Wikipedia talk:WikiProject Elements/Archive 57

On the periodic table article more generally (2)
It is also listed in for WP:ELEMENTS. See also (recents changes). -DePiep (talk) 20:22, 10 December 2020 (UTC)

Group 12, on the basis of what Andrew D. said above about our audience
There may be some grounds for returning group 12 to the transition metals in the colouring like we had until 2017. In fact, here are all ten of them that I see, going from simplest to most complicated.


 * 1) We are writing for beginning students, and they don't really need the detail. Anecdotal evidence: in my first year in school chemistry, Zn, Cd, and Hg were called transition metals as they are part of the d block. Only in later years were they excluded on the grounds that they don't really have the transition metal properties. More on that point at #8.
 * 2) Group 12 is included in transition metals by Britannica, ptable.com, LANL, ACS, and RSC (the five sources Sandbh mentioned when justifying categories' inclusion above). I think this supports point #1 that it is generally felt that beginners don't need the detail. Rather strongly in my opinion because while they disagree on some details, this is one thing that they are all in unison about, and we can't say some of these aren't authoritative.
 * 3) Regarding the big three inorganic textbooks that Sandbh mentions often: Cotton & Wilkinson exclude the Zn group from transition metals, but Greenwood & Earnshaw (p. 905) as well as Holleman & Wiberg include it. So that's 2/3.
 * 4) According to W. B. Jensen, writing in 2003, "virtually every recent general chemistry text" (p. 952) he examined included the Zn group in transition metals. While he did mention that many inorganic texts that do this later contradict themselves, he also mentioned that most general chemistry texts simply did not even cover the chemistries of these elements in enough details for any real problems to arise. (Regarding whether or not there really are big problems coming from it, see #8, #9, and #10 if you are up for more chemistry as described in RS.) The 50-50 divide he referred to was about "advanced monographs on coordination chemistry and organometallic chemistry". (Note that Jensen opposes the inclusion of the Zn group in the transition metals and considers it wrong, but these are the results he admitted he got while surveying the literature. Regarding his argument, see #10.) I think this substantiates my point that beginners are usually taught that the Zn group is a transition group, and if it is excluded (which is not universal), then that is only done at a later and more advanced stage. So the 50-50 argument doesn't seem to hold that well in hindsight given the level we should be writing for. Besides, if something is universal at the basic level, and even at the advanced level half of all authors agree, then should we really be taking the view of the dissenters?
 * 5) Since this is basic, we should not just look at textbooks, but also at even more basic things like periodic table posters. It seems to me that in that sort of sphere, including the Zn group into the transition metals is absolutely the dominating option judging by Google Images results.
 * 6) We had it this way from 2002 to 2017 and no one complained that much.
 * 7) That's the definition IUPAC gives first in the Red Book: "For example, the elements of groups 3–12 are the d-block elements. These elements are also commonly referred to as the transition elements, though the elements of group 12 are not always included". And that seems to be their primary one, since on p. 228 onwards when they talk about naming organometallic compounds of main group elements, group 12 is not included as a main group (implying it has been shunted off to the transition groups).


 * 1) The lack of significant transition-like chemistry of group 12 is one of the reasons usually given to exclude it. But it is not really decisive. Silver lacks it too for the most part, yet effectively everybody today calls it a transition metal. Besides, as noted at Group_12_element there are some significant similarities between group 12 and group 11 in chemistry (e.g. Cu2+ vs Zn2+); and as noted at Post-transition_metal, the group 11 elements themselves behave more like the group 12 and later elements when in the +1 oxidation state, and already show the low melting points and high electronegativity characteristic of the group 12 and later elements. This all suggests that the difference between groups 11 and 12 is not an absolute and sharp division and that it is not something that is worth forcing on the beginning readers right from the start. Sometimes the group 12 elements can be made to behave more transition-like, and sometimes the group 11 elements can be made to behave less transition-like (in the case of silver one hardly needs to try very hard). It seems to me that the details of this are just not worth going into for the beginners that should be most of our audience, which is consistent with how general chemistry textbooks and posters generally include the Zn group (#4 and #5), and it is generally the higher-level books that sometimes (not even always!) start excluding them.
 * 2) The exclusion of group 12 seems to also come from the lack of ionisation of d electrons for chemistry and the definition that says a transition metal must either have or be ionisable to an incomplete d subshell (IUPAC Gold Book). But (0) taken literally this means that thorium [Rn]6d27s2 and uranium [Rn]5f36d17s2 are transition metals, which nobody actually believes; (1) this is controversial for Hg (does HgF4 exist, and if so does it count because it is present only at weird conditions?); (2) as noted in the periodic table article lack of ionisation does not equal lack of contribution into the bonding; and (3) this definition may get seriously problematic not far from where we are at with predictions about Cn ionising its 6d before its 7s. There have even been predictions for Nh and Fl 6d usage (in the literature, although unlikely, so they have to be thought about), and then all of a sudden we may be in the situation where an element has been characterised but we cannot colour it in due to reasons that don't seem to make that much difference for the actual chemistry, judging by how the reverse order of 6d and 7s compared to normal seems to make no difference for Sg earlier in the transactinides. (I understand that concerns about elements 112, 113, and 114 seem esoteric, but already preliminary chemistry studies have been carried out experimentally on all three and there seems to be no reason why we might not get results in the near future. Indeed we kind of already did, if you remember how elements 112 and 114 at some point got coloured and then uncoloured.) All these concerns are a bit awkward and at the same time completely unproductive and confusing for beginners; especially the concerns here about the superheavies that are opened by the exclusion seem to me a bit like rearranging the deck chairs on the Titanic considering how short-lived they actually are.
 * 3) It seems that this is partly correlated with Jensen's article (linked above) arguing against considering the Zn group as transition metals. The only problem is that his argument is against the conclusions of reliable sources. On p. 956 Jensen states that there is no convincing evidence for the involvement of the d electrons of the Zn group atoms in any sort of bonding interactions, and gives theoretical studies of ZnS as justification. Leaving aside the point I made in #9 that this is surely too high a level of argument to use for something we're trying to present at a basic level, it is a bit odd that he only considers ZnS because already in 1977 such a 3d bonding interaction had been investigated for some simple Zn compounds by Tossell, in which indeed none was found for ZnS, but a significant one was found for ZnF2. Given that alleged lack of bonding involvement of d electrons in the Zn group seems to be Jensen's only argument against considering them as transition metals, and it seems from other sources to be based on a faulty premise, this must be regarded as problematic.

