This page is within the scope ofWikiProject Chemistry, a collaborative effort to improve the coverage ofchemistry on Wikipedia. If you would like to participate, please visit the project page, where you can jointhe discussion and see a list of open tasks.ChemistryWikipedia:WikiProject ChemistryTemplate:WikiProject ChemistryChemistry
The article will be discussed atWikipedia:Articles for deletion/1,5-Diisocyanonaphthalene until a consensus is reached, and anyone, including you, is welcome to contribute to the discussion. The nomination will explain the policies and guidelines which are of concern. The discussion focuses on high-quality evidence and our policies and guidelines.
Users may edit the article during the discussion, including to improve the article to address concerns raised in the discussion. However, do not remove the article-for-deletion notice from the top of the article until the discussion has finished.
Lately I've been thinking about having some kind of framework for notability of compounds, as seen for other topics inWikiProject notability advice.WP:NUMBER is the sort of framework I would take inspiration from: a series of questions that can loosely establish either notability or a lack thereof. I'm inclined to lean for inclusion rather than against in most edge cases, but I'm sure people disagree - and hashing out the concept in general can be a useful reference for those cases as they arise.Fishsicles (talk)14:02, 15 September 2025 (UTC)[reply]
Developing some consensus on notability of chemical compounds is an admirable idea but probably a doomed effort. Often it seems, when a very chemical theme is AfD'd, suddenly out of nowhere arrive editors thoroughly knowledgeable of rules and regs but ignorant of the chemical literature. Any effort would be thwarted by these virtuous souls.Anyway, it's not that big of a deal because dumb articles don't interfere with our main mission.--Smokefoot (talk)18:03, 15 September 2025 (UTC)[reply]
The only way to change the rules and regs is to propose an alternative, though. If all we can cite are the informal expectations of different WP:CHEM editors, nobody's going to really pay much mind - but having an internal standard is the best first step to suggesting actual official guidelines/policies through the proposal process.Fishsicles (talk)12:54, 16 September 2025 (UTC)[reply]
[Writing this when I'm tired, hopefully this isn't too long-winded and makes enough sense]
Thanks @Fishsicles, I'm glad you raised this. Ultimately, I think it would be helpful to have aWP:SNG for (non-medical) scientific topics. I agree the best place for WP:CHEM to start would be to develop a topic-specificWikiProject notability guidance page, albeit these tend to be less influential for AfDs (as they aretreated asessays [that] do not establish new notability standards, lacking the weight of broad consensus of the general and subject-specific notability guidelines).
The usefulness and distinctiveness criteria mentioned in your19:07, 5 September 2025 userpage note seem reasonable to me. I also came across the essayWP:How Wikipedia notability works (full disclosure: authored by an unsuccessful RfA election candidate), which argues that, in practice, WP assesses notability using three broad criteria: source availability, importance, and encyclopedicness. It suggests SNGs can help with assessments of importance / encyclopedicness, and can sometimes alsoset up alternate criteria [or alternate assessment systems for criteria] for topicswhere GNG is not commonly used or usable.
As per @Smokefoot, I think low-quality articles on obscure topics should be relatively harmless as long as they don't violate other core policies/practices (e.g. FRINGE, BLP, POV forks). But notability guidelines also play a role in training/shaping WP's community of editors. I gather there have been cases in the past (e.g. ROADS) where subcommunities of editors have diverged far enough from the expected norms, for long enough, that the inevitable correction has caused significant conflict, disruption, and editor loss (understandably so, when editors have put so much work into articles prior to theiren masse removal). To quote fromSlate's article on WP:ROADS:
One flashpoint that inspired the recent revolt was a strict interpretation of the site’sreliable sourcing policy. ... A volunteer editor seeking to contribute content to the page might use information found on the West Virginia Department of Transportation’s website as a reference source. However, policy sticklers are likely to deny this usage because DOT is a primary source for highways ... According to the site’s policies, Wikipedia should be based mainly on reliable secondary sources, such as newspapers. ... Why not allow Wikipedians to cite from DOT, which is responsible for publishing highway routes? ... it’s worth remembering that most of the time Wikipedia has good reasons for the prohibition against primary sources, especially with government entities. ... The question is whether there is some way to recognize an exemption, granting that some types of primary sources may be reliable while still protecting the integrity of the rule.
In particular, I was struck by the recent AfD comment (from a fellow science editor)We have a long standing general consensus to keep stuff covered by multiple reliable academic sources, even if they are considered technically primary. I can imagine situations where this would be warranted. E.g. if secondary sources exist, indicate a topic is notable, do not provide sufficient detail to write a non-stub article, but multiple independent primary sources exist and are broadly in agreement (a scenario which is much more likely for scientific topics), it should be possible to collate information from primary sources without falling foul ofWP:SYNTH orWP:DUEWEIGHT. But the more frequently we deviate from GNG guidelines, the more important I think it is to explain/codify our reasoning. I'd like us to start doing that, beyond the limited scope for discussion provided by individual AfDs.Preimage (talk)07:23, 17 September 2025 (UTC)[reply]
Might be interesting to bring in the supporters of1,5-Diisocyanonaphthalene:@Cyclopia,Dclemens1971,My very best wishes, andVery Polite Person:. I retract my comment that dumb articles (e.g.,1,5-Diisocyanonaphthalene) are harmless. They become intertwined (linked, cited) into legit articles, which confuses readers who think that we are providing notable information vs some lame crap. The other thing is that this exercise in self-citations encourages losers to dump more %$& onto Wikipedia in an effort to gain a modicum notability as authors (bibiometric tools at journal sites recognize Wikipedia mentions). As one criterion for the proposed notability would be that the article is not based on self-citation thing, as is1,5-Diisocyanonaphthalene.--Smokefoot (talk)14:46, 17 September 2025 (UTC)[reply]
My (quick) $0.02: whatever standards are set on a subject-matter level like this for more particular notability conditions, are, remember, supplemental or alternative paths TO notability, or guidance on interpretation of notability againstWP:GNG,WP:BASIC, et al. A per-topic guidance framework, even ridiculously over-the-top ones likeWP:FRINGE, are wholly subservient and lessor to, in all circumstances,WP:GNG. EvenWP:MEDRS, which anyone sane supports, will factually lose 100% of battles in a straight up fight vs roteWP:GNG (as it should; as should any of them).
If the question is: can we make it a little harder to get stuff into Wikipedia in some context like this? The answer is and must be no.
If the question is: can we make it more clear what is consideredWP:RS for things derivative of or created bychemical synthesis broadly defined? Sure, lots of topics do that. You'd just need to carve up and build out ideas, and then refine them in a few RFC type fires. Just remember: no matter who tight you may or may not try to make it, anyone, me, you, a stranger, can blow up any opposition by simply dropping severalWP:SIGCOV levelWP:RS, and walk away whistling. I've done that on a few AfDs that I suspect left heads scratching. —Very Polite Person (talk/contribs)14:56, 17 September 2025 (UTC)[reply]
The comment above that is really delicious (or kinda self-canonization): "If the question is: can we make it a little harder to get stuff into Wikipedia in some context like this? The answer is and must be no." Also, one might add, have you stopped beating your wife? Which books have your burned today? And in what ways do you support slavery?--Smokefoot (talk)16:10, 17 September 2025 (UTC)[reply]
Huh? I'm supporting the idea of refining the standards or carving them out. I was just explaining that if it was the goal to make it more restrictive (which I'm not objecting to) then it would be around defining what counts as RS in the first place, like how MEDRS handles it. Just emulate that approach. I was just saying it can't be done by making it simply harder from a GNG/BASIC POV. —Very Polite Person (talk/contribs)16:15, 17 September 2025 (UTC)[reply]
Of course rules being discussed would make it more difficult to "get stuff into Wikipedia." What else could a rule do? The effect would be incremental and net beneficial, at least that is the intent. Yes, one always worries about "slippery slopes" etc. But anyway, sincere thanks for the advice. --Smokefoot (talk)16:20, 17 September 2025 (UTC)[reply]
I'm 99.872% sure we're on the same page and would prefer the same outcome. I was just trying to say the smoothest route would be the chemistry equivalent of MEDRS. Launch your ship from Miami aimed for Lisbon, but tweak the heading a fraction of a degree off and you end up in the Med. HavingWP:CHEMRS (I will !voteSupport forWP:CHEMRS to be the name) would help make sure most ships get to Lisbon or their intended destination--accuracy.
I think some specific guideline might be helpful. Obviously, not everything inPubChem qualify for inclusion. One can reasonably argue that a compound should be included if (a) it is described in several publications, and (b) there is a substantial information about the compound to be included in the page. Speaking on1,5-Diisocyanonaphthalene, I think it satisfies (a). As about (b), this is something debatable. One can say it does not satisfy (b). One of important criteria could be inclusion of a compound into tertiary sources, such as encyclopedia or even reputable databases. Would everything inMerck Index qualify automatically for inclusion? Maybe. Everything from theHuman Metabolome Database, i.e. [[1]]? I am not sure. Everything fromDrugBank? Yes, probably. Assuming that there are also publications about the compound in scientific journals.My very best wishes (talk)17:01, 17 September 2025 (UTC)[reply]
I think what needs to be codified most is when primary sources are appropriate for establishing notability. In particular, I would say primary sources that areabout a compound itself are less important than primary sourcesusing a compound to achieve some end.
