" to mean that each pair of vertices is separated by paths four edges long. As there are ten pairings, that makes 40 edges, but a tesseract has only 32 edges. Could I trouble you to explain, please? Thanks, cm&#610;&#671;ee&#9094;&#964;a&#671;&#954; 01:13, 11 March 2021 (UTC)
 * The hypercube can be described as having all subsets of a four-element universe {a,b,c,d} as its vertices, and edges between subsets that differ in a single element. The empty set has four neighbors {a}, {b}, {c}, and {d}. Each two of these neighbors can be connected by a path of length TWO (not four), that avoids the empty set: {x} – {x,y} – {y}. These five vertices, the four edges from the empty set to the singletons, and the six paths between pairs of singletons, form a subdivision of K5. In contrast, what your picture shows is not a subdivision of K5. Therefore it is incorrect and false to assert (as some of your uses of this image assert) that it provides an example of Kuratowski's theorem. —David Eppstein (talk) 01:18, 11 March 2021 (UTC)
 * Thanks again, Dr Eppstein. I think I understand now: each vertex can be specified using four coordinates (0,0,0,0)...(1,1,1,1), the empty set (say, vertex O) being (0,0,0,0) and the neighbours having a single 1 – say, A (1,0,0,0), B (0,1,0,0), C (0,0,1,0) and D (0,0,0,1); OA, OB, OC and OD are edges of one segment as only one coordinate must change, whereas each pair of neighbours is an edge with two segments as two coordinates must change. I've updated the diagram accordingly and hope it's now true that Kuratowski's theorem can be applied to the coloured subgraph. Cheers, cm&#610;&#671;ee&#9094;&#964;a&#671;&#954; 00:51, 12 March 2021 (UTC)
 * Yes, it looks better now. —David Eppstein (talk) 00:53, 12 March 2021 (UTC)

Research Gate
David, I can't find this particular Andrew Hill (pharmacologist) in Google Scholar. But there's a link to ResearchGate where it says he has 324 publications and 9,800 citations. About 20-30% listed have a preprint tag, the rest say article. Although being ResearchGate, I'm not sure how their system works. Are these published in peer-reviewed journals? I don't think I've seen ResearchGate stuff much before. How would you evaluate this one? Thanks for your help.  Onel 5969  TT me 22:13, 10 March 2021 (UTC)
 * ResearchGate hosts preprints. By themselves, they count as self-published sources from the point of view of reliability. But as you say many of them are preprints of things published elsewhere, for instance in journals. I don't know much about how believable their citation numbers might be. Also I'm more interested in peak citations per paper than in total citations. I think he may have some triple-digit publications in Google Scholar but it's really hard to tease out, and this is a high-citation field so that may not be enough. In this case I'd like to see a little more than bare citation counts before we declare him notable. —David Eppstein (talk) 23:42, 10 March 2021 (UTC)
 * Thanks. There are two other Andrew Hills that I came up with in Google Scholar, both with nice citation counts, but I couldn't find him.  Onel 5969  TT me 00:10, 11 March 2021 (UTC)
 * His papers are there but they're mixed up among all the other Andrew Hills. He doesn't seem to have a separate profile, or if he does I didn't find it. —David Eppstein (talk) 00:13, 11 March 2021 (UTC)

A barnstar for you!

 * You're welcome! —David Eppstein (talk) 19:02, 12 March 2021 (UTC)

March 2021
Hello, I'm Kammill. I wanted to let you know that one or more of your recent contributions&#32;to Dragon curve have been undone because they did not appear constructive. If you would like to experiment, please use your sandbox. If you have any questions, you can ask for assistance at the Teahouse. Thanks. – Kammill  ⟨talk⟩ 06:06, 14 March 2021 (UTC)
 * Really? I remove unsourced cruft and improve the sourcing (including removing spam from sockpuppet Sockpuppet investigations/Badtoothfairy/Archive) and this is the thanks I get? From a three-month-old account, no less? Next time, try looking a little less superficially at your watchlist. Oh, and since you're three months old and might not have seen it before, WP:DTR. —David Eppstein (talk) 06:14, 14 March 2021 (UTC)
 * , my mistake. I apologize. – Kammill  ⟨talk⟩ 06:17, 14 March 2021 (UTC)
 * , I'm really sorry! Actually I made a few problematic edits, you see. I thought I was filtering the changes at RCP, but I didn't knew that I was actually reverting those edits. Once again, I'm sorry! – Kammill  ⟨talk⟩ 06:24, 14 March 2021 (UTC)