The impression I get from #8 and #9 is that excluding group 12 is based on higher-level considerations from those our likely audience should concern themselves with. It opens up a can of hard-to-answer yet unproductive questions that are really not on the radar of that target audience (chemistry of elements that don't even last an hour, MO theory, weirdo compounds of mercury at cryogenic temperatures), and also has part of its basis on a definition that is not even taken at its word because it produces strange results like Th and U as transition metals. Moreover, the arguments often seem fairly qualitative and debatable (#8), since already group 11 (and especially Ag) are starting to be quite weak in showing some of the characteristic transition metal properties too but are rarely excluded anymore, and the clearest argument produced for the exclusion of the Zn group (#10) is based on what is apparently a false premise (as it's refuted elsewhere in the literature). On pragmatic grounds, group 12 in transition metals appears to be simpler and avoids such murky waters. This is substantiated by the fact that the exclusion of the Zn group is generally a phenomenon encountered at or close to the university level, is not even universal there (arguments #3, #4), and that basic things for beginners generally include it (#1, #2, #5). The fact that this was uncontroversial for an amazing fifteen years (#6) on the project, and that IUPAC seems to include the Zn group in the Red Book after paying a little bit of lip service to those who exclude it (#7), suggests to me that we should have left well enough alone. Which is a sentiment expressed by our founder way back in 2008 when the group 12 thing was first discussed; since it seems to me that the situation had not changed much in the literature since then, it strikes me that that should perhaps have been the end of it.

And remember, this is just for the beginning periodic table article. We do not need to follow it all the time if we are writing about a topic where the other definition makes things easier: after all, context matters for categorisation. Probably, however, that will be an advanced topic if the literature distribution found by Jensen is anything to go by. I submit that in the beginners' context, including the Zn group as a transition group seems to be the standard way to go. By excluding it three years ago we have opened up a complicated university-level can of worms into an article on a topic often taught to schoolchildren, when the issue at university level is not even fully agreed, and I think that we may have thus done our readers a disservice by introducing that instead of an absolutely standard simplification.

TL;DR summary. Excluding the group 12 elements from transition metals is based on advanced information that I do not think the reader actually needs to know, that this advanced information is not even actually agreed on the advanced sources, and that in general basic-level sources for beginners stick to calling it transition metals as we did from 2002–17. Considering that an average general chemistry book will discuss the different parts of the PT including transition vs main-group elements, I feel that we should follow what even Jensen admits to be a universal in that context.

As noted: Keep in mind that any model is inevitably a simplification of reality that spares the reader overwhelming complexity. A central question for each article using the PT is "Who is the audience for this article?" What we often miss at Wikipedia is that for every editor there are hundreds of silent readers. We need to think about what will serve the reader best. Sometimes less complexity is more valuable. We can have daughter articles that go into greater depth. We can provide links to deeper treatments of the topic. I submit that the place to go into this at greater depth is at the main articles on transition metal, post-transition metal, and group 12 element, and not periodic table.

Double sharp (talk) 13:22, 15 November 2020 (UTC)


 * Double sharp (talk) 13:26, 15 November 2020 (UTC)

Side discussion about periodic table placement
''This sub-section was originally part of >  >, but seems to no longer be directly related to discussion & reasoning related to that poll, much less discussing the application of WP:NOTAFORUM, so I have moved it into a separate "==" level thread. If anyone feels strongly that it should be a "===" level subsection, please make the change. YBG (talk) 04:25, 18 December 2020 (UTC)'' Specifically, this thread was begun by this response to this post.YBG (talk) 08:52, 19 December 2020 (UTC)


 * , yes, the article on H over C by Cronyn (deceased) was published in well regarded, peer-reviewed Journal of Chemical Education. It has 41 citations, including by Schwarz; Scerri; Philip Stewart; Jensen; Sacks (deceased); Laing, and Rayner-Canham. I referred to H over C in in my peer-reviewed article on organising the metals and nonmetals (6,500+ downloads)–none of the referees, nor the editor, batted an eyelid. I'm co-authoring another article (2nd proof forthcoming) with Schwarz on the periodic table in which we mention, among other things, H over groups 1, 13, 14, and 17. Even Imyanitov referred to H over group 14: "Besides block boundaries, the position of certain elements in the Periodic Tables is also a subject of debate. 4 positions are proposed for H, i.e. above Li, B, C or F" (p. 170). . In the same way, I don't count He over Be as fringe. Sandbh (talk) 05:57, 16 December 2020 (UTC)
 * If H over C (or He over Be, or anything else for that matter) can be found in 2RY or 3RY sources, there is no need to go hunting for 1RY sources. If they're not referenced in any meaningful way in 2RY or 3RY sources, then it would take an enormous number of 1RY sources to overcome the very reasonable presumption of FRINGE. YBG (talk) 07:19, 16 December 2020 (UTC)
 * Double sharp (talk) 07:57, 16 December 2020 (UTC)