To use one of my own AfCs as an example,sodium tetrapropylborate: I'd started looking into it mostly on a whim as it is isoelectronic totetrapropylammonium. The De Smaele et al ref (doi:10.1016/S0021-9673(97)00886-8) isabout the compound: if this was the only paper I'd found going into it in any depth, I'd probably have ditched the draft. Chemists make new compounds with Amazing Potential™ all the time and I'm not interested in hyping up someone else's grant applications. It's useful for establishing properties of the compound, but not establishing notability.
Huang et al (doi:10.1016/j.scitotenv.2004.04.015) hasa specific use of the compound (and establishes that it's commercially available to researchers, a fact that in of itself raises notability to me) - this isn't a silver bullet, but is a much stronger notability point even if it doesn't have as many topic-specific details compared to theabout paper. Schubert et al (doi:10.1007/s002160050072) isabout using the compound, and moreover a comparison to a related compound - this sort of paper is to me the "strongest" kind of source for notability short of a proper literature review.
I agree withVery Polite Person: «If the question is: can we make it a little harder to get stuff into Wikipedia in some context like this? The answer is and must be no.» As they correctly stated, any further guideline would only supplement or clarifyWP:GNG, not substitute it. That said, I reject the concept that (sourced) obscure compounds are not encyclopedic or even "lame crap". When I first started contributing to WP more than 20 years ago, I did it for itsWP:PURPOSE: «free access to the sum of all human knowledge». If there are reliable sources on anything, be it a chemical compound or else, we can include it, and if we can include it weshould include it.--cyclopiaspeak!10:35, 18 September 2025 (UTC)[reply]
I think the problem is thatWP:GNG is rooted firmly in secondary sources, and that there's not good secondary sources for many compounds that I would argue are notable. The question that is most relevant to this issue is "what kind of primary sources establish notability for chemical compounds, and to what degree?". Many chemical articles are rooted mostly or solely in primary sources, which would makenone of them notable under a strict reading of WP:GNG - which is obviously not the goal. The question I think is not "How do we make it harder to get things into Wikipedia (compared to GNG)?" but rather the concurrent questions "How much easier should we make it to get things into Wikipedia (compared to GNG)?" and "How much harder should we make it to get things into Wikipedia (compared to current informal process)?"Fishsicles (talk)12:56, 18 September 2025 (UTC)[reply]
It ought to be relatively straightforward. If you could amendWP:RS directly just for purposes of whatsort or caliber of primary sources is sufficient for defining notability of chemical substances on Wikipedia...
What's the gold standard? What's the minimum floor? What specific two sources, as an example?
Tailor a version 0 draft against that: ID examples. Agree on two examples. Tailor initial outline and sketch describing sources that would include both of those sources, but "likely" cut off anythingbeneath the floor source. —Very Polite Person (talk/contribs)14:48, 18 September 2025 (UTC)[reply]
Thanks for pointing this,Fishsicles. I was a bit carried out becausede facto, academic papers, while being formally primary sources, are considered equally as secondary across a wide range of scientific fields. Species, for example, or astronomical objects.WP:NASTCRIT could be a starting point if we need to make this explicit.cyclopiaspeak!20:39, 18 September 2025 (UTC)[reply]
I agree withWP:NASTCRIT as a solid foundation, though there's obviously some practical differences.
The object is, or has been, visible to the naked eye.
This one to me says "things should be significant outside of the field's interest in discovering new things". I would probably propose "the compound is or was commercally available in meaningful volume".
The object is listed in a catalogue of high historical importance
There's notreally historical equivalent of astronomical catalogues for chemicals, as it took us a lot longer to figure out what chemicals were, but some level of "known and significant in historical records" is probably worth having. Is there a WikiProject Alchemy?
The object has been the subject of multiple non-trivial published works, which contain significant commentary on the object.
This is probably where the haggling is to be done. I'd say "multiple non-trivial published works" needs a chemical caveat of "from different labs" - in particular, while a source on the first preparation or characterisation of a compound is usefulfor preparation and properties of that compound, I don't think it should be counted towardsnotability.
The object was discovered before 1850, prior to the use of astrophotography or automated technology.
I don't think this directly translates at all as it's hard to "automate" discovery of chemicals. I suppose "discovered prior to modern chemistry" is also a criterion for notability, but that fits more into the "significant in history" rather than its own line.
Part of the issue is what does "primary" mean. Because most of our articles are about people or organisations, we don't want to use sources where people write about themselves, or where organisations issue a press release about themselves. These are primary sources. However chemicals do not write about themselves, and theoretically at least have some existence. However if one person writes about a chemical, we would want it to be confirmed by another writer. Many chemical articles have a mini-review at the start, summarizing previous work. So I would recommend that notability is proved where independent people write about that chemical topic. Even for chemicals that don't exist, but are in chemists imagination, there may be notability if several articles are on the topic. However most of the prod and AFDs I have started for chemicals are for those that do not exist. On this board we have discussed some unsuitable sources. These would not be reliable sources. Some databases such as PubChem and ChemSpider do not prove notability, as they have entries for all sorts of stuff, including errors and things that do not exist.Graeme Bartlett (talk)06:00, 19 September 2025 (UTC)[reply]
Yes, the inclusion to PubChem and ChemSpider do not prove notability of any compounds, although the inclusion to DrugBank might be. The only reason for keeping1,5-Diisocyanonaphthalene could be its studied anti-fungal activity. This is on the borderline of notability, I agree. It could be deleted or kept, whatever.My very best wishes (talk)17:52, 20 September 2025 (UTC)[reply]
Not to relitigate a recent AfD too much, but I'd argue againstpotential applications as a criterion for notability unless very widely studied. Everything has potential applications. If a compound can be developed to the extent that it's competitive with existing compounds, it should get an article once that hits the market, not while it's in basic research.
I think the goal of this effort should not be to make it harder (or easier) to get articles into WP, but rather easier to determine notability in our specific subject area, and harder for articles that meet our specific standards to get AfDeleted. I agree that a key part of this is determining RS for our subject area, and to codify appropriate use of primary sources. Eventually this should make it into a Chen-specific RS guideline and N guideline, to be accepted as part of the general WP standards. But a good first step is to hash something out in our project space so we as a project are agreed on how the general WP standards should be enhanced and we are reasonably confident that it would be accepted as part of the general standards.YBG (talk)06:38, 23 September 2025 (UTC)[reply]
A perceptive editor complained that the pKa listed fordiphenylamine is really the pKa of its conjugate acid. That kind of "you know what I mean" thing can be confusing. So, I doctored up the entry in the chembox (format police allowed that), but there may be a more elegant approach to this problem (definining that pKa for amines often refers to conjugate acid). Also, I wonder how pervasive is this issue? --Smokefoot (talk)18:15, 15 September 2025 (UTC)[reply]
The{{chembox}} template accepts pKb. Couldn't that be used instead? Seems better to report the basicity of a base, rather than the acidity of the base's conjugate acid.Marbletan (talk)19:40, 15 September 2025 (UTC)[reply]
Isn't organic chemistry almost universally taught from the perspective of pKa for acid-base equilibria? For that, I think it's a little more useful to record a pKa. Though nothing really wrong with reporting both (likeEthylamine), unless space is at a premium.
In this case I would format it as:0.79 (conjugate acid,Ph2NH+2), but that is basically the same thing. I like the addition of the chemical formula for clarity, most of the time it is omitted likeAniline.⇌Synpath01:08, 16 September 2025 (UTC)[reply]
I would say the best solution is to have a separate argument in chembox that determines if the pKa line is titled as just "pKa" or as "pKa (conj. acid)" for amines etc. (It should probably also have a check requiring that if the "conjugate pKa" line is used, a conjugate acid formula is provided in the chembox.)Fishsicles (talk)12:59, 16 September 2025 (UTC)[reply]
Isn't this just a semantics issue ... for both pKa and pKb it is not about the acid or its conjugate base, it is about the equilibrium between the two and a proton in a certain media. For the chembox, that is generally water at 25°C, where it is then an interplay between the subject acid and subject base and not a 'free' proton but the 'hydrated' proton. I think that the link toacidity in front of it does it all, it does not need the explanation.Dirk BeetstraTC07:40, 17 September 2025 (UTC)[reply]
Yep, as things stand, there are a heap of inconsistencies between WP articles. I'll start with a quick look at simple polyprotic acids.Sulfuric acid's infobox readspKa1 = −2.8 [line break]pKa2 = 1.99, whereasphosphoric acid has no mention of any pKa values in its infobox, instead reporting pKa1 to pKa3 in the article body.