Your GA nomination of Borromean rings
The article Borromean rings you nominated as a good article has been placed on hold. The article is close to meeting the good article criteria, but there are some minor changes or clarifications needing to be addressed. If these are fixed within 7 days, the article will pass; otherwise it may fail. See Talk:Borromean rings for issues which need to be addressed. Message delivered by Legobot, on behalf of Tayi Arajakate -- Tayi Arajakate (talk) 02:20, 16 March 2021 (UTC)

Your GA nomination of Borromean rings
The article Borromean rings you nominated as a good article has passed ; see Talk:Borromean rings for comments about the article. Well done! If the article has not already appeared on the main page as a "Did you know" item, or as a bold link under "In the News" or in the "On This Day" prose section, you can nominate it within the next seven days to appear in DYK. Bolded names with dates listed at the bottom of the "On This Day" column do not affect DYK eligibility. Message delivered by Legobot, on behalf of Tayi Arajakate -- Tayi Arajakate (talk) 07:22, 16 March 2021 (UTC)

Carol Cleland Page
Hello,

I apologize if this isn't the right forum for discussion, I am a neophyte what lies behind the curtain on Wikipedia.

I am working with Carol Cleland to add some content to her wikipedia page, and she provided me with the list of publications and book chapters to add. If that isn't a sufficient reason for having this particular list then I can work with her to trim down the list to something like 6 works that are representative of her work. I appreciate any general guidance for what principles should guide this kind of thing.

Thanks for your help!

- Joe
 * So you have a conflict of interest? All conflicts of interest must be explicitly declared by Wikipedia and Wikimedia policy, and conflict-of-interest changes to biographies should be proposed on the article talk page rather than performed directly. Also see our prohibition on autobiographies. To be blunt, Cleland, and you as her agent, should not be editing our article about her. This is an encyclopedia, not a social media site and not a site for promotional self-description. If there are important inaccuracies in the article, bring it up on the article talk page and they will get fixed. If you just want to polish it up and make it more to her taste, don't. —David Eppstein (talk) 19:55, 18 March 2021 (UTC)

Thanks David, that all makes sense. I will refrain from making or proposing any changes to the page. Thanks for your help in understanding how all of this works. — Preceding unsigned comment added by Joeman714 (talk • contribs) 20:36, 18 March 2021 (UTC)

Arnold's limit
Hi, Professor David Eppstein. I would like to create an article about Arnold's limit, but the best sources I could find for it were a YouTube video and a MathOverflow thread. What do you think? Thank you! MathKeduor7 (talk) 12:49, 21 March 2021 (UTC)
 * Neither of those is usable as a source, so you should keep looking for better ones. —David Eppstein (talk) 16:22, 21 March 2021 (UTC)
 * Thank you. I'll wait for someone to write about it in a journal or magazine before I create an article about it here! MathKeduor7 (talk) 16:45, 21 March 2021 (UTC)

April editathons from Women in Red
--Megalibrarygirl (talk) 20:15, 22 March 2021 (UTC) via MassMessaging

Recamán's sequence
I saw your edit to Recamán's sequence. It didn't sound plausible to me either, but a year or so ago I contacted him and asked how he did it. I didn't follow all that he said, but he said that it got sub-linear because of the self-similarity in the structure, I think. Bubba73 You talkin' to me? 21:18, 22 March 2021 (UTC)
 * You can certainly get sublinear. I would believe 10^23 based on that. But 10^230? I would need to see a lot more specific an explanation and a lot more reliable a publication of it. —David Eppstein (talk) 21:19, 22 March 2021 (UTC)


 * I couldn't find my messages about it. I contacted him through https://oeis.org/wiki/User:Benjamin_Chaffin - email user in the lower left.  He said that he was thinking about writing it up.  Note that what he said he did is a visualization of 10^230 terms - it doesn't mean that he calculated 10^230 terms (which doesn't sound possible). Bubba73 You talkin' to me? 23:37, 22 March 2021 (UTC)
 * Sure, but you need to calculate enough information about the terms to visualize them accurately, for instance meaning that we should be able to determine from this calculation whether 852,655 is among them. Anyway, if he does publish we can definitely revisit the issue. —David Eppstein (talk) 00:16, 23 March 2021 (UTC)


 * You might ask him for an update and to explain what he did. Bubba73 You talkin' to me? 01:50, 23 March 2021 (UTC)