 * , 1RY sources are useful, among other things, to establish provenance, even where there are 2RY sources. There is no prohibition on the use 1RY, in the absence of 2RY or 3RY. WP policy is that primary sources that have been reputably published may be used in Wikipedia, with care. A generic example would be: "It was recently reported in Nature that…". WP:FRINGE is essentially about not giving undue weight. Even here, the advice is that the policy “is best treated with common sense, and occasional exceptions may apply.” Questions about the location of hydrogen have a long standing history, and are still raised today. I see nothing FRINGE about that. Equally, Sc over Al has a long history (1893 to 1983+), and is still mentioned today as per the examples I gave. It doesn't merit an article of its own. It certainly warrants a mention in the historical part of our periodic table article, including a reference to its dramatic decline in popularity with the rise in modern electronic theory. Sandbh (talk) 06:07, 17 December 2020 (UTC)
 * When questions about H are raised today, in reliable general chemistry textbooks, it is always only about H-Li vs H-F. The usual way in which it is discussed is that it's pointed out that H has similarities to both Li and to F, but does not fit either group totally. That's what Greenwood & Earnshaw does, that's what Scerri in The Periodic Table: Its Story and Its Significance does, that's what the Russian pop-science book 107 Stories About Chemistry does. Then, among most modern sources, the conclusion is either that it stays in the now traditional place in group 1 over Li, or it is left to float outside both groups as not entirely fitting either. In the past, duplicating H over Li and F was more common, but now it is rarer. H over F is also pretty rare, but all right, Clayden et al. show it in their famous organic chemistry textbook, so I grant it a pass. There's not a single 3ARY source about general inorganic chemistry around that treats H-B or H-C on a par with either of those. Those are what I would consider FRINGE. Incidentally, I also think the same of He-Be, which is why I've removed it from the periodic table article despite the fact that I personally agree with Wojciech Grochala and think it's actually correct. Why? Because not a single 3ARY textbook has actually been convinced by Grochala or anybody else about this yet. I think it sucks, but as long as that's the situation, I think keeping it out of the main periodic table article is necessary to keep WP as more like an encyclopaedia than like a journal where primary research is published.
 * You may have noticed that I have not objected to discussing B-Al-Sc as a matter of periodic table history. Because it was, indeed, a part of periodic table history. And there is, indeed, a reliable 3ARY source mentioning its role there. You already showed it: Mellor 1943 reimpression of 1939. (Although even that already says that the difficulty of IIIA vs IIIB was a past one and that B-Al-Ga-In-Tl was then already "generally agreed". I am pretty sure every chemist who was professionally active before 1939 is either deceased or optimistically retired by now...) Curiously, the same is true for Be-Mg-Zn (which clung on a little bit longer, see Sidgwick 1950 who still feels the need to defend why he doesn't adopt it), but I don't see you calling for it to be described. I also can't help but notice that your analysis is in favour of one but not the other, which makes me agree with EdChem that only the literature supporting something you like is being scoured. But anyway: both were a part of periodic table history, as parts of the hiccups that arose from expanding the 8 column table into the 18 column table. Indeed, those hiccups are still around wrt the f elements, whence the Sc-Y-La vs Sc-Y-Lu thing. But: unlike the group 3 dispute: the situation has already been resolved for groups II and III. Among 99.999% of practicing chemists today: Be-Mg-Zn and B-Al-Sc are as dead as a doornail. Discuss them in history? Sure! There are secondary sources for that. Show dead-as-a-doornail already-refuted B-Al-Sc outside history, in a picture purporting to give general isodiagonality trends, that are not generally accepted by chemists outside the usual cases of Li-Mg, Be-Al, B-Si? No, I very much oppose that. A totally OR idea of chemical B-Al-Sc vs electronic B-Al-Ga? That too.
 * I don't have anything against you posting forgotten chemistry here, if you can find sources for it from history. I don't think anyone disputes that something like B-Al-Sc deserves coverage as a historical matter. I am less enthused by your OR here to support their rehabilitation – questions on whether it is based on a modern understanding of chemistry aside, I would prefer it if it was kept to userspace – but, since these issues tend to result in flame wars even in the literature sometimes, I also think it's not intrinsically bad to know where everyone personally stands on each issue so that we may all be scrutinised to make it very clear that we're not being biased one way or the other. I, however, am not convinced if the rest of us agree with your stand on how sources ought to be used on WP. YBG is making an excellent effort to clarify where each of us stands. Then, if it becomes clear that the divide I suspect exists really does, between your view and the others here, then we need to think about how to resolve the conundrum that our vision of how the periodic table article should look is probably going to be very, very different from yours. Double sharp (talk) 06:46, 17 December 2020 (UTC)

, interesting post, thanks.

Hydrogen: General chemistry textbooks, along with the well-documented phenomenon of textbook errors,, , are not the sole source upon which WP is built. Nor are 3RY sources. Specialised sources are more reliable in their areas of focus.

References dealing with H over B, H over C, and He over Be appear in reputable peer reviewed sources. There is nothing FRINGE about that. I would at least add a footnote about the first two. He over Be deserves a mention in the main body of the periodic table article given how many sources refer to this, including in the LSPT appearing in our PT article:

Groups 2 and 12: This story is missing from our article, and merits inclusion. There is some interesting history and chemistry here.

That my analysis has focused on group 3 and 13 was prompted by the IUPAC project. When Schwarz learnt his craft in the 1950s he told me that he was taught B-Al-Sc-Y-La.

I posted about group 2 and 12 early on:


 * I mapped the trend-lines going down these two options using the same eight properties I used for group 3 and 13.


 * The average goodness-of-fit values were:


 * Be-Mg-Ca-Sr-Ba = 0.84
 * Be-Mg-Zn-Cd-Hg = 0.66


 * This is better than the figures for group 3 and group 13, which I did not find surprising. It is consistent with 16 of the 17 B-Al-Sc-Y-La-Ac tables. Only Hubbard shows Be-Mg-Zn-Cd-Hg. It is consistent with C & W (1999): "The Group 12 elements differ markedly from those in Group 2 in nearly all aspects except having II as their only important oxidation state." (p. 599).

Isodiagonality: There is Mingos (1998) who counts diagonal relationships as one of the three patterns that characterise the periodic table, along with vertical and horizontal trends: Essential Trends in Inorganic Chemistry. Oxford University Press, Oxford.

Or Rayner-Canham (2011): Diagonal relationships are, "…a general attribute of the properties of the chemical elements. For example, the metal-nonmetal divide forms an almost diagonal demarcation (Edwards and Sienko 1983 pp. 691–692). Similarly, the elements often considered to be semimetals fall on a roughly diagonal border between the metals and nonmetals."

Note Mingos and Rayner-Canham are general chemistry textbooks.

Several authors have written on diagonal relationships extending beyond periods 2 and 3, from as far back as 1928.

The PT image "purporting" to show isodiagonal relationships comes from a peer-reviewed reputable journal article, as did its predecessor. Images are only words in pictorial form; there are supporting RS for all those relationships.

Chemical B-Al-Sc vs electronic B-Al-Ga: Is hardly OR, let alone "totally" OR. It was called out by Mellor (1939). The two relationships are still seen in the n + 10 relationship, which goes back to DIM.

Sources: My stand on how they can be used is based on WP policy.

Flame wars: I note prefers to aim first then shoot. I sometimes fire first then aim. It's odd that this time I've engaged in "aim first" by way of discussion here and that this approach has gathered as much opprobrium as firing first. Sandbh (talk) 02:33, 18 December 2020 (UTC)

Why is the periodic table article now headed with a B-Al-Sc table?
Permanent link to revision of the image after 's recent update (. Edit summary mentions only change of "Metalloids" to "Metalloid". While that changed indeed, he did not mention this rather big change that also happened. (Note, this means it has been up for almost 17 hours already.) Double sharp (talk) 15:22, 18 December 2020 (UTC)


 * Could we please have an explanation for this?
 * For your views, please. Double sharp (talk) 15:24, 18 December 2020 (UTC)
 * Also. Double sharp (talk) 15:27, 18 December 2020 (UTC)
 * Revert this immediately. We might as well then place thorium in the d-block, hydrogen in the middle of nowhere, and recolor a whole bunch of elements. There's no consensus for this change, and per some comments I wrote above, this is extremely unsound from a physical and conventional point of view. ComplexRational (talk) 15:44, 18 December 2020 (UTC)
 * I definitely wanted to do this, but given my previous history with, I felt I had to wait for someone else to confirm. Reverted. Double sharp (talk) 15:54, 18 December 2020 (UTC)
 * Looks like the imposed version is this one:, 22:30, 17 December 2020, change summary "Metalloids --> Metalloid".
 * I note that removing the -s from metalloids is inconsistent. And: it happened after a relateds requist by me (misunderstood? ):.
 * The B-Al-Sc change was not put up for discussion, not within 17h and not after being asked to do so, a request (above), did not contain a proper commons changesummary, did not answer a partial request correctly. I have found no explanation for this editing behaviour. A sort of abuse of "Bold", and so "Revert" is in place. -DePiep (talk) 16:52, 18 December 2020 (UTC)

This is the 2nd or 3rd time I have heard about B-Al-Sc being at Periodic table. but each time it seems to have been removed by the time I look at the article. YBG (talk) 21:32, 18 December 2020 (UTC)


 * I haven’t gone mad. I made a mistake. I removed the s from metalloids in response to a request from on my talk page:


 * ”I'd suggest you make this edit to the original & upload. In the categories-legend, in the below-row add the word "metalloid" (singular). It would complete that row, instead of making metalloid(s) confusingly and inexplicitly standing out.”


 * I inadvertently did this to the wrong file, that’s all. Glad to see it was rightly reverted.


 * DePiep’s suggestion has merit. While there are multiple categories of metals and nonmetals there is only one category of metalloid. Sandbh (talk) 22:10, 18 December 2020 (UTC)


 * It seems what happened is that I mistakenly edited the B-Al-Sc table that I'd posted on our talk page, in response to a request from DePeip, rather than editing its predecessor. Sandbh (talk) 22:52, 18 December 2020 (UTC)
 * An understandable mistake, sure, but it felt bad. OK, enough.
 * Anyway: I think you misunderstood my request. My suggestion is, given current version File:Taxonomic_PT_wth_halogens.jpg. In the bottom row of categories, "alkali ... noble gas", please add the word "metalloid" to complete it. This would be the only change. The fact that it is mentioned already in the top row "metals - metalloids - nonmetals" is not enough, the bottom row better be complete by itself. (using plural -s or not is another issue, but at least we should be consistent within a row). -DePiep (talk) 00:34, 19 December 2020 (UTC)
 * Just a FYI, I think it would be better with the metalloid label only on the top. My preference is not a strong one, and I am happy to have it changed so that I can see it the other way for comparison; it may very well be that after seeing it I change my mind.
 * This whole incident points out one unfortunate consequence of having this graphic in an image that only one member edits. But the alternative of having it in wikitext would eliminate having the image show up in the mw:Page Previews, which would be unfortunate. YBG (talk) 01:24, 19 December 2020 (UTC)


 * Thanks, all fixed now. Sandbh (talk) 01:25, 19 December 2020 (UTC)

Psst psst, it's easier to edit the svg at File:Simple Periodic Table Chart-en.svg. Even a text editor will do. ;) Double sharp (talk) 07:45, 19 December 2020 (UTC)
 * Thanks. I'll keep that in mind in the future. YBG (talk) 08:17, 19 December 2020 (UTC)
 * ... and if an image not text graphic is needed/preferred, an image can be made from that public svg file :-) -DePiep (talk) 09:55, 19 December 2020 (UTC)
 * I propose to archive this thread shortly. -DePiep (talk) 23:41, 19 December 2020 (UTC)

Some editing to Periodic table
Based on conversation with I have done some cutting of the history and alternative-table sections, since those already have separate ancillary articles. I also cut out all the "periodic trends and patterns" that were added after the 2012 FA status was achieved, since as above I and others have serious misgivings about whether they DUE-ly reflect the literature. (That not only includes things added, but also the primogenic-symmetry thing that I added.)

I do feel that some of these things deserve to be mentioned: primogenic-symmetry, as well as secondary and internal periodicity, deserve a mention as they are common in Russian sources. As do diagonal relationships. However, I feel the current version is so overgrown with UNDUE stuff that I think cutting like this, to what sections the article had when it got the FA, to the trends most common for high schools to mention, is an improvement. Let's not forget that our readers, for this article about a topic that comes in so early that it might as well be the gateway to chemistry, probably would rather have a high-school or at most first-year university summary of the situation. Not throwing Cotton & Wilkinson at them and certainly not throwing the latest cutting-edge literature at them.

I also feel that the sections on periodic trends, as well as the basics, need to be rewritten/expanded based on standard high-school and early-university texts. But, I think this is a good first step. Double sharp (talk) 06:24, 19 December 2020 (UTC)

P.S. strikes me those general trends deserve to be rather in a 1st section on where the PT comes from and how to read it. Next bit instead should probably be about how to deduce stuff like the IE etc trends. Double sharp (talk) 06:41, 19 December 2020 (UTC)
 * I think it would be good to have a section "Overview" reinstalled. Problems with this section were pointed out by #On the PT article more generally (Oct 2020): "Is this [current Overview content] really what a reader needs?", and "... from the perspective of encyclopaedic content". The section was removed completely on Nov 15, referring to this discussion but not its content (being: change, not delete). The challenge is to distill the right overview content, especially since the TOC appears to be volatile.
 * Personally I think the first, or the main PT in there, can be in black and white, as the main features are visible by table structure itself (is why it is so great). That is: Z-sequence, rows, columns, blocks (to explain the steps up/down). Colors and further details can have their own place being a detail only (in sections, articles). Of course important trends sholod be used in OVerview too, but restriction is key in this.
 * Reinstallment also invites us to rethink the TOC as an article structuring base. Some content my be UNDUE indeed etc. as Double sharp here notes. -DePiep (talk) 09:47, 19 December 2020 (UTC)
 * Agreed, agreed! I just agree with EdChem that the previous Overview content was not doing a good job. I want to rewrite things based on his version...but it will require looking at some tertiary sources. As a stopgap I may work with Britannica first, and then do some refining. In a day maybe. ^_^ Double sharp (talk) 10:09, 19 December 2020 (UTC)
 * P.S. I think first PT should be coloured by blocks only, just because Z-periods-groups-blocks are the four important things. The first three are obvious (LTR obvious, rows, columns); the last one is not quite obvious (helium is s-block, and as long as Sc-Y-La stays you have to make it clear that a split d-block is indicated) and I feel colouring may be a good thing for that. Like WebElements (except they're naughty and call He p-block, very naughty XD). Category colours can be introduced when we introduce categories, that seems to make more logical sense to me. Double sharp (talk) 10:11, 19 December 2020 (UTC)
 * P.P.S. Inspiration: Physics Hypertextbook. Double sharp (talk) 10:15, 19 December 2020 (UTC)
 * (ec) Of course Overview can wait for the TOC to be stabile. I'd object to coloring the blocks in the basic PT, because it stresses one feature over the others. And, just as columns & rows are visibly present, so are the blocks -- by the steps of course. -DePiep (talk) 10:18, 19 December 2020 (UTC)
 * (Topic split and issue delays maybe?) -DePiep (talk) 10:21, 19 December 2020 (UTC)
 * Not really I think. Because I think you and Physics Hypertextbook just persuaded me that B&W first table is the way to go, so this discussion won't need to spiral out like a hydra. ;) Double sharp (talk) 10:23, 19 December 2020 (UTC)
 * Of course, in section Blocks (===?) you eh Block-editors can show their fireworks. -DePiep (talk) 10:33, 19 December 2020 (UTC)

Notification of intention to revert
I intend to revert several of Double sharp's recent edits to our periodic table article. RL obligations will delay the timing of my reverts. Once the reverts go ahead I expect the usual discussion will follow at the periodic table talk page.

I'm concerned about the undue removal of citation-supported content and that what's left is becoming too low level, and lacking in nuance, sophistication, and the wow (I didn't know that!) factor. Our article should be written up to a professional standard not down to a mundane textbook measure. Sandbh (talk) 12:08, 19 December 2020 (UTC)


 * For reference: the diff between "before my edits today" and "after my edits today".
 * I would like to request for others' views on my changes. Double sharp (talk) 12:21, 19 December 2020 (UTC)
 * P.S. Regarding a professional standard: I note that the Britannica article, which being in a standard encyclopaedia and being written by noted chemists J J Lagowski and Linus Pauling is about as professional as you could ask for, does not include the periodic trends that I deleted mention of. Double sharp (talk) 12:28, 19 December 2020 (UTC)


 * (ec) re Sandbh. Don't revert, discuss right away. Since the article is in full reconstruction since a few weeks, every GF edit has its place under the sun.
 * If differences of opinion exist about future of this article, all involved are invited to agree on general headlines first. Quibbling about every sentence will bring no good for the article. -DePiep (talk) 12:30, 19 December 2020 (UTC)


 * I generally agree with your removal of material. I had looked at those sections and thought myself that they seemed way too much in the weeds for an intro article to the PT. If the rest of the article were to be expanded to the same level of detail, the article would be completely unmanageable, IMO. However, I have one serious caveat. I agree with on the helpfulness of the wow factor. On this basis, I think that it might be good to incorporate some of the deleted material in one or more subsidiary articles, and retain a one-sentence summary in the main PT article. YBG (talk) 19:55, 19 December 2020 (UTC)
 * Please, please discuss before reverting. As far as WP:BRD, I analyze the situation as (B) your recent BOLD extensive expansions, (R) DS's recent trimming of much of this content, which brings us to (D) the appropriate next step is discussion. This approach is also in keeping with your ideas about continuous cooperative improvement editing expressed in . YBG (talk) 19:55, 19 December 2020 (UTC)
 * I have replied at . In short: so-called "continuous cooperative improvement editing" is not defined, and anyway WP:POLICY is complete enough to cover issues at hand. The "description" you refer to is scanty, and omitting any invitation or initiative to discuss. I cannot and will be hold to such an idea as editing rule, whatever it may actually be. I note that apparently, editing by those invented guidelies has brought us into this impasse: bold edits, and discussion about Revert, whil the content discussion is still not opened. -DePiep (talk) 20:43, 19 December 2020 (UTC)

, our periodic table article attained FA status for the second time, on 8 Jan 2018. At that time the Periodic trends and patterns section included:


 * 3.7 Linking or bridging groups.

Why was this cut?

The following sections were also in the Open questions section:
 * 6.3 Groups included in the transition metals
 * 6.4 Elements with unknown chemical properties
 * 6.6 Element with the highest possible atomic number
 * 6.7 Optimal form

Why were these removed? Sandbh (talk) 07:08, 20 December 2020 (UTC)
 * Firstly, the article periodic table did not "attain FA status" in 2018, it simply appeared as today's featured article (TFA) for the 2nd time. The article attained FA status for the 2nd time in 2012, after having lost it.


 * The sections you list were cut or merged because there are already ancillary articles, which the reader is directed to, that are supposed to cover these matters in more detail than the main periodic table article.
 * Section 3.7 "Linking or bridging groups" was cut because it is not a trend that appears in this generality in a majority of reliable sources (i.e. introductory textbooks on periodicity). Of course, the idea that changes across the table are gradual is, but the specific examples normally are not. BTW we also already have an article on periodic trends that presumably deserves some expansion, with just the most common examples being left to the basic article on the periodic table.


 * Section 6.3 was removed because it is not a debate about how the periodic table is shown, but a debate on categorisation. Whether or not an author thinks group 12 elements should be considered as transition elements or not, it remains that s/he will put them in the same place: between the group 11 elements and group 13 elements. There is no difference in structure to the table. Ergo, to me this belongs in the article on the transition metals proper and the article on the group 12 elements. There seems to be no reason why this particular categorisation debate should be given a whole section here, for example, when the debate on which elements are metalloids is not. Again, for that, we have a specialised article.


 * Sections 6.4 and 6.6 were merged into 6.5. Again, elements with unknown chemical properties is a matter of categorisation. The point at which it becomes relevant ties in to what Oganessian and others have said: if Og turns out to not be a noble gas, some questions have been raised about whether it is the end of periodicity or not. And we know that most people who have formulated eight-row extended periodic tables have decided that, since E172 is probably far more like a noble gas than E168 (probably a Sn homologue) is, 172 should be placed under Og, although Scerri recently suggested that it shouldn't. Therefore, to my mind, the relevant things were what I merged: whether the predicted "trend breaks" in period 7 and 8 should affect the shape of the PT. Since we already have articles on these matters (extended periodic table, transactinide, the individual articles on the superheavies), it seems to me that we only really needed a basic summary here and a link allowing readers who wanted to go deeper.
 * Section 6.7 was merged into the discussion of alternative periodic tables, as that's where it naturally fits to me.


 * Again, not saying I necessarily have the right way to do things. I just present my rationale for having done so and contend it had logic behind it. What matters is whether most of us think the logic was sound: I'm open to changes and will respect it if consensus turns out against what I did. I nevertheless think that some sort of cutting and shortening, with the reader directed to those ancillary articles for more should s/he want to go there, remains justified. Double sharp (talk) 09:17, 20 December 2020 (UTC)

, I was wrong to say the PT article attained FA status for the second time, on 8 Jan 2018. Sorry. I will instead say the article was judged worthy of being TFA, and it included the section on linking groups.

On the cuts generally, I suggest it is not a question of consensus turning out against these. Should I revert, in accordance with custom and practice, it would be a question of attaining consensus for reinstating the cuts.

On links to ancilliary articles, quite a few of the sections of the periodic table, have these. I nevertheless feel, for example, that diagonal relationships are one of the three key relationships, and that there should be a section on these in the PT article. Sandbh (talk) 06:50, 23 December 2020 (UTC) Sandbh (talk) 06:50, 23 December 2020 (UTC)

My impression: irreconcilable differences of opinion
My impression, as stated above, is that full agreement is very unlikely to be possible.

Looking at the above poll by YBG at, and ensuing responses between Sandbh and others (myself, EdChem, YBG, you) at sections , , and , my impression is that there are two opinions here on the future of this article based on how we feel that DUE, OR, and other policies ought to be applied. One seems to be held by Sandbh alone, and is in favour of the retention of sections like the ones I removed from "Periodic trends", based nearly wholly on primary-source material that is not common in the literature. The other, judging from comments above, seems to consist of everybody else here on the project, and be in favour of their removal.

Judging by how you've convinced me of some details and I've convinced YBG of some other details, it seems to me that a full agreement of all ELEM editors except Sandbh is very much possible and should not take much time to achieve. However, it seems that getting Sandbh to accept something the rest of us could agree on is going to be fruitless based on his responses when I and others talked to him. Especially given that we are back on the same tack of policy being viewed as an impediment to improving the article at above. Likewise, I do not think that the rest of us could really agree to accept what Sandbh would prefer here. That gives us an impasse.

By how consensus usually works as I understand it (counting both by numbers and how policy-compliant the arguments are felt to be – which is why WP:ILIKEIT as an extreme example won't get you anywhere at a deletion discussion), it seems to me that as the majority group we should be able to decide the future of the article. Since we also apparently have a majority view of how policy should be applied based on, that based on EdChem's words seems to be fairly consistent with general WP norms, it seems to me that we should also have a consensus to apply it that way. If that is how it really is (and the polling makes me quite sure that it is), and Sandbh is willing to accept that consensus interpretation of policy here is not the way he sees it and accept our consensus, then that's well and good. But this makes me wonder if Sandbh is going to accept it if we do so, and not continue to advance his own interpretation of policy. That would simply give us an impasse.

I hope I am wrong. But if I'm right, then I really do not know how to solve this, and ask all of you for opinions. Especially those who have had less of a conflict with Sandbh previously.

Perhaps I am being entirely too pessimistic because of past conflicts. To that end, I think the perspectives of the following four users would be the most helpful, to see if my pessimism is warranted or not:, who has been gallantly mediating the whole issue; , whose brilliant idea it was to gauge the differences in the poll I am now referring to; and , who managed to stay totally out of most of the fights and could therefore provide a fresh perspective; and you (DePiep), because AFAIK you've got no stake in the content issues that sparked all of this off in the first place. Of course, anyone else is equally welcome to give opinions. Double sharp (talk) 12:50, 19 December 2020 (UTC)


 * I was going to return to Wiki in the end of this month or in January, but it looks like this issue is spiraling too fast for a reconciliation of differing opinions.
 * here's a genuine suggestion: if this becomes too difficult for you two to find common ground on how this article should be organized, one way out of this for both of you could be to voluntarily resign from editing periodic table, and I will bring the article back into shape instead. I am now suggesting this not in order to overtake the article to improve it the way I like as if I didn't like either or both of your trains of thought; in fact, I haven't even thought too closely on how the article should be reorganized if at all so far, though I definitely will if I am entrusted to do this. I am suggesting this under assumption that both of you agree that it would be better for the future of both your relations and our segment of Wikipedia more generally if this argument didn't go any further than it has already, and since the article does require improvement, I offer myself to improve it bearing in mind that both of you have known for several years now and hopefully see at least some image of credibility in me when it comes to article writing.
 * There is a plenty of articles where, hopefully, you will not step on each other's toes. If this doesn't hold for this article, you can outsource the task. If that is agreed upon, I will not consult either of you on how to improve the article, and both of you will equally have little effect on what the final result will be. Perhaps knowing that and that the other one is willing to go for such measures can help overcome and abandon this dispute.
 * I will get someone to copyedit my writing when I'm done.--R8R (talk) 13:58, 19 December 2020 (UTC)
 * I am quite happy to resign from editing the article as well as offering opinions, and let you get it into shape instead, provided that is also agreeable to this course of action. Presumably the other ELEM members will still have and want to give opinions, though, and this will be a good test of whether the problem is really with my vision for the article vs Sandbh's, or if it's (as I pessimistically suspect) the views of everyone else's against his. More importantly, I do indeed trust that you will be able to write a great article. ^_^ Double sharp (talk) 14:33, 19 December 2020 (UTC)
 * if this becomes too difficult for you two [ie, Double sharp & Sandbh] to find common ground on how this article should be organized, one way out of this for both of you could be to voluntarily resign from editing periodic table, and I will bring the article back into shape instead.
 * I am surprised, and I object. One cannot request editors to "voluntarily" refrain from editing, and even discussing an article. It is not about common ground between editors, it is about the content of an article. Also, where would this leave other editors? How is this not a path to WP:OWN as an argument in content disputes?
 * It surprises me that when the article was edited boldly initially a few weeks ago, everything was acceptable, but with a similar bold reorganisation now there seems to be a problem.
 * The only route, and the WP route, is to discuss beforehand the setup and lines of periodic table. At least 's earlier comments should be used ('encyclopedic').
 * I note that all of this does not follow from the article itself. It also pertains to other articles. And so, we can expect to see (or already see) the same problematic editing practice in other articles. Also, the suggestion does not address problematic edits, only common ground of ideas. All in all: no discussion ahead by this, which is bad for articles and not WP-like.
 * So, R8R, I object to your proposal and I find excluding editors from editing this way is unacceptible. -DePiep (talk) 21:04, 19 December 2020 (UTC)

And now?
@Everyone: This is now looking at a single article and would be better on talk:periodic table. A link back to here might be useful but I wouldn't move all the discussion. It would be good to get some other editors involved.

Also, please can we avoid any more huge bold edits and / or huge reversions. We don't need to be arguing about whether we have an edit war, nor to start one.

I am deliberately making no comment on the content ideas here as the process needs comment more urgently. I do have some thoughts to add later, however.

, it sounds to me like you have made the removals with the thought that a blank page is the best starting place for that section, and that some of what was removed might be used in reconstruction. I had thought of the same approach with the previous overview section. With that approach, might I suggest: It's also worth remembering that having consensus doesn't mean that any other views / suggestions should be disregarded without consideration. I agree that unanimity is neither expected nor necessary, but minority views can still need inclusion under DUE. Suggestions from those we disagree with can help us achieve NPOV / balance, and may be better than some of our own ideas. I'm not saying you want to simply disregard Sandbh, I'm just reminding you and everyone else here that collaboration means considering all perspectives.
 * 1) Starting a section at talk:periodic table that outlines this approach, with a link to the removed content and perhaps the removed text as a comment, to facilitate the editing by yourself and others.
 * 2) One topic per thread, so I would separate the overview and trends discussions.
 * 3) Once you are ready, start a draft (open to everyone to edit, obviously), or with alternatives with the intent of seeking consensus before returning the content to article space.  One approach might be like the discussion I facilitated at the TFA page a while back.
 * 4) Be ready to set up a formal RfC if necessary to determine consensus.

, there have certainly been circumstances on WP where two conflicting editors stepping back to leave room for others to sort out an article can work. ArbCom have forced this at times with topic bans, as has the community, and voluntary ones are reasonable to request... but if a request is declined, then the choices are to request community / admin support or to give up on the idea. I am speaking in theory here, please note; none of this means that the request is necessarily wise or desirable in this specific case. There are certainly situations in which a request would be an inappropriate attempt to remove an editor whose perspective is necessary to NPOV coverage of views with DUE weight. And you are right that such an approach can be a step towards WP:OWN problems.

, given the prior conflicts amongst WP:ELEM members and differences of perspectives on various topics, I think any single editor effectively leading changes is a high-risk approach. We need project members to be better able to work together, and that is only going to happen by trying to work collaboratively. If that proves that the members cannot work as a team, it would provide strong support that the composition of the team is not functional... and that would lead us back towards ANI or ArbCom. Removing disagreeing editors on a case-by-case basis risks hiding the underlying problems, which I believe need to be exposed and addressed. I thank you for your suggestion, which I see as made in good faith and appropriate in some circumstances, and I hope that you will join in with the work... but it is my opinion that the voluntary stepping-back approach is not the best one in this case.

, thank you for posting to state your intent rather than just going ahead with a reversion. Discussion is good. As I see it, you are advocating that any changes / additions / removals be made from the starting point of the content that was included. Double sharp is advocating starting from a blank section and choosing what to include. Both are viable approaches. Which one is "best" is a matter of opinion, one where reasonable people can disagree. We could have a discussion and seek consensus on which to adopt... or we could devote time to discussion of the changes themselves by accepting the present version as the starting point. EdChem (talk) 00:30, 20 December 2020 (UTC)
 * Thanks EdChem. I am primarily interested in the response by Sandbh. Given recent responses, I will look for hooks that signal involvement in conversation (talk, discussion, consensus). -DePiep (talk) 00:41, 20 December 2020 (UTC)


 * I see where you're coming from. I'd like to add a couple of considerations, however, that may (or may not) add to the picture to somewhat change the thinking on what the way forward should be.
 * One of the reasons why I'm taking this sort of break where I watch the discussions but don't participate in them is because I found myself exhausted at first, and I have found myself irritated a couple of times over the last few weeks still. That may have been because what I saw was justifiably irritating or it may have been that I'm unreasonably irritable. One way or another, I've seen other editors handle those events better than I would have, so I'd say my abstinence is the best thing I could be doing for the time being. Despite that, this hiatus of mine will not be indefinite: either I clear my mind or things settle down.
 * It appears to me that we might be here in a similar situation here. It appears quite possible to me that a collaboration will not work for similar reasons, at least not now, and the best thing would be to leave the article alone for the time being. However, I don't really expect either DS or Sandbh to walk away freely when nothing is being done, and that's where I come in. The purpose is to make sure both of them know some progress is being made on our project's most signature article, even if it happens to be not quite what either of them at times.
 * I deliberately said that this was one possible way forward rather than the path to walk. We could simply work together on it, that would be great, but I realize that while this idea may work, it's not certain to work at all. We could also abandon the article for now and see how things will look in a year, but I remain doubtful that the article will really be abandoned, as someone will likely make an edit that will spark another conflict, not to mention that it means that we'll have an article in this state for another year before it gets better. Another idea could be to merely revert to whatever version was live when the article was assigned a bronze star, with perhaps some small tweaks if those prove necessary.
 * I'm not going to force this thinking onto either DS or Sandbh; I made the offer, and it's up to them to either accept it or reject it. I take it DS accepts it, but I will not make my move until I get a similar response from Sandbh, and I'll take a lack of response as a no.
 * Is this risky? Perhaps. Either DS or Sandbh may not be satisfied with my edits and pressed to add something the other one will take negatively and a breach of this compromise. There is also a possibility that what I come up with is not up to the highest standard, even though I hope this risk does not appear too significant in light of my previous FAs. As for content ownership, this issue relies on trust. Both DS and Sandbh have known me for years; I'm certain both of them can formulate an opinion on whether I'm prone to that and either trust me I'll be able to improve the article without descending into that or politely deny me that trust. However, to me, this risky option seems the surest way of those that I can think of, and that's why I proposed it. If either other editors (particularly, the two editors involved, Double sharp and Sandbh) don't agree, or choose another path forward, or even if they agree but there's a consensus in the project that my proposal is not a good solution for the problem at hand, then I won't implement it.--R8R (talk) 14:01, 20 December 2020 (½UTC)
 * However, I don't really expect either DS or Sandbh to walk away freely when nothing is being done – FWIW, I have considered it, but wanted to see how things would play out. Double sharp (talk) 14:31, 20 December 2020 (UTC)
 * re circumstances on WP where two conflicting editors stepping back to leave room for others to sort out an article can work. (R8R's idea). I understand, but my impression is that that is not solving our current situation here. For starters, and not irrelevant: it brings down questioning edits to questioning editors (a good distinction I picked up from you). Also, it leaves other editors chopped liver. Bottom line: it does not solve the problem at the core: editors cooperation issues. That includes general talkpage interaction, and discussing towards consensus (for example re source handling, DUE/UNDUE, encyclopedic aim, full article redesign, bold & revert). -DePiep (talk) 20:36, 21 December 2020 (UTC)
 * , I agree with you that it is not the best approach in the current circumstances, which I said above to R8R: I thank you for your suggestion, which I see as made in good faith and appropriate in some circumstances ... but it is my opinion that the voluntary stepping-back approach is not the best one in this case. EdChem (talk) 23:32, 21 December 2020 (UTC)
 * OK then. -DePiep (talk) 06:49, 22 December 2020 (UTC)


 * I find it very worrying and troublesome that now it has come this far . -DePiep (talk) 09:59, 22 December 2020 (UTC)
 * To be clear: the problem I find is not in the first part (ignored for now). It's the last sentence. -DePiep (talk) 13:29, 22 December 2020 (UTC)

Some items to consider
There are some items to consider:

On the question of whether or not to revert, I'm torn between (a) wanting to edit on a cooperative continuous improvement basis; and (b) the custom and practice of BRD as a consensus-seeking mechanism.

In (a), the obligation falls on me to demonstrate why the content should be restored. Whereas in (b), the original editor, rather than the reverting editor, needs to demonstrate why their edit should be restored.

Perhaps I should keep my power dry for the time being. --- Sandbh (talk) 03:25, 21 December 2020 (UTC)


 * I personally don't agree with this, sorry. My reasons are as follows.
 * The topic of chemical periodicity is one that appears very early on when people start learning chemistry in middle school. Well, they are likely to check the article on Wikipedia. They are not likely to