Hydrazine's infobox readspKa = 8.10 ([N2H5]+) [with citation] andpKb = 5.90, with article body reporting pKb values for bothN2H4 and[N2H5]+. (It would probably be an improvement to refer to these as pKb1 and pKb2 for hydrazine.)
The most complicated example I found wasarginine, whose infobox currently readspKa = 2.18 (carboxyl), 9.09 (amino), 13.8 (guanidino). While this conflates the pKa of arginine with its conjugate acid and base, it's the cleanest way of presenting its acid-base (buffer) properties. I get the argument we should be reporting pKb for bases rather than pKa for their conjugate acids, but for arginine that would mean we'd have to list pKb1(amino), pKb2(guanidino), and pKa(carboxyl), which I think is going to be more confusing in practice. (And theisoelectric point is going to be more of a pain to calculate if we do things this way.)
If we needed to be absolutely precise, I guess we could say the conjugate acid of arginine (ArgH+) is a triprotic acid, with pKa1(ArgH+) = 2.18 (carboxyl:−COOH ⇌ −COO− + H+), pKa2(ArgH+) = 9.09 (amino:−NH+3 ⇌ −NH2 + H+), and pKa3(ArgH+) = 13.8 (guanidino:−NHC(NH2)+2 ⇌ −NHC(NH)NH2 + H+). Though I doubt that's going to fit in the infobox :)Preimage (talk)14:17, 17 September 2025 (UTC)[reply]
I like the idea of having an annotation in the infobox if the value is actually for the conjugate acid. I strongly oppose my interpretation of Fishsicles's parenthetical about "formula is provided in the chembox" because we universally try to avoid including details of "compound B" in the infobox for "compound A" on the article that is focused on compound A, and especially because it's only for purposes of clarifying one infobox item. Instead, I'd support a parenthetical or similar note specifically in that one infobox item, such as "conjugate acid" in the title (which can be usefully linked), and something more specific (parenthetical on the value) if there are several to distinguish from each other.DMacks (talk)18:29, 17 September 2025 (UTC)[reply]
My parenthetical is more in reference the existingTemplate:Chembox Properties conjugate acid/base entry, not an additional field. If the pKa is defined as that of the conjugate acid, it should probably warn if there's no entry on the conjugate acid in the template.Fishsicles (talk)13:06, 18 September 2025 (UTC)[reply]
I've been cleaning up the chemical categories on Simple English, with a goal of matching (where enough pages exist) the categorisation here on enwiki. One thing I've noticed, though, is that we countCategory:Diazepines underCategory:Aromatic nitrogen heterocycles. Now, it's been a while since I did heteroaromatics in any particular detail, but this fails the Huckel smell test to me - shouldn't diazepines be antiaromatic? They've got two more pi electrons than imidazole. Is there a corner case here I'm forgetting re: the lone pair on the NH?Fishsicles (talk)11:48, 24 September 2025 (UTC)[reply]
The enwikiCategory:Imides is categorised underCategory:Lactams. While many imides of interestare cyclic, and AFAIK IUPAC nomenclature uses "diacylamine" for acyclic imides, the functional group itself isdefined as including anyRCO−NR−COR (but "especially" cyclic compounds). Wiki pages seem inconsistent as to which standard is applied.
The main pages for bothimide anddiacetamide refer to the latter as an imide, but it is only classified underCategory:Acetamides.Phenacemide, meanwhile, is an acyclic imide thatis in the category - and is very much not a lactam.
For Wikipedia purposes, isCategory:Imides meant to contain all di/triacylamines, or only cyclic ones? If the former, it should not be classified under lactams, and pruned of the acyclic compounds; if the latter, should we haveCategory:Cyclic imides that implies lactams? In either case, the category description(s) should probably specify.Fishsicles (talk)15:51, 29 September 2025 (UTC)[reply]
I think we should follow IUPAC, where "imide" is not necessarily cyclic; having an subcat "cyclic imides" intersection of "lactams"+"imides", each of which are subcats of "carboxamides" makes sense. And also per IUPAC, "imide" is specificallydiacyl. If we have anytriacyl articles, that could go in a sister-cat of "imides" as a subcat of "carboxamides".— Precedingunsigned comment added byDMacks (talk •contribs)16:14, 29 September 2025 (UTC)[reply]
Are cylcic imides even lactams? Can a cyclic anhydride be a lactone, a linear anhydride as an ester? I doubt it is useful to subdivide functional groups like this.⇌Synpath18:39, 29 September 2025 (UTC)[reply]
The convention I am familiar with is that "lactams and cyclic imides" can be grouped together like that, rather than the latter as a type of the former. That said, it's not something I can point to a particular definition for, so I'm inclined to bow to existing wiki taxonomy until a more formal definition shows up one way or another.
If I were writing from scratch, I'd be inclined to keep them separate, on the grounds that we don't normally call amides amines even if it still has theC−N bond - but again, nothing formal I can point to that decides things "properly".Fishsicles (talk)18:51, 29 September 2025 (UTC)[reply]
Indeed by functional-group definition, acid-anhydrides are not esters, and imides are not amides and neither of them are amines (amides and imides arederivatives of amines). But many specific compounds are named as if they were, via the derivative/substituent concept (example:N-acylethanolamine is an amide). How about using Commons cats as a basis, which means easier interwiki mapping.DMacks (talk)03:01, 30 September 2025 (UTC)[reply]
Commons cats tosses a new wrench into the works: they don't put imides under lactams, but theydo call imides secondary amides. Which only makes sense to me (mathematician brain wants universal rules) if amides are considered amines. (And I agree they're not!)
I think chemistry categories in general have an issue of not distinguishing "member of a class" from "derivative of a class". I've noticed a few hydrogenated heterocycles being grouped under their aromatic parent, withlevamisole being a recent case - it was categorised underCategory:Imidazothiazoles, despite both rings being partly or wholly saturated. It's clearly aderivative of an imidazothiazole, but it's equally clearly not one itself.
I haven't gone on a whole big hunt for them because I suspect that the terms get fuzzy when looking at chemical definitions versus pharmacological definitions? But still, iftetrahydrofuran isn't inCategory:Furans (and I don't think it should be!), complex hydrogenated aromatics also shouldn't be in their categories from where I sit.Fishsicles (talk)12:54, 30 September 2025 (UTC)[reply]
I took it upon myself to clean up some of the subcategories ofCategory:Esters, grouping them based on the shared component, and noticedCategory:Glucuronide esters. Asglucuronic acid is polyfunctional, I went to check if this was "esters of the carboxylic acid group on glucuronic acid" or "esters of the hydroxyls on glucuronic acid", only to discover the answer was apparently neither: a randomly selected page,codeine-6-glucuronide, showed me that the compound in question was aglycoside of glucuronic acid, without any visible ester groups. Dittoestradiol 3-glucuronide,estrone glucuronide,scutellarin.
Are glycosides esters, rather than ethers? Is there something special about glucuronic acid derivatives specifically that warrants classifying these seemingly esterless compounds as esters? Am I losing my mind?Fishsicles (talk)18:06, 30 September 2025 (UTC)[reply]
I thinkCategory:Glucuronide esters is a category filled with incorrect entries - glucuronidation is a glycosidation reaction, not an ester-forming one. Glucuronides have a free carboxylic acid group and could in principle form esters, but glucuronides themselves are not esters. I suspect people got confused about which end of glucuronic acid forms the chemical bond in a glucuronide. I don't see any actual glucuronic acidester article inCategory:Glucuronide estersJo-Jo Eumerus (talk)07:05, 1 October 2025 (UTC)[reply]
As summarized atTalk:Carbohydrate:Glycobiology redirects toCarbohydrate,Glycan redirects toGlycoconjugate (even though glycan is the same thing as a polysaccharide, workers in the area seem to use the term in the context of hybrids such as glycoprotein/glycolipids), most mono-, di-, oligo-, andpolysaccharide has been pulled out ofcarbohydrate and redirected to those four articles, history was mostly removed fromsugar tohistory of sugar.Carbohydrate is more of an overview with emphasis on food.
Our portfolio in this area is hopefully less confusing. The nomenclature seems almost as complicated as the structures. One problem is that sucrose and glucose are dominant on the minds of most editors, but they may not so dominant in biochemistry and biology. A related challenge is contextualizing the role ofhigh fructose corn syrup. Other editors are welcome to comment.--Smokefoot (talk)19:11, 10 October 2025 (UTC)[reply]
I noticed most of theWikipedia categories named after chemical compounds are in the set categories of their corresponding chemical compound. This introduces issues likethis guy being considered an organic compound with 1 carbon atom, as his article, through a long chain of categories, is in the methane category. This is whyset categories shouldn't mix with topic categories and we should remove all topic categories from the chemical set categories. I'd like some help chipping away at these, so I'm doing a post here.HansVonStuttgart (talk)10:11, 2 October 2025 (UTC)[reply]
Is this an official WP policy? If not, I suspect that strictly enforcing this separation will cause more issues than it solves. While I support strong organisation of categories within a topic (e.g. my multiple posts earlier about functional group categorisation), those are non-trivial distinctions relevant to someone who might be browsing e.g.Category:Esters. If someone is looking at a category describing chemical compounds, and gets a list of chemical compounds, it's important that all those chemical compounds are accurate! But anyone who is looking into a list of compounds should know to exclude a man from Liverpool unless they are going into existential questions. On the other hand, someone looking for e.g. general background on alkanes probably doesn't want to look in two different places, one strictly for their chemistry and one for everything else. To me, you should be able to get to a category named Methane from a category named Hydrocarbons, because the category Methane is about a hydrocarbon even if everything in it isn't a hydrocarbon itself.Fishsicles (talk)12:57, 2 October 2025 (UTC)[reply]
No, it is a guideline. For that use case it's convenient to have those articles there, but it pollutes search results with deepcat and other automated browsing methods.HansVonStuttgart (talk)14:06, 2 October 2025 (UTC)[reply]
This strikes me as the responsibility of the developers of the automated tool. In my opinion, the concern of contributors should always first and foremost be people browsing Wikipedia directly. If the set category/topic category distinction is important for the backend, there should be a way to explicitly distinguish them in the frontend, e.g. an argument in the[[Category:]] tag, that does not impose restrictions on content and organisation from the user side.Fishsicles (talk)14:58, 3 October 2025 (UTC)[reply]
Searchability is important for maintaining Wikipedia. I was searching through articles about chemicals to try and find ones without structure diagrams, but there were so many people, products and companies in the results that I was having a hard time. It seems to me that the most painless way to fix issues like this is to take these 100 or so topic categories out of set categories, rather than develop a new argument for the category tag that then needs to be added to all the same pages.HansVonStuttgart (talk)19:07, 3 October 2025 (UTC)[reply]
It seems to me the problem is downstream (uh, no pun intended) of poor categorization: "natural gas" is in category "methane", when TBH we should probably have "methane" in category "natural gas". TBH a simple fix might be to scan for all pairs (topic categoryt, set categorys) witht ∈s, and then swap them so thats ∈t. Is that too simplistic? Do you know any other examples of this problem?Bernanke's Crossbow (talk)14:21, 19 October 2025 (UTC)[reply]
The main pagecrotyl group makes some interesting but unsourced claims of delocalisation in the carbanion, which makes intuitive sense to me, though though I'd expect the resonance forms to be the allyl anion-esqueR−CH2−CH=CH−CH−2 andR−CH2−CH−−CH=CH2, rather than involve "both terminal carbons" as the text describes (I'm assuming they are counting the carbon bound to R as "terminal"?) If this is just the same structure as any other allylic carbanion, I doubt it needs a dedicated section on a dedicated page. Crotylation appears to be a reaction with some presence in the literature, but I've not dived deeply into its applications.
This page was drafted earlier this year by a user who is likely a student of (now Nobel Prize-winner)Omar M. Yaghi.Jannatulbaqi (talk·contribs) moved it to mainspace, but that account was banned as a sock and had a pattern of accepting AfCs with no apparent review (as described onWikipedia:Sockpuppet investigations/Darshak.parmar/Archive). The existing content is very promotional and looks more like a grant application than a Wikipedia page. I believe it is too late to redraftify the page. I'm not sure what do do about this.
Omar M. Yaghi has a history of COI edits and the entire Research section needs some help. Maybe an expert, maybe just someone who understands this field of chemistry more than I do.Apocheir (talk)21:37, 9 October 2025 (UTC)[reply]
It is very obviously AI, with both Sapling.ai & Quillbot being quite certain. However, I don't see blatantAI hallucinations, but I have not checked the sources in detail -- I to do not know the topic. If someone does check and finds evidence for hallucinations then a CSD becomes possible. Otherwise there is the question of whether, in terms of content, this is aWP:CFORK ofMetal-organic framework in which case you could send to AfD.Ldm1954 (talk)22:12, 9 October 2025 (UTC)[reply]
AfD just filed. I have had it up to my eyebrows with science-themed LLM slop on Wikipedia. We have much better things to do with our time than hunt down each and every claim to make sure the cited source actually said it. If the article's creator cares enough about the topic to write an article, let them write the article properly.WeirdNAnnoyed (talk)22:31, 9 October 2025 (UTC)[reply]
You will require luck in deleting this thing. Here is the pattern: out of nowhere, editors will arrive at the AfD who are steeped in rules, who know no chemistry, and who are dedicated to an iota of truthiness. They will out-vote and out-argue the best of us.--Smokefoot (talk)20:08, 10 October 2025 (UTC)[reply]
He is a living person. Specially a famous living person with a Nobel Prize (that makes it high importance in the project scale). Everything in his article should be backed by a reference.--ReyHahn (talk)21:10, 10 October 2025 (UTC)[reply]
Kitagawa is looking great now, there is only one template left. Can somebody also help withOmar M. Yaghi? It has sections based only on his or other research without secondary sources. Can somebody either remove part of that content or back it up?--ReyHahn (talk)09:02, 11 October 2025 (UTC)[reply]
In June 2009,user:Theinfamousfinatic uploadedthis photo and called it "Hydrogen.JPG", then added it to the article on hydrogen, then removed it two minutes later. They last edited in 2015.
Based on their other edits, I'm confident they were well-intentioned, and most likely took the photo themself in a high school chem lab.
Does anyone have any idea what specifically the people in the photo are doing, and thence what would be a better name for the file such that I could transfer it to Commons? Or is it not worth it and I should just delete it?DS (talk)23:17, 11 October 2025 (UTC)[reply]
The lack of protective equipment alone would suggest that this had better bedry ice orliquid nitrogen.
Hi. I'm in no way a chemist, but started a sandbox version of a new article on the mineral Tsugaruite here:User talk:MisawaSakura/Sandbox2. I could use some help beefing it up before moving to mainspace. Feel free to directly edit my sandbox. Two questions: 1) are there any practical uses for this mineral at the moment and 2) is mindat considered a reliable source?Thanks!MisawaSakura (talk)18:27, 17 October 2025 (UTC)[reply]
Pb, As, S, and Cl are abundantly available as by-products from higher value minerals so I doubt this one has any special demand as an ore if you aren't turning anything up for unique structural properties.
What is the structure of those non-metals As S and Cl? It is an unusual combination, so it would be good to describe the nature of ions and covalent bonds.Graeme Bartlett (talk)07:33, 18 October 2025 (UTC)[reply]
I don't have access todoi:10.3749/canmin.2000005 to check details, but the abstract suggests the structure is Very Complicated - 8x15x38 is ahuge unit cell even for a zeolite, let alone something without silicon or oxygen. I have to assume sulfur's doing a lot of the heavy structural lifting.Fishsicles (talk)02:36, 19 October 2025 (UTC)[reply]
I will say that even if the structureseems notable - and I'll restate that it certainly seemsunusual, going off the numbers I can see - an article claiming such is probably going to run afoul ofWP:SYNTH unless you can find a review or something that discusses its peculiarities in wider context. Hunting further resources turned updoi:10.1127/0935-1221/2008/0020-1778, which doesn't have too much to say but is at least a better citation for "this seems weird" than Wikipedia User Fishsicles.Fishsicles (talk)03:10, 19 October 2025 (UTC)[reply]
Fishsicles Thanks again! I've never had an intention of trying to write a robust detailed article on this as I know it's beyond my knowledge level. But I am knowledgeable enough to write a short stub/start article that meets the DYK requirement of 1500 characters of prose, which I've now done. I also find this particular mineral very interesting, for some reason. The full text of the doi link you just found ishere, in which it says "The crystal structure of tsugaruite is unknown, but is probably complex as is indicated by the large unit cell(V=4678 Å3).3". Note that article is 13 years older than the one I emailed Biagioni for. I hope Cristian Biagioni sends me his article because I can only find an abstract of that one. For now I have enough for a DYK and I could add a couple more sentences if Biagioni sends me his article (I'll let you know if he does); but beyond that a science person would have to work the article. I'll make you a co nom on the DYK. Thanks so much again!MisawaSakura (talk)12:12, 19 October 2025 (UTC)[reply]
It's just a really complicated thing with non-stoichiometry, a big unit cell, obtained from super-obsure locations. And the aspects that might possibly be worth fretting over (Cl-As bonding?), charge balance, nature of the disorder (Cl vs S in a big cell would take a very skilled crystallographer) would require someone with deep knowledge. So, I worry that developing this article by someone who is "no way a chemist" seems like a really bad idea. Just leave it alone and work on subjects where you do understand the topic. Also I am unsure if it is worth fretting over. --Smokefoot (talk)03:17, 19 October 2025 (UTC)[reply]
I'd tentatively call it notable from the claim that it's the first (well described) chlorosulfosalt of lead and arsenic, but I'm pretty sure that "chlorosulfosalt" itself is a category of maybe ten minerals? It's definitely beyond my ken to delve into the details though. It's got some unusual properties, but they're unusual properties for geologists and crystallographers to poke at a bit more, rather than unusual properties that merit a full page on Wikipedia.
That said, there's alot of mineral redlinks, and I can't help but wonder if some kind of "list of minerals" pages similar to the way akin to e.g.10000#Selected numbers in the range 10001-19999 is done, using the Mindat version of N-S? The current mineral classification pages have no shortage of issues beyond the redlinks - I'm still tilting at windmills about who moved the non-tetrahedral vanadates into the phosphates class, and having a larger restructuring plan would greatly appeal to me.Fishsicles (talk)13:46, 19 October 2025 (UTC)[reply]
I recently came acrossChemical intermediate in the new pages queue after it was split fromReaction intermediate byBernanke's Crossbow. From my brief search online, it seems like there is some overlap in how the termsreaction intermediate andchemical intermediate are used, so it might make more sense to cover them in the same article. I'm not very knowledgeable about chemistry, however, so I'm not in the best position to evaluate the sources. Can any WikiProject editors provide their perspective on whether the articles should be split? Thanks.Zeibgeist (talk)02:25, 19 October 2025 (UTC)[reply]
Theseare two different concepts, though the former is very poorly sourced - this definition of "chemical intermediate" is mostly used in an economics context, while the sole Britannica link describes the chemical sense shown inreaction intermediate. The compounds a market analyst would call "chemical intermediates" are simply products of one reaction and reactants in another; nothing a chemist is going to label a reaction intermediate is going to (meaningfully) last into the next process unit, let alone survive a trip to "other industries".Fishsicles (talk)02:52, 19 October 2025 (UTC)[reply]
Thank you both for your comments. If I've understood you correctlyFishsicles, it seems like "chemical intermediate" is a more niche industry term. The fact that Britannica uses the term as a synonym for "reaction intermediate" seems to reinforce this. I agree withJohnjbarton that the sourcing is very weak, so unless better sources can be found I'm inclined to mergechemical intermediate back intoreaction intermediate. If the article is going to remain separate, better sourcing definitely needs to be added.Zeibgeist (talk)03:23, 19 October 2025 (UTC)[reply]
@Zeibgeist: Why are you even involved in this discussion? Bernanke's Crossbow is a pro in the chemistry editing sphere, and you have no track record. If you need us to explain the difference between a chemical intermediate and a reaction intermediate, then you are out of your depth. I just dont understand editors who take it upon themselves to cross-examine technical articles well outside of their knowledge base. What is the point?--Smokefoot (talk)12:49, 19 October 2025 (UTC)[reply]
I am a new page reviewer, andChemical intermediate is in the new pages queue because it was recently expanded from a redirect. The fact that I'm not an expert in chemistry is precisely the reason I opened this discussion – so that people with expertise can help me make an informed decision. This kind of rude commentary really isn't helpful.Zeibgeist (talk)13:23, 19 October 2025 (UTC)[reply]
@Smokefoot: If I don't make a decision on whether to review the article, that decision will fall on another reviewer (who will also probably not be an expert in chemistry). This is Wikipedia after all; most of us are not subject matter experts. With that in mind, do you have anything productive to contribute to this discussion?Zeibgeist (talk)13:29, 19 October 2025 (UTC)[reply]
I would definitely prefer a split article, even a stub, rather than combining the two pages - the two terms are very different, even if the existing stub doesn't have much to talk about it. I suspect there's plenty of secondary sources about e.g. the size of "chemical intermediate" markets that exist on the economics side of things, I just don't have them to hand - and it's difficult to assess sources on economic data for their reliability, as there's a lot of investorbait slop on the open Internet and I'm not really inclined to pay a "reputable" analyst firm for their reports just to cite a wiki page.Fishsicles (talk)13:54, 19 October 2025 (UTC)[reply]
There are some things that probably would justify, e.g.https://www.britannica.com/science/chemical-intermediate does ("All the substances generated by one step and used for the succeeding step are considered intermediates", clear distinction from a reaction intermediate that cannot be isolated) . Note that pharmaceutical chemists also talk about 'chemical intermediates' as the intermediate compounds between starting point and final API. Some of those chemicals also have no other reason to be made then to be converted further, and often will not even leave the lab where they are made. Sometimes these compounds are just stable enough to observe (spectroscopically) but not stable enough to properly isolate, they react A and B, observe until C is optimally formed and then add D to convert C to E.Dirk BeetstraTC13:33, 19 October 2025 (UTC)[reply]
Thank you all for your comments. It seems like there's rough consensus to have separate pages, so I went ahead and markedChemical intermediate as reviewed. Additional sourcing would definitely be helpful if it can be found.Zeibgeist (talk)14:04, 19 October 2025 (UTC)[reply]
(Edit conflict) Wow, this discussion proceeded far while I slept. No, I don't know any sources immediately. But this is not the first time that a new page reviewer has complained that my split produced a poorly-sourced draft, so I'm going to try to dig up some sources at the local college library.Bernanke's Crossbow (talk)14:06, 19 October 2025 (UTC)[reply]
I think sourcing this split may be more difficult than we think. "Chemical intermediate" has two meanings, not one. Every "reaction intermediate" is also a "chemical intermediate" in many sources. If we had a disambiguation page for "chemical intermediate" thenreaction intermediate would be the main use case according to sources by a wide margin.
(Well, the main thing is that the patroler would "appreciate" our efforts). More seriously, lots of termscould have multiple interpretations, but in this case, there is little room for ambiguity. A "reaction intermediate" is certainly a chemical, but no professional chemist would call such an intermediate a "chemical intermediate" because that would be redundant terminology. At least IMHO.--Smokefoot (talk)16:54, 19 October 2025 (UTC)[reply]
Yeah, I think the main issue here is that search engines see "chemical intermediate" and then immediately leap to "reaction intermediate" as their suggested topic; I'm certainly not familiar with chemical literature using the term explicitly. Searching for e.g. "chemical intermediates market" gives more concrete examples of the "chemical intermediate" usage but I'm not calling any of themWP:RS.Fishsicles (talk)22:27, 19 October 2025 (UTC)[reply]
As I'm gearing up for another bout with Simple English's chemistry pages, I've noticed a particular claim on enwiki:
In chemistry, a derivative is a compound that is derived from a similar compound by a chemical reaction.In the past, derivative also meant a compound that can be imagined to arise from another compound, if one atom or group of atoms is replaced with another atom or group of atoms, but modern chemical language now uses the term structural analog for this meaning, thus eliminating ambiguity.[citation needed]
While this may be a formal definition - and if so, definitely needs a citation - I'd hardly call the use of "derivative" to mean "we traded some parts out on a blackboard" something only seen "in the past". It's certainly not how it'sde facto used on Wikipedia, in categories likeCategory:Nitrobenzene derivatives - if someone can makeflutamide fromnitrobenzene in one reaction I'd love to see their notes.
I'd normally look to IUPAC for prescriptivism in chemical terminology, but IUPAC has no official definition of a derivativeor structural analog. I suspect this definition might be something from the legal world on defining derivatives of a drug (where the "ambiguity" mentioned is much more of a problem)?
If this language has been deprecated, it should have a very robust source backing it - otherwise we should probably represent the practical usage rather than a particular rigid framework. As much as I do love rigid frameworks.Fishsicles (talk)15:42, 27 October 2025 (UTC)[reply]
I agree with the concern about stating which (if either) meaning of "derivative" is obsolescent, even if "analog(ue)"is a more recent and non-ambiguous term for one of them. You're right that there are legal entanglements when a term has an imprecise meaning. Seedoi:10.1039/9781839166631-00153 for some discussion of that aspect.DMacks (talk)16:09, 27 October 2025 (UTC)[reply]
This source itself seems to contradict the idea that one notion is obsolete - per the abstract, "its precise meaning depends on the circumstances", which suggests a lack of a single standardised meaning in 2022, 12 years after the relevant edit,Special:Diff/393553874. I'm not sure there's good reason to leave an uncited assertion from a banned IP user standing.Fishsicles (talk)16:19, 27 October 2025 (UTC)[reply]
I don't know how much banned IP comes into it since that range wasn't blocked until 15 years after this edit, but I agree that unsourced claims as significant as this should be commented out at minimum.. --Reconrabbit19:32, 27 October 2025 (UTC)[reply]
Scribunto recently added support for outputting images. I've been trying to come up with cool uses for that. I made a template (based on CineMol) that takes in a .mol file and outputs a visualization of the molecule:Module:Sandbox/Bawolff/CineMol(edit: nowModule:CineMol). I don't know much about chemistry, would that template be useful for this wikiproject?Bawolff (talk)23:33, 27 October 2025 (UTC)[reply]
This is pretty exciting. It makes it easier to upload views of 3D structures while potentially unifying the presentation style in infoboxes. Sourcing 3D structure coordinates is not too difficult, but it is somewhat obscure. I would rely on theProtein Data Bank for most of the molecules I would be interested in. The first concern that comes to mind is if the image generation slows the loading of the article too much. 3D views are nice, but aren't worth sacrificing reading experience.
As an aside, since you mentioned you're looking for uses for the new functionality, automatic generation of 2D line drawings and reaction schematics would be very desirable for me. The ability to input a fewSMILES strings into a template and output a schematic would save a lot of time for many enzyme reaction articles. The primary reason why I don't add images to articles with inline reaction schemes likeIndole 2,3-dioxygenase is that it takes up too much time to generate images. If I could reduce the work to to copying and pasting SMILES strings and setting wikilinks I would spot fix every one I come across.⇌Synpath02:02, 28 October 2025 (UTC)[reply]
Playing with the module a bit the most salient issue is including the full text of the .sdf file as input. It would be more editor friendly to host a text file on Commons or Wikidata. The easy incorporation of CSS is very nice for us dark mode enjoyers.⇌Synpath 02:21, 28 October 2025 (UTC)Updated image to use new PDB ligand code functionality. The full text coordinates issue is partly solved.⇌Synpath21:12, 29 October 2025 (UTC)[reply]
I think the ideal would be supporting the smile format as wikidata already has a property for that and it also seems to be a much more concise format. At first glance though converting between smile and mol seems a bit complex, but i'll have to read more about it. We can only use .mol text from commons in a template at wikipedia if its in a Data namespace, which is a bit unideal, but i guess workable. I agree that putting the .mol directly on the article feels bad. I guess one could create a template for each compound, and store the .mol in the template but that also feels not great.
As far as darkmode goes, we can make this support dark mode. Is there an official colour scheme for this sort of think in dark mode. For that matter, is there an official light mode colour scheme for this type of diagram that Wikipedia uses? I just copied what cinemol does.
For performance - there are two things. First of all this puts images directly into the article which can theoretically slow down page loads. Generally these images are small enough (~10kb) that i think the effect would be insignificant. Second, the template is more complex than the average template so it may slow down parse time (by ~500ms). However, most of the time a view does not parse the page, so this should not slow things down in the average case. Of course all of this needs more real world testing on real data to be sure.Bawolff (talk)07:45, 28 October 2025 (UTC)[reply]
I was referring toline-bond drawings when converting from SMILES. There are existing open source packages that do this (imperfectly, but it is a hard problem to solve well!). Converting from SMILES to 3D coordinates is something I haven't seen yet, but I also haven't actively looked for that.
Storing the coordinates in a template is a decent enough solution, but it would be more appropriate to store that in a more searchable and accessible space like Wikidata. A wiki I know next to nothing about, so take that with salt.
Re: dark mode, there's no official color scheme, but lightening the shade of grey representing carbon would help dark mode contrast. The outlining in the cartoon look helps a lot with that.
Re: performance, at least personally these images haven't hampered my experience, though part of that may be the caching.⇌Synpath12:51, 28 October 2025 (UTC)[reply]
Personally, I think that those images are not helpful, possibly even deceptive and possibly original research. Maybe I can be convinced otherwise, but my impression is that these images are used as a substitute for conveying information, which is our core job. If one wants to contribute, then describe properties, uses, preps, reactions, safety, etc. I would welcome arguments for the inclusion, much less the proliferation, of such eye candy.--Smokefoot (talk)21:14, 28 October 2025 (UTC)[reply]
Creating an image like this is no more original research than taking a photograph of a landscape and has similar issues with deceptiveness as photos. Projecting the 3D onto 2D necessarily loses information. As for helpfulness, it does vary from person to person, but 3D views of molecules generally help to communicate properties (sterics, at least). They arerelevant.⇌Synpath20:37, 29 October 2025 (UTC)[reply]
@Synpath I think wikidata mostly wants their properties to be about one sentence at the most, so i think mol files are too big. However I managed to import PDB's Chemical Component Dictionary into commons, which is linked together with wikidata viad:Property:P3636. So its now possible for the template to use a pdb ligand id, or wikidata if wikidata has that property filled out.Bawolff (talk)04:45, 29 October 2025 (UTC)[reply]
I see many problems with this type of image rendering and few advantages. The example Bawolff used isbenzylpenicillin and that article already has an accurate 3-D structure based on an X-ray source (click on the image here to see the details). Note that Bawolff's space-filling version a) has used different co-ordinates and b) does not expand to a larger size or give image details when clicked. Molfiles obtained from most sources will not correspond to averifiable 3D set of coordinates but will be as drawn by an unknown author. Incidentally,ChemSpider stores molfiles and Wikipedia articles should always have a ChemSpider ID in the Chembox/Drugbox so in principle the molfile could be obtained from there rather than Wikidata. Our Chemboxes are already set up to provideJmol links so that users can not only see coloured ball-and-stick representations but can manipulate them by rotation to see the "back" of the image, which in my opinion is much more useful than a single static picture.Mike Turnbull (talk)11:48, 29 October 2025 (UTC)[reply]
I don't see there being a problem with auto-generating a static thumbnail-sized visualization based on coordinates deposited in a curated database. Something like thePDB Chemical Component Dictionary would be a reliable and verifiable source of molecular coordinates. Among others, I'm sure. Also, the image could be made to automatically link to Jmol or do so in a caption alongside the sourcing, since Jmol is an external link. The automatic image could be overwritten with a higher-quality, custom-made version when an editor has the time to do so.
That said, I don't think this could be easily integrated into the chembox since there isn't a dedicated image field for a 3d view. We wouldn't know which boxes should already overwrite an automatically generated image.
I added support for clicking to zoom in, as well as the ability to specify an additional caption when zoomed in. The template now also has a link parameter to allow it to link somewhere else if desired. The copyright situation of the ChemSpider database is a bit unclear to me, I don't think we can use it in an automated fashion. Wikidata is used to find the pdb ligand id, which is then used to lookup the mol file in the pdb chemical component dictionary (or more specificly a copy of it that I uploaded to commons). The actual mol files are not coming from Wikidata. Alternatively the template supports directly specifying the pdb id instead of using wikidata.Bawolff (talk)08:17, 30 October 2025 (UTC)[reply]
The question remains: why do this? "I don't see there being a problem" is hardly justification. Showing one of myriad conformations implies some insight that is not available to the contributing editor. I say: (1) contribute core content (properties, uses, preps, reactions, safety), (2) improve our English, or (3) many articles where some improved graphics are needed, lots of badly drawn ChemDraws associated with organic mechanisms.
Page has a lot of potential-application promises and few specifics or sources, and may haveWP:COI issues from researchers cited. I will be looking into it myself but more eyes would be appreciated.Fishsicles (talk)23:34, 30 October 2025 (UTC)[reply]
Clearly AI. Since the original paper only has 29 cites I think it fails notability. Either PROD or AfD.
May I suggest that we discuss a bit further what we consider asWP:SIGCOV for pages such as this. I don't think there is any policy or an essay, and I have seen claims that something is notable ranging from a small number of (peer reviewed) cites to not notable despite having hundreds or even a thousand.
I suggest criteria similar to when we look at academics forWP:PROF#C1. I have seen arguments that "a number of papers with more than 100 cites" starts to be impactful, with > 1K definitely impactful. Hence to me we should use a cutoff of at least 100 to justify an argument ofWP:SIGCOV for a scientific advance. Comments welcome, maybe later this can be opened up to a wider discussion if we have any consensus here.Ldm1954 (talk)13:03, 31 October 2025 (UTC)[reply]
This overview
Defay, E. (2022). Cool it, with a pinch of salt. Science, 378(6626), 1275-1275.
does a good job of putting the ionocaloric effect in perspective. A topic with this level of coverage and fitting in to the overall story of refrigeration and caloric effects is notable.
So does that mean that AI wrote this section or that AI learned about this topic from the Science article + the Wikipedia article written by the same person?Johnjbarton (talk)15:09, 31 October 2025 (UTC)[reply]
The web sites detect AI from various characteristics in the style of writing. They are not infallible, and can be confused by math/citations, but useful. When two give such a high score it means that it was almost certainly written by AI. There is alsoGPTzero (which marked the page as 84% AI generated 16% Mixed 0% Human) as a third source; there are probably other web pages you can use.Ldm1954 (talk)15:23, 31 October 2025 (UTC)[reply]
I don't think there is necessarily a conflict between AI generated and COI; I dislike speculation, but I can imagine a major LLM use case is prompting it to summarise something written for an academic audience into something for a general audience. (Certainly I've seen enough students try and do this...)Fishsicles (talk)15:59, 31 October 2025 (UTC)[reply]
There is also anarticle in Physics Today (March 2023). However, the Defay 2022 article in Science and this one are published soon after the original paper was published in Science in December 2022 so I don't think this article yet meets the SIGCOV requirement usingWP:GNG criteria, i.e. is this articleWP:TOOSOON?Nnev66 (talk)15:56, 31 October 2025 (UTC)[reply]
I'm definitely leaningWP:TOOSOON. As I've done a bit more reading, itdoes appear to be a reasonably novel approach to refrigeration - at the very least, they've built a working prototype, which is more than can be said for many "potential applications" puff pieces, but my subjective judgment of "interesting, potentially promising" doesn't mean much if it stays a curiosity.
On consideration, my preferred solution is currently a page forcaloric refrigeration in general, which can discuss this topic as well as the aforementionedelastocaloric materials and related experimental work. Looking atelectrocaloric effect also sets off some of my puff piece alarms, although that page at least demonstrates to my satisfaction that there's a lot of ongoing work in the field. Themagnetocaloric effect could also be discussed there in brief, as by far the most mature example.Fishsicles (talk)16:09, 31 October 2025 (UTC)[reply]
When does a scientific advance have enough SIGCOV?
I have moved this from the PPP topic above to get more specific input.
I suggest that we discuss a bit further what we consider asWP:SIGCOV for pages such asionocaloric refrigeration. I don't think there is any policy or an essay, and I have seen claims that something is notable ranging from a small number of (peer reviewed) cites to not notable despite having hundreds or even a thousand.
One option is criteria similar to when we look at academics for WP:PROF#C1. I have seen arguments that "a number of papers with more than 100 cites" starts to be impactful, with > 1K definitely impactful. Hence to me we should use a cutoff of at least 100 to justify an argument of WP:SIGCOV for a scientific advance.
An alternative, that is close to the same is criteria used for notability of chemicals, e.g. the ongoingAfD here (although I am not sure what numbers would be used).
At least among physics pages I know a numerical cutoff is helpful but I think other criteria and multiple goals are involved. Is the topic extraordinary (anti-gravity)? Controversial (Hubble's constant)? Supported by work from a single person (Black hole electron)? Then we need secondary sources and not just press releases or news reports. Could the topic fit with other similar topics? The bar for a paragraph is much lower than the bar for an article.Johnjbarton (talk)14:04, 31 October 2025 (UTC)[reply]
Thank you Ldm1954 for starting this conversation as I was curious myself about what sources count as SIGCOV for an early stage scientific discovery. I see there's a proposal that a certain number of citations could be a loose metric, which seems reasonable. However, I don't think individual journal articles noting a given discovery/method in the context of other research should count toward notability. Would articles written in journals like Science for a generalist scientist audience count towards this (e.g. Defay 2022 noted by Johnjbarton for the "Ionocaloric refrigeration" section above)? I would think so but would like to see what others think.Nnev66 (talk)15:50, 31 October 2025 (UTC)[reply]
My personal sense is that for scientific advances, one has to note their coverage by independent researchers. Whether this coverage is "in the context of other research" doesn't really make a difference, whether this coverage is substantial probably should. Regardingionocaloric refrigeration there seems to bea little independent coverage, but whether it's substantial I didn't check.Jo-Jo Eumerus (talk)07:46, 1 November 2025 (UTC)[reply]
...matters for discussion raised with today's date, at the Talk page ofPotassium peroxymonosulfate, regarding split of the article into two—a short one for the pure title compound, and a longer one for the commercial reagents historically known, e.g., as Oxone® (which contains the pure compound in various formulations). We have done work today to make clear which content elements of some content sections apply to which agent—the pure or the commercial "triple salt"—but as other registered editors have noted (including some contributing here), the article currently continues to confound the pure agent and the triple salt formulation. So, if effort is going to be made to split the article (i) it is easier now than it was earlier, and (ii) it will make further, future edits all the easier to have the article split. (After today, we will reserve editing time for other articles, until this is addressed, so that we better use our limited time in future.) Signed, a former registered editor of many years.73.45.21.125 (talk)17:52, 3 November 2025 (UTC)[reply]
I've made the{{Chemical reaction}} template meant to ease the creation of high-level chemical reaction schemes. Rather than spending 10-30 minutes creating a high-quality svg and uploading it to Commons you can now opt to create a medium-quality version from images of single compounds already on Commons in 1-2 minutes. Great for improving less-viewed corners of the project that use only text to express reaction schemes.
I'd appreciate if you could take a look at the template and give some feedback on readability, accessibility, usability and practicality. I've constructed some examples atCitrate synthase,Tryptophan 2,3-dioxygenase,Strictosidine synthase andGlutamate synthase (NADPH) and you can see them all at the/testcases page. My bias is enzyme chemistry as you can see, but the scheme should work for any reaction(Note that the arrow does not accommodate verbose reaction conditions well, at the moment). I would like to improve the template to reliably make schemes that editors can use easily+quickly, while rendering well on all skins and small screens. I have a few ideas on where to go next, but I think the template is in a good enough spot to ask for opinions here before making more changes.⇌Synpath15:55, 8 November 2025 (UTC)[reply]
Wow, that's impressive, though I'm not sure that your estimate of 1-2 minutes is realistic given the steep learning curve for use of the template! Nevertheless I'll certainly give it a try. I liked the fact that I could easily tweak the size of each individual component using the |upright= parameter, which allowed me to make both citrulline and aspartic acid have the same bond lengths (set aspartic to |upright=0.6), which was otherwise what would be a minor irritation about the example on your template page. I also liked the fact that you are supporting other-language versions of Wikipedia by separating the captions from the drawings. It has been a problem with chemistry reaction schemes that some editors are tempted to put language-specific text into their .svg so that they internationalise less well. Well done indeed.Mike Turnbull (talk)16:40, 8 November 2025 (UTC)[reply]
I noticed that you had used an image of unspecified stereochemistry for the product atGlutamate synthase (NADPH) so Itweaked it to use the correct L-Glutamate product by swapping the image file. In doing that I noticed a very odd effect. You had an overall width parameter of 50% but if I kept that value the product's image came out looking absolutely tiny in Preview and adding an |upright parameter had no effect. However, when I enlarged the whole width parameter to 60% everything looked OK. Is that something to think about?Mike Turnbull (talk)17:18, 8 November 2025 (UTC)[reply]
Thanks for taking a look! Part of the reason I'm looking for feedback is that I'm getting blind to how easy/intuitive this thing is to use, and how quickly I can get others up to speed on how the template group is supposed to operate.
As forGlutamate synthase (NADPH) I gave ita pass of my own and the issue was that the width of the image was being suppressed by the lack of a text caption. An un-intuitive effect I will need to fix. Perhaps with a minimum width CSS rule or by making the caption mandatory.
The width parameter is a bit of a pain point for me. There are some values that improve the look of a scheme, but it needs to be balanced against how the page is rendered at different screen sizes. I set it to 50% as a conservative width that accommodates the fixed width of the Infobox. I can partially fix this by forcing the scheme to scroll when the sizes don't match up in borderline scenarios, but using scroll bars like this aren't recommended for mobile I think.⇌Synpath19:35, 8 November 2025 (UTC)[reply]
Glutamate synthase (NADPH) was an old semi-automated creation from 2007 when rather clunky wording was used for the reaction description. I've noticed that in many other examples. I've tweaked the offending sentences and also removed the now-duplicated Wikilinks which your reaction scheme provides. I think I'll probably have many more questions and comments about the template as I try it out but I'll put them on your talk page rather than clutter up this thread. BTW, I now realise that some editors might like to use your template to create reaction diagrams as .png, which they can do by taking a screen-grab of the rendered output using a snipping tool. Good for people who haven't got access to a chemical drawing program or don't want to learn how to use one.Mike Turnbull (talk)14:26, 9 November 2025 (UTC)[reply]
From an editor who is not good at this technology, the template is really impressive. The products are however inferior to ChemDraw in the hands of an experienced chemist. Another thing is the inclusion of links and words within equations.... isn't that discouraged? Small rant: just use ACS or RSC settings. And use professional designations like RCHO for aldehydes, do not omit the H. --Smokefoot (talk)14:48, 9 November 2025 (UTC)[reply]
Most of these shortcomings are because the template relies on existing drawings, which are often not quite up to our standards, especially regarding ACS settings. The advice not to include links and words in equations is, I think, because when placed on Commons this makes a drawing less useful in other-language edition of WP. This template doesn't have that problem as it would allow for foreign captions/links when used elsewhere because the individual images are themselves "clean". I've just used the template atAcetylserotonin O-methyltransferase#Reactions catalyzed, where the earlier version had lost all its diagrams: I think you'll agree this is much better thanwhat had been there!Mike Turnbull (talk)16:29, 9 November 2025 (UTC)[reply]
The arrow is too thin, small and narrow to be comprehensibly seen, at least from a phone user's perspective; you should make the arrow longer to give a decent space between the|fwd_in= and|fwd_out= contents/outputs, too. The generated diagram should also automatically become scrollable horizontally, the test case in§ test1 cuts out to the sides. I think you could take inspiration from the§ All recognised symbols and codes function of{{chem2}} so that common key terms (e.g.: NADH, coenzyme abbreviations like CoA, the 3 letter abbreviations for amino acids, etc.) are automatically wikiliked to their respective articles (rather than having to spell out the entire wikitext for their link, with an additional pipe-trick for the abbreviated name), or just take inspiration from{{chem2}} in general. I mean, you're pretty much already there, except each substrates/products – and even the plus signs – were in separate templates for some reason. Test cases in §§ test1,test3, andtest4 could all have easily been 3 templates long (discounting{{chemical reaction}} itself and other unrelated templates) if each substrates and products are defined separately in their respective (unnamed) parameters in{{chemical reaction/compound}}, and the plus signs automatically added between each compounds; this might mean {{#invoke:Separated entries|function}} would be needed in{{chemical reaction/compound}}, I guess.Heck! You could probably make this into just one template – with one parameter for the substrate compounds, 3/4/however-many parameters for the arrow, and one parameter for the product compounds. My idea for this is thusly:
{{chemical reaction}} will have the following parameters:|subs=,[a]|prod=,[b]|ucap=,[c]|lcap=,[d]|dir=,[e]|fwd_in=,[f]|fwd_out=,[g]|rev_in=,[h]|rev_out=,[i] and other possible parameters for micromanagement
|subs= and|prod= will act like the{{chem2}} input – except with compound names instead of chemical formulas – where "Phe" will link toPhenylalanine for example
You can add single curly brackets around certain names if necessary (example:{UDP}–Glc so it's interpreted asUDP–Glc instead ofUDP–Glc)
With the curly brackets function, you can also disable image display (putative example:!i_{ATP} will display text only, skipping the diagram)
Going by the2006 ACS style guide (p.272,379) and what I could see from the modern one locked behind ACS' paywall, there is no explicit guidance on arrows outside of that it should be legible. At the moment the arrows are not legible when crowded, and limiting them to become no smaller than 55px wide helps with that (see testcases). To my eye this is a reasonable trade off for these pictures of arrows when downscaling. Otherwise I'll need to make some new pictures of arrows or figure out a CSS trick to size them.
The font size issue is a tradeoff between readability and being able to render the text without overlap. For the smallest screen sizes a font size of 85% often overflows its HTML element making captions unreadable. Part of my justification here in shrinking the text below what MOS:SMALL recommends, is that the heteroatoms in the chemical structures themselves often violate MOS:SMALL at small screen sizes. This is especially true for schemes with several compound images. The larger font can also look inconsistent next to the text size in the images. Of course, that could indicate that this template shouldn't be used in such a scenario. Though, even proper SVG schemes suffer from this, but they have the option to be zoomed in on when clicked.⇌Synpath 23:44, 12 November 2025 (UTC)a hefty edit for clarity⇌Synpath01:39, 13 November 2025 (UTC)[reply]
Thanks for really getting into this! I've made a few changes to how the arrow,|fwd_in= and|fwd_out= are rendered and they shouldn't ever get too small while maintaining space between the labels. If the arrow does still get too small I'll have to spend some time on figuring out a CSS trick or making new arrows.
Also, the testcases should now show that the reaction schemes never overflow the boundaries of the page or surrounding column of text. I don't think the sizing options are ideal quite yet, but they at least shouldn't need scrolling. I'd like to avoid scrolling, but if the sizing issues become too pathological/unacceptable for small screens a well-signposted scroll bar/overflow indicator will be the best mobile remedy.
As for the overall design of the template—I wanted to keep the template transparent, simple and flexible. To me that meant foregrounding as much familiar wikitext as possible like[[File:...]] and normal wikilinking. That lets an editor maintain a lot of control over display. For example, theupright parameter still works to scale the images with no extra template parameter. Part of that approach led me to decide that the scheme content should be inputted as a sequence of templates with as few assumptions on order and use as possible. It's also why I would be reluctant to add more structure to the base template through extra parameters. However, that flexibility means the custom text replacement/linking like{{chem2}} can be integrated as a new subtemplate and an editor can opt to use that as their preferred text input choice.
Unfortunately, automating everything to pure text input does not work when using{{wikidata}} (the underlying template of{{chemrxn/cpd}}) since image retrieval from Wikidata often does not work using a name alone.
Regarding the "+" joiners—the current implementation does not allow them to be added automatically (that I know of). That would need either a verbose parameter set (e.g.|subst1=|subst2=|subst3=|prod1=|prod2=|prod3=) or Lua code.
Lastly, integrating the arrow in the base template is something I would like to avoid for only one reason: the current arrow has a really narrow use case. It's only useful for these generic biochemistry-leaning schemes and would absolutely fail to render even a simple chemistry reaction scheme like that atSwern oxidation. One of my next steps is to make a new arrow template that can take numbered lists above and below the arrow (Template:Chemical reaction/orgarrow?Please pick a better name for me. Perhaps the current arrow should be called /biochemarrow).⇌Synpath01:30, 13 November 2025 (UTC)[reply]
Oh, I meant the arrow itself is too small (the arrowhead blends in/becomes invisible), but it's good now; it's fixed.
Also, the testcases should now show that the reaction schemes never overflow the boundaries ...
It seems that the tradeoff here, weirdly enough, is that the captions now cut off.
... custom text replacement/linking like{{chem2}} can be integrated as a new subtemplate and an editor can opt to use that ...
That would be great, actually! I forgot to mention that the + signs are still manually added for the third (just one template) example, just like in{{chem2}}. The reason I even came up with the automatic "+" joiner idea was because I noticed that{{chemical reaction/text}} was even needed to add the + signs.
It seems that the tradeoff here, weirdly enough, is that the captions now cut off. This has been a hard thing for me to avoid. I have found no universal, reliable solution in pure CSS, other than scaling text down to too small sizes or resorting to scrolling. The documentation probably should recommend chemicals with long names be abbreviated (with abbreviations linked and defined in prose) and to check how the template is displayed at other screen sizes by resizing the browser window. That may be adequate.
Also, I tend to agree that {{chemrxn/arw|+}} is a bit verbose just to add a plus sign, but I'm not convinced it negatively impacts usability too much. Maybe, with time, that will become clearer.⇌Synpath01:34, 14 November 2025 (UTC)[reply]
Scrolling overflow might be something thatI just don't like and I'll get over it when I see enough examples with sketchy presentation. There are issues with hiding content behind a scroll, but hiding text with no way of recovering it aside from the HTML is worse.⇌Synpath04:37, 14 November 2025 (UTC)[reply]
Can someone please remind me/us about what problem this template solves?I mean, the technical aspects of the coding impresses me, but why do we need a "reaction template"? What is difficult or broken about our usual rather flexible approaches to introduce drawings? Is this an effort to standardize format or to distinguish chemical reactions" from other chem graphics? --Smokefoot (talk)13:16, 13 November 2025 (UTC)[reply]
@Smokefoot Having now taken a look at all the articles in{{One carbon transferases}}, I have been able to rather quickly add enzyme reaction diagrams to those which lacked them or had poor representations. In its latest version, the template can access individual chemical structures from Wikidata, so when I notice that some component is missing I can draw just that chemical, upload it to Commons and make the link in Wikidata. Now anyone can use my drawing in any other chemical reaction. This will give a flexible and consistent approach to articles on enzyme mechanisms. It may in the long term also be useful for simple A -> B chemical reactions. SeeArginine:glycine amidinotransferase andN-acetylornithine carbamoyltransferase for examples and compare them to what we previously had.Mike Turnbull (talk)14:04, 13 November 2025 (UTC)[reply]
Just ask me via my talk page. There are very many more compounds in Wikidata than have articles in en:WP. Wikidata also has images (from Commons, of course) for many things that don't exist here as articles, some of which are not ACS-standard. Creating .svg files for known compounds that exist in ChemSpider or PubChem is something that can be done easily once you know the tricks. The main technical issue is that chemical drawing programs can't import and convert .svg back into the proprietary vector graphics files thatMDL Information Systems invented originally (.skc files), which makes the drawing of whole new reaction schemes in a chemical drawing program much more tedious than using the new template, whose modularity is its strongest feature.Mike Turnbull (talk)15:10, 13 November 2025 (UTC)[reply]
Amino acid has been nominated for a good article reassessment. If you are interested in the discussion, please participate by adding your comments to thereassessment page. If concerns are not addressed during the review period, the good article status may be removed from the article.Z1720 (talk)15:55, 22 November 2025 (UTC)[reply]
The redirectMolecular form has been listed atredirects for discussion to determine whether its use and function meets theredirect guidelines. Readers of this page are welcome to comment on this redirect atWikipedia:Redirects for discussion/Log/2025 September 10 § Molecular form until a consensus is reached. --MYCETEAE 🍄🟫—talk19:16, 10 September 2025 (UTC)[reply]
You are invited to join the discussion atTalk:Lyonium ion § Proposed merge of Lyate ion with Lyonium ion, which is within the scope of this WikiProject.DMacks (talk)05:18, 26 October 2025 (UTC)[reply]