Fermat's right triangle theorem
Regarding "According to Fermat's right triangle theorem, it is not possible for all four lengths a, b, c, and d to be integers.", my inner snot-nosed kid wants to say, "Zero is an integer" :-) — Preceding unsigned comment added by RoySmith (talk • contribs) 20:47, 24 March 2021 (UTC)
 * This is in the caption to the figure of Fermat's right triangle theorem, right? I think the caption is ok without needing more qualifiers, because it is also not possible for the length of a side of a triangle to be zero or negative. —David Eppstein (talk) 20:51, 24 March 2021 (UTC)

Your GA nomination of Fermat's right triangle theorem
Hi there, I'm pleased to inform you that I've begun reviewing the article Fermat's right triangle theorem you nominated for GA-status according to the criteria. This process may take up to 7 days. Feel free to contact me with any questions or comments you might have during this period. Message delivered by Legobot, on behalf of RoySmith -- RoySmith (talk) 21:21, 24 March 2021 (UTC)

American women mathematicians
Hi, what is the reasoning for reverting many of my changes? Per wikipedia policy, the most specific categories should be used. "American women mathematicians" is a subcategory of "American mathematicians". Would you please revert your rollbacks? Thank you. TiMike (talk) 02:47, 25 March 2021 (UTC)
 * Did you not read your talk? Do you not understand what a NON-DIFFUSING SUBCATEGORY means? It means you should not remove them from the parent category when they are in this subcategory (unless they are also in other subcategories of the same parent). No, I will not reinstate your bad edits, and if you start making them again after being told not to you are likely to be blocked. —David Eppstein (talk) 05:19, 25 March 2021 (UTC)
 * I read this before my talk page. I was not aware of non-diffusing subcategory. Thanks for informing me. I'd recommend following Wikipedia's fundamental principle of assuming good faith going forward. All I did was ask for a reason. Good day.

Your GA nomination of Fermat's right triangle theorem
The article Fermat's right triangle theorem you nominated as a good article has passed ; see Talk:Fermat's right triangle theorem for comments about the article. Well done! If the article has not already appeared on the main page as a "Did you know" item, or as a bold link under "In the News" or in the "On This Day" prose section, you can nominate it within the next seven days to appear in DYK. Bolded names with dates listed at the bottom of the "On This Day" column do not affect DYK eligibility. Message delivered by Legobot, on behalf of RoySmith -- RoySmith (talk) 14:21, 25 March 2021 (UTC)
 * Thanks! —David Eppstein (talk) 16:27, 25 March 2021 (UTC)

DYK for Theodore Cohen (chemist)
— Maile (talk) 13:10, 26 March 2021 (UTC)

Knuth Prize question
David, I notice that you have many contributions to Wikipedia, and you also work across disciplinary boundaries in CS and Math (I do too). I'm noticing that the page for the Knuth Prize has been classified as: "relies too much on references to primary sources" and also: "may not meet Wikipedia's general notability guideline". https://en.wikipedia.org/wiki/Knuth_Prize At the time I saw this, there were only 3 references, and they were all ACM references! So I added 11 new references, from several other sources, and I might add even a few more references. You have a lot more Wikipedia experience than I do. Do you have suggestions about this page? If so, I would be happy to work hard to improve it. I did not create this page, but it seems worthwhile to me, and I want to help. I'm spending more time recently doing some Wikipedia editing, and I thought this was a nice place to pitch in. Thanks for any suggestions that you might have! Best wishes! MDW333 (talk) 03:09, 27 March 2021 (UTC)

WOW! By pure coincidence, while I was still working on some references for the Knuth Prize page tonight, I saw that you are actually the person who took the picture in the caption on this page! Very cool! https://en.wikipedia.org/wiki/Knuth_Prize#/media/File:Strassen_Knuth_Prize_presentation.jpg https://www.ics.uci.edu/~eppstein/pix/strassen/KnuthPrizePresentation2.html I promise that this is pure coincidence. OK, I was delighted when I saw this, and it made me feel happy that I wrote to you earlier tonight about whether there are ways to improve the page. Sometimes there really is such a good thing as luck. MDW333 (talk) 03:36, 27 March 2021 (UTC)
 * Yes, some of these academic societies and their awards and publications can be very difficult to find non-primary sourcing for, especially because the same societies publish the publications that you're likely to find sources in. A note re the sources you added: Sources from the employer of an academic saying "our professor won a prize" generally don't count as independent. And the source you've listed as coming from the AAAS is really just a repackaged ACM press release so it's also not independent. Other societies that might say something relevant include cra.org and eatcs.org but I didn't see a lot there on the prize in general (although there were some sources on individual winners). What you won't see on the Strassen photo description was that I shot it with Erik Demaine's camera — I was taking the photos semi-officially, meaning that someone associated with the event (I forget who, probably Gary Miller) asked me to, and I was using two cameras for greater reliability and so that I wouldn't have to waste time swapping lenses. —David Eppstein (talk) 04:17, 27 March 2021 (UTC)

David Eppstein thank you very much for the lightning fast reply, and for these insights! I truly appreciate your feedback. I'm relatively new at editing pages on Wikipedia. OK, I see two issues here: 1. "relies too much on references to primary sources"  I agree with you that it is really hard to find non-primary sourcing for this Knuth Prize. 2. "may not meet Wikipedia's general notability guideline"  This one should be easier to address, because the prize has existed for many years, and the prize winners are some of the most well known in their field, and Knuth himself is a giant in science. I think that this article definitely meets Wikipedia's general notability guideline. Do you have some advice about how I explain this to the editor who posted that concern? I truly appreciate your advice, especially about this second point. Thank you very much for considering! MDW333 (talk) 04:24, 27 March 2021 (UTC)
 * The problem is that, to meet Wikipedia's general notability guideline, by definition, it must be covered in-depth in multiple independent reliable sources. Sources that merely say that one of the recipients received that award among others are not in-depth, and the sources we have on the award itself are for the most part independent. "Notability" does not mean "importance". I agree that this award is important enough that we should have an article on it, but convincing Wikipedia editors who are less closely connected to the topic looks tricky at this point. —David Eppstein (talk) 04:28, 27 March 2021 (UTC)

Thank you very much for the guidance David Eppstein I appreciate this insight! MDW333 (talk) 04:33, 27 March 2021 (UTC)
 * Incidentally, if you do need additional independent sources for the individual winners, many of them are in the Notices of the American Mathematical Society. Search ams.org for the winner name and knuth prize. I didn't see any coverage of the overall prize there though. —David Eppstein (talk) 05:47, 27 March 2021 (UTC)

OK, thank you again, David Eppstein for your help! Much appreciated! Incidentally, I started to go down this path because I have been assembling a new page for the Flajolet Lecture Prize https://en.wikipedia.org/wiki/Draft:Flajolet_Lecture_Prize and I am learning a lot, as I go. Thanks again! MDW333 (talk) 11:27, 27 March 2021 (UTC)

'Actor'
Most pleasant indeed to read from someone in the OC. I grew up there, and we had a long-running joke about how the birds all fly in circles because they have only a right wing. 🤪🖖🏻—ATS (talk) 21:56, 27 March 2021 (UTC)
 * I live in one of the most blue neighborhoods of OC (the one Katie Porter comes from), but although there are still also pockets of deep red, the county as a whole has moved more to the middle of the road (for US politics) in recent years. —David Eppstein (talk) 18:48, 28 March 2021 (UTC)
 * And may she be in her seat for decades to come.
 * Meanwhile, I wrote the Featured Article Kona Lanes, so you can guess where much of my childhood was spent. 😁 —ATS (talk) 22:37, 28 March 2021 (UTC)

DYK for Borromean rings
Cwmhiraeth (talk) 12:01, 28 March 2021 (UTC)

Your GA nomination of Keller's conjecture
The article Keller's conjecture you nominated as a good article has been placed on hold. The article is close to meeting the good article criteria, but there are some minor changes or clarifications needing to be addressed. If these are fixed within 7 days, the article will pass; otherwise it may fail. See Talk:Keller's conjecture for issues which need to be addressed. Message delivered by Legobot, on behalf of Pbrks -- Pbrks (talk) 06:41, 30 March 2021 (UTC)

Draft:Renée Mauborgne
Hi What are your views on this girl. I was planning to promote it, copyedit it and find some additional sources; if she is notable. It has been sitting in the queue for some weeks now, moved out to draft due to tone issues. I must have looked at it about 6 times now.  scope_creep Talk  19:20, 31 March 2021 (UTC)
 * Her citations look like a pass of WP:PROF. The draft looks kind of spammy, though, especially in the listing of awards, which look dubiously significant. I would instead look for published reviews of her books to bolster the draft. —David Eppstein (talk) 19:50, 31 March 2021 (UTC)
 * Right. Thanks David.  scope_creep Talk  19:58, 31 March 2021 (UTC)