Talk:Infinity symbol/GA1

GA Review
The edit link for this section can be used to add comments to the review.''

Reviewer: Theleekycauldron (talk · contribs) 22:06, 9 February 2022 (UTC)

Hi there! I think I can take this review, although this is my first time coming close to the mathematics section. Comments should be finished in at most a week. Thanks! theleekycauldron (talk • contribs) (she/they) 22:06, 9 February 2022 (UTC)
 * Thanks! As mathematics goes, I think this should be mostly pretty non-technical. —David Eppstein (talk) 23:09, 9 February 2022 (UTC)
 * Yep, that's why I took this :) and I had such a promising future in math, too... theleekycauldron (talk • contribs) (she/they) 09:00, 11 February 2022 (UTC)

General comments

 * I'm seeing that four years ago, you removed a reference to the infinity symbol as it addressed neurodiversity; are you able to find better sourcing for that idea now?
 * Yes, I think so. There are lots of sources but in my cleanups in prep for GA I replaced a dubious primary source (some specific neurodiversity group's web site) with a published academic journal article, which gives a rainbow infinity sign as Figure 1, with the caption "Many people with autism and their families like the rainbow infinity sign as a symbol of neurodiversity, representing not just the tremendous diversity of people on the spectrum but also of human cognitive and intellectual capabilities." —David Eppstein (talk) 17:26, 11 February 2022 (UTC)
 * is contains special characters useful?
 * I don't know, is it? Are you asking whether it should be added to this article? It doesn't appear to be there already, and its documentation says not to use it on mathematics articles. —David Eppstein (talk) 01:25, 18 February 2022 (UTC)
 * ah, my mistake. theleekycauldron (talk • contribs) (she/they) 02:04, 22 February 2022 (UTC)

Lead

 * Is it necessary to include three different variations of the infinity sign in the lead sentence (plus one in the infobox)? Surely once in the lead is enough
 * Cut down to one. I guess the lead of this article is not the place to provide a tutorial to Wikipedia editors on the different choices they have for formatting this symbol in wikitext and the different appearances those choices result in. —David Eppstein (talk) 17:28, 11 February 2022 (UTC)
 * Any reason lemniscate is linked on second mention, or why a distinction appears to be made between lemniscates and lemniscate curves?
 * First mention = what this symbol is called. Second mention = similar thing also called a lemniscate, linked to the article on that similar thing. "Lemniscate curve" = an attempt at differentiating the lemniscate-as-symbol meaning here from the lemniscate-as-curve meaning in the linked article. —David Eppstein (talk) 01:28, 18 February 2022 (UTC)
 * or "lazy eight", in the terminology of livestock branding: pretty sure that comma's not supposed to be there
 * I'm pretty sure that when we say "it is also called X explanation-for-X or Y explanation-for-Y" we should punctuate both of the parallel clauses in the same way as each other. Do you think the comma should be removed both from "lemniscate, after the lemniscate curves of a similar shape studied in algebraic geometry" and from "lazy eight, in the terminology of livestock branding", or do you think it should be kept in both? —David Eppstein (talk) 01:28, 18 February 2022 (UTC)

History

 * Wallis did not explain his choice of this symbol, but it has been conjectured to be a variant form of a Roman numeral for 100 million, which resembled the same symbol enclosed within a rectangular frame, of other Roman numerals such as the notation CIↃ used to represent 1,000, or else as a variant of ω, the lower-case form of omega, the last letter in the Greek alphabet.: can this sentence be broken up? Also, there's two spaces in "of ω".
 * Ok, split. The second space was intended to prevent bad line breaks but because of the first space was ineffective at doing so; I change to a nowrap template. —David Eppstein (talk) 01:34, 18 February 2022 (UTC)

Usage

 * It's probably worth it to make it clear, in the Pale Fire bit, that "Pale Fire" (the poem) is still a work of Nabokov.
 * Reworded. —David Eppstein (talk) 01:36, 18 February 2022 (UTC)
 * Side view of a camera lens showing infinity symbol on the focal length indicator: should be lens, showing
 * I'm not convinced it's obligatory, but it doesn't hurt, so ok, done. —David Eppstein (talk) 01:36, 18 February 2022 (UTC)

Sourcing
Quality and verifiability: version review :D
 * General comments:
 * Make sure your date formats are consistent, I'm seeing MM yyyy; yyyy-mm-dd; MM dd, yyyy; and dd MM yyyy.
 * I added a Use mdy dates template specifying my intended format, "cs1-dates=ly", one of the standard and allowed date formats, which uses MM dd, yyyy for publication dates and yyyy-mm-dd dates for access- and archive-dates. This is not an inconsistent format, and adding the template has the benefit that if some of the date formats are in a different format in the source citation templates, they will be automatically standardized. As for what level of detail to specify for dates: I think it is reasonable to use only years for book publications (more precise dating is not really meaningful in the world of book publication and even the year is often fudged by one for marketing reasons), and months for academic journal publications (the level of dating that journal issues are usually specified as having). I don't generally care to look up web page publication dates myself but for those a day of the month does make sense since they can change arbitrarily; same for access-dates.
 * Many linked sources are missing URL access dates, although you may not see that as necessary.
 * I didn't add those sources so I didn't add access-dates for them. —David Eppstein (talk) 01:44, 18 February 2022 (UTC)
 * Make sure you're consistent about whether you're using ISBN10 or ISBN13.
 * Really? Really?? You do realize that the GA criteria do not actually require citations to be consistently formatted at all. Read them. What they say (in 2a) is that the LAYOUT of the citation section meets the layout style guidelines, and (in 2b) that the sources be reliable. I can understand preferring properly hyphenated ISBNs over unhyphenated ones, but if a book was published before ISBN13 was a thing, and shows only an ISBN10, I don't see anything wrong with citing it as it is rather than as you would like to pretend it might have been. —David Eppstein (talk) 01:44, 18 February 2022 (UTC)
 * Statements that appear to be unsourced:
 * Tarot deck image doesn't provide a source that it's an infinity symbol, and not merely a lemniscate; may want to mention its usage in body
 * Added a source. —David Eppstein (talk) 02:16, 19 February 2022 (UTC)
 * Rows 2–5 of the first table, and rows 3–5 of the next one (starting from encodings)
 * Those rows are provided automatically by the charmap template. Many of them could be sourced from the compart unicode page but the charmap template does not provide any way to specify a source for them, nor to disable these rows. —David Eppstein (talk) 16:55, 19 February 2022 (UTC)
 * 1) Rucker 1982: Reliable; could include a google books link, it seems freely accessible to verify (you'd want a retrieval date); verifies information
 * 2) *I could not find a Google Books link for the edition I cited. Unless there is a good reason I prefer to cite the original edition of a work. A good reason might be that a better treatment of the subject can be found in a later edition, or that I only have access to a later edition and can't check that the text I want to cite is really in the original edition. Availability of a Google Books link that might or might not continue to work is not a good reason. —David Eppstein (talk) 02:18, 19 February 2022 (UTC)
 * 3) Erickson 2011: Reliable, but the formatting doesn't seem to check out—as far as I can tell from Cite book, the  param is for afterwords, prefaces, dust jackets—parts of the book not necessarily written by the author (you'll also need a URL access date). Also, this might be a little finicky/WP:SKYBLUE, but I'm not seeing the source say that the lemniscate is similar to the infinity symbol, nor that it's studied in algebraic geometry. The first bit of that could be cited to Rivkin et al.
 * 4) *The only distinction between the chapter and contribution parameters is in the mind of whoever wrote that documentation. They are synonyms from the point of view of the citations they produce. I prefer to use contribution because it is inaccurate to call sections chapters, to call appendices chapters, etc., and also because that's the parameter that is understood by the software I use to format references. —David Eppstein (talk) 02:23, 19 February 2022 (UTC)
 * 5) Humez and Humez et al. 1993: Reliable; missing URL access date; verifies information
 * 6) *I do not believe in providing access dates for courtesy links on book or periodical publications. The actual citation is the book or periodical publication, and would be a valid citation regardless of whether the link were present or not. The link is there to make things easier for readers, not because it is a necessary part of the citation. As such, the date the link was accessed is irrelevant for the purpose of access-dates, verification of content and recovery of the correct version in the case of bad links. Because it is irrelevant information, it should not be presented to readers as if it were relevant. Access-dates should be used only for references for which the link itself is the primary access to the reference (generally, only cite web). —David Eppstein (talk) 02:23, 19 February 2022 (UTC)
 * 7) van Riel 2017: Reliable; I would suggest making the page numbering more specific, and missing URL access date; verifies information
 * 8) *It is mandatory to provide the full page range of a journal publication. Our citation templates do not provide a way to provide both the full page range and a pointer to a more specific selection of numbers. Instead, I added a quote after the citation template, with a pointer to its page number. Re access date, see above. —David Eppstein (talk) 02:38, 19 February 2022 (UTC)
 * 9) Wallis 1655: Primary; missing URL access date; AGF on foreign language content, but I don't think this one's necessary. Between it being written in latin(?) and it being used to source a statement crediting him for a discovering, it's not overly helpful.
 * 10) *I think that providing links to primary but historical sources is an important part of verifiability. It is not being used to source the claim that Wallis was the one to invent the notation (a secondary source is needed for that, and two are provided) but instead with the intent of providing a valuable resource for readers who want to find out exactly what he said when he invented it. Why the question mark on "in latin"? The citation is properly marked as being in Latin. And being in a non-English language is only a problem when a suitable replacement source is available in English; for historical documents, that cannot be the case. —David Eppstein (talk) 02:38, 19 February 2022 (UTC)
 * 11) Scott 1981: Looks reliable; missing URL access date; AGF on inaccessible content
 * 12) *Pretty sure it was accessible when I accessed it. This is why I don't trust Google Books links. They come and go too randomly. Access date would not help; Google isn't going to change the book at that link, only whether or not you could see it. —David Eppstein (talk) 02:48, 19 February 2022 (UTC)
 * 13) Cajori 1929: Reliable, missing URL access date and you might want to consider making the page numbers a bit more specific with sfn
 * 14) Looks like this one's all you need for this sentence; verifies content in whole.
 * 15) Hmm. Verifies content as explicitly written, but if you don't dig into the source, it seems as though you claim that Euler's variation is also due to space concerns, which isn't verifiable from the source.
 * 16) See above
 * 17) *I did specify page numbers. I think a 5-page range is short enough to condense it down to one footnote rather than trying to break them out by pages. Also "space limitations" is a misreading: the limitation is on what glyphs the typesetter can typeset. But the intention of "perhaps in some cases" was to avoid ascribing a particular case to Euler; typesetting limitations was the only reason given by Cajori for these variations but he was vague about whether it was the only reason people had, and I wanted to reflect that in our text. An earlier version of our article was worded in a way that made it appear that Euler's use of a different symbol was a deliberate choice used to reflect a different meaning for how Euler thought of the infinite, which we didn't have any source for, and I also wanted to avoid that editorializing implication. But I'd be happy to take suggestions for how to even more strongly disconnect the reason for some typographic variation from the lack of known reason for Euler's variation, without getting too lost in the weeds. —David Eppstein (talk) 02:48, 19 February 2022 (UTC)
 * 18) Maor 1991: Reliable, missing URL access date, verifies information
 * 19) *See above re access dates.
 * 20) Clegg 2003: Reliable, properly formatted, but it doesn't verify that CIↃ was used for 1,000. Barrow 2008 (p. 340) seems to be usable for this.
 * 21) *Yes it does. "Some have suggested that the symbol is derived from the old Roman sign for 1000 'CIↃ', or from a closed-up version of the last letter of the Greek alphabet, omega (ω).", third-to-last paragraph of Chapter 6. I didn't specify a more precise page number because the version I have isn't marked with page numbers. Incidentally, in case all this disagreement with your comments is coming off as testy, let me add how much I appreciate it that you are actually checking that the references verify what they are claimed to verify. Far too many GA reviewers skip that step and it's an important one. —David Eppstein (talk) 08:50, 19 February 2022 (UTC)
 * ah, funny, the symbol doesn't show up in a Google Books preview. Also, thanks, I do appreciate that :) I think I'm about ready to pass the article, as soon as I finish looking through your responses. theleekycauldron (talk • contribs) (she/they) 02:26, 22 February 2022 (UTC)
 * 1) Euler 1744: This note should probably include the page number in the citation itself; it could also be formatted as a note, but that's less important. It does verify the content. Missing URL access date.
 * 2) *See above with respect to the impossibility of including both the full page range (mandatory) and the specific page in the same journal citation template, and on access-dates for journal papers. —David Eppstein (talk) 19:26, 19 February 2022 (UTC)
 * 3) Barrow 2008: Reliable; Same issue with title vs. contributor as Rucker, and missing URL access date; verifies information
 * 4) *See above for chapter/contributor and access-dates. —David Eppstein (talk) 19:26, 19 February 2022 (UTC)
 * 5) Shipman April 2013: Inaccessible at the moment, although I'll try to get access to taylor & francis via WP:TWL. In the meantime, AGF, looks reliable. Do DOI links need an access date?
 * 6) *I don't think they do; see above. —David Eppstein (talk) 19:26, 19 February 2022 (UTC)
 * 7) Perrin 2007: Reliable source; missing URL access date; looks to verify based on the words i can string together, but I'm not up to speed on algebraic geometry.
 * 8) *We're really using very little algebraic geometry from that source: only "the projective line is a (thing) to which we add a unique point at infinity" and "unique point &infin;" (the same point, but given in notation rather than words). —David Eppstein (talk) 19:26, 19 February 2022 (UTC)
 * 9) Aliprantis et al. 2006: Reliable; missing URL access date; is there a reason the book seems to use tau while the article uses T?
 * 10) *Actually, the book uses X where we use T, but it is a little sloppy about notation. When it is being formal and careful it uses notation like (X,tau) to describe a topological space, where X is the set of points of the space and tau is the family of open sets, but in less-formal text it uses X alone (I guess as synecdoche). So it refers to the one-point completion both carefully as (X_infinity,tau_infinity) and sloppily as X_infinity. Our own article on the one-point extension doesn't use the same notation (it uses X*). And the sentence Aliprantis is used to source is mainly about what to call the one added point, not what to call the space. So I think maybe it's easiest (and better from the point of view of technicality) just to remove the notation from that sentence, leaving "and the point added to a topological space to form its one-point compactification". —David Eppstein (talk) 19:26, 19 February 2022 (UTC)
 * 11) Zboray and Zboray 2000: Reliable; missing URL access date; verifies content
 * 12) *See above re access date. —David Eppstein (talk) 19:34, 19 February 2022 (UTC)
 * 13) Crist et al. 1990: Reliable, properly formatted; inaccessible, AGF for now
 * 14) *I hesitate to suggest that you use pirate sites like Sci-Hub or Z-Library to verify sources (I certainly wouldn't include direct source links to them, even in a discussion page like this one), but if you really want to look at the sources and don't have some sort of institutional subscription, they often provide a working alternative. —David Eppstein (talk) 19:34, 19 February 2022 (UTC)
 * 15) O'Flaherty 1986: Looks reliable enough; missing URL access date; verifies content
 * 16) *See above re access date. —David Eppstein (talk) 19:34, 19 February 2022 (UTC)
 * 17) Toker 1989: Reliable; missing URL access date; verifies content
 * 18) *See above re access date. —David Eppstein (talk) 19:34, 19 February 2022 (UTC)
 * 19) Bahun 2012: Reliable, properly formatted, verifies content
 * 20) *Nothing to add. —David Eppstein (talk) 19:34, 19 February 2022 (UTC)
 * 21) Natalini 2013: Reliable, properly formatted; inaccessible, AGF for now
 * 22) *Nothing to add. —David Eppstein (talk) 19:34, 19 February 2022 (UTC)
 * 23) Healy et al. 2003: Reliable; missing URL access date; specified page doesn't mention 1800s creation, is that on the previous page I can't se?
 * 24) *The specified page says (lines 2-3), quoting someone else, brackets in original: "[The flag] was first used by Metis resistance fighters prior to the Canadian Battle of Seven Oaks in 1816". I think it's fair to describe "first used prior to 1816" (what the source says) as "since the early 19th century" (what our article says). We don't say when it was created; we say that it has been used since a certain time, but that doesn't rule out an even earlier creation date. Our Métis flag article (based on the same original source material) is similarly vague, again saying we don't actually know when it was first created or what the original intent of its design was, only that it was already used before a certain date. Even the intended meaning of the difference between the red and blue versions of the flag is historically unclear. —David Eppstein (talk) 19:52, 19 February 2022 (UTC)
 * 25) Gross 2016: Should I be reading into the fact that this marks itself as not peer reviewed? Properly formatted and would verify content if confirmed reliable
 * 26) *It's a published editorial, more or less equivalent in reliability to a newspaper editorial, but we're only using it for a factual claim so I don't think we need to mark it as opinion. One can also find more recent pop-psych books mentioning this, but that's a genre I mistrust the reliability of in general and using very recent sources risks WP:CIRCULAR problems, so the fact that this one is from 2016, before any attempt to add this material to this article, reassures me in that respect. —David Eppstein (talk) 19:52, 19 February 2022 (UTC)
 * 27) Crespy et al. 2008: Reliable, properly formatted; inaccessible, AGF
 * 28) *Nothing to add. —David Eppstein (talk) 19:52, 19 February 2022 (UTC)
 * 29) Rivkin et al. 2005: Reliable; missing URL access date; verifies content
 * 30) *See above re access date. —David Eppstein (talk) 19:52, 19 February 2022 (UTC)
 * 31) Willmes 2021: On the one hand, people named claudia are notoriously trustworthy. On the other hand, this looks a little... press release-y? Properly formatted and verifies information, at any rate.
 * 32) *I take it you're not a fan of Pamela Dean's The Secret Country trilogy. Yes, it is somewhat promotional as a source, and again an editorial rather than a peer-reviewed journal article. But all we're using it for is the factual claim that the resemblance between this notable organization's logo and an infinity symbol is intentional. —David Eppstein (talk) 19:52, 19 February 2022 (UTC)
 * 33) Al Jazeera 2019: Reliable; creation date format is MM dd, YYYY, missing URL access date; verifies content
 * 34) *It's a newspaper or news publisher (not sure of the distinction nowadays), so a daily rather than monthly date format is the correct choice. —David Eppstein (talk) 19:52, 19 February 2022 (UTC)
 * 35) Compart AG: Reliable, date format is yyyy-mm-dd and I'm not sure AG Compart is a name, verifies content
 * 36) *Grr badly formatted cite web. I particularly dislike cite web templates that repeat the hostname as the name of the web site, but formatting "Compart AG" as a person rather than as a company is worse. Publisher is the company "Compart AG". It doesn't appear to have any person listed as author. The name of the website (what you get if you click on "home" at top left) is "Unicode". The date is an access-date (formatted as numeric in the article date format I am using); I don't see an obvious date that the web page was created and am not convinced that the 2021 copyright date it displays is accurate, so I didn't list a publication date. Fixed now. Good catch. —David Eppstein (talk) 19:52, 19 February 2022 (UTC)
 * 37) OEIS: Looks like a wiki, albeit a selective one? Date format is yyyy-mm-dd, missing link to OEIS (resolved), verifies content
 * 38) *The main part of OEIS has a strong editorial process, but there's also a wiki attached on the side. This source went to the wiki, not the main part. I don't know that there's much editorial control over content there. I'll try looking for a better source; it should be possible to find published book sources for this material. —David Eppstein (talk) 20:17, 19 February 2022 (UTC)
 * 39) *Update: Replaced by Pakin, below. —David Eppstein (talk) 20:38, 19 February 2022 (UTC)
 * 40) Steele 1996: Reliable, date format is yyyy-mm-dd but missing URL access date, verifies content
 * 41) *Added access-date for web source. —David Eppstein (talk) 22:17, 19 February 2022 (UTC)
 * 42) Apple 2005 (2.1+): Reliable, date format is yyyy-mm-dd but missing URL access date, verifies content
 * 43) *Added access-date for web source, checked date formats, and adjusted authorship for unicode sources. —David Eppstein (talk) 22:17, 19 February 2022 (UTC)
 * 44) Apple 2005 (4.0+): Reliable, date format is yyyy-mm-dd but missing URL access date, verifies content
 * 45) *Added access-date for web source, checked date formats, and adjusted authorship for unicode sources. —David Eppstein (talk) 22:17, 19 February 2022 (UTC)
 * 46) Unicode Consortium 1994/2015: Reliable; date format is yyyy-mm-dd, missing URL access date; verifies content
 * 47) *Added access-date for web source, checked date formats, and adjusted authorship for unicode sources. —David Eppstein (talk) 22:17, 19 February 2022 (UTC)
 * 48) Unicode Consortium 2006: Reliable; missing URL access date; verifiable if we wanna stretch WP:PRIMARY
 * 49) *Added access-date for web source, checked date formats, and adjusted authorship for unicode sources. —David Eppstein (talk) 22:17, 19 February 2022 (UTC)
 * 50) Unicode Consortium IBM-970: See above
 * 51) *Added access-date for web source, checked date formats, and adjusted authorship for unicode sources. —David Eppstein (talk) 22:17, 19 February 2022 (UTC)
 * 52) Steele 2000: Reliable; missing URL access date; verifies content
 * 53) *Date format inconsistency with the other Steele reference fixed. Added access-date for web source. —David Eppstein (talk) 22:17, 19 February 2022 (UTC)
 * 54) Unicode Consortium 2011: Reliable; date format is yyyy-mm-dd but missing URL access date; verifies content
 * 55) *Added access-date for web source, checked date formats, and adjusted authorship for unicode sources. —David Eppstein (talk) 22:17, 19 February 2022 (UTC)
 * 56) van Kesteren: Appears to be self published by someone not objectively (formally) a subject matter expert; missing URL access date; fails verification, i can't see the decimal or hex values in there
 * 57) *He is notable enough to have an article on Wikipedia about him, describing him as "an open web standards author"; I think that makes him an "established subject-matter expert, whose work in the relevant field has previously been published by reliable, independent publications". And when I look in his table at row A1 column DB I see the infinity symbol right there. The hex values are the gray labels at the start of the row and the top of the column. Our article gives them in decimal rather than hex but see WP:CALC for the non-requirement of sources for routine calculations like this sort of unit conversion. —David Eppstein (talk) 22:25, 19 February 2022 (UTC)
 * 58) Pakin 2020: mmmm, primary, but looks reliable enough; date format is yyyy-mm-dd but missing URL access date
 * 59) verifies content
 * 60) verifies content
 * 61) *In what sense is Pakin primary? See WP:PSTS. It is a book about LaTeX, written by someone was not directly involved in developing LaTeX. It looks secondary to me. The corresponding primary source for the same information would be the stix package, where most of these symbols are defined, and its documentation. It appears not to be published in the traditional way by a print publisher (instead appearing in CTAN) but it appears that CTAN exercises at least some level of editorial control over what it publishes. So the question should be, is it reliably published rather than is it primary. It's possible that some of the same content can be found in a similar book, Typesetting Mathematics with LaTeX by Herbert Voss, but I didn't find an accessible version of this that matched its publication information, and I'm not convinced that being published by UIT Cambridge is any more respectable than being published by CTAN. I did add a specific page reference within this since otherwise it's quite a long source (hundreds of pages) to search through for the relevant parts. —David Eppstein (talk) 20:38, 19 February 2022 (UTC)
 * 62) Joypixels: Looks to be self-published, iffy on reliability; URL access date is missing and it should probably link here; would verify content if reliable
 * 63) *Agree on unreliability. Searched for better but emoji is a maze of bad sources and I didn't find anything I considered good. Removed. —David Eppstein (talk) 23:01, 19 February 2022 (UTC)
 * 64) Unicode on Github: Primary and reliable; URL access date is missing and this should probably signal that the website is github while the repo is cldr; verifies content
 * 65) *I added GitHub to the citations that link to GitHub. It's not the website, and it's not the publisher (that's the Unicode Consortium), but it's the web provider through whom the published content can be found; that's the role of the via parameter. —David Eppstein (talk) 23:01, 19 February 2022 (UTC)
 * 66) Unicode consortium: Primary and reliable; date format is dd MM yyyy; verifies content
 * 67) *See above for edits made to all Unicode sources. —David Eppstein (talk) 23:01, 19 February 2022 (UTC)

Imagery

 * This won't be disqualifying for GA, as far as I can tell, but it appears you've got an MOS:SANDWICH issue with your image placement.
 * I tried replacing the left-right image placement with a multiple image to improve this issue. This article has a fairly high ratio of images to text making placement tricky. —David Eppstein (talk) 17:32, 11 February 2022 (UTC)
 * All images (with the possible exception of the tarot deck; see sourcing section) are relevant, well-captioned, freely licensed, and clear, but they need alt text. theleekycauldron (talk • contribs) (she/they) 08:05, 17 February 2022 (UTC)
 * Tarot image caption sourced; alt text added to actual images. (The thing that appears to be an image in the infobox is I think actually a math formula. Is there any way to make alt text visible, for instance on mouseover, in browsers that display images properly? It seems silly to require me to add something I can't actually see and can't check that I have added correctly.) But although adding alt text is demanded by MOS:ALT, and is a helpful suggestion, requiring its addition as part of the GA criteria has been explicitly rejected in past discussions. See —David Eppstein (talk) 21:17, 19 February 2022 (UTC)

Overall
Possibly not done with adding comments, but I'm comfortable putting this for now. Fantastic work! theleekycauldron (talk • contribs) (she/they) 08:27, 17 February 2022 (UTC)
 * All comments responded to. Please take another look. —David Eppstein (talk) 23:01, 19 February 2022 (UTC)
 * all right, your responses seem to pretty much check out! Thank you, by the way, I've learned quite a bit about the GA criteria through this one :) passing the nomination. theleekycauldron (talk • contribs) (she/they) 02:36, 22 February 2022 (UTC)
 * Thanks! —David Eppstein (talk) 06:49, 22 February 2022 (UTC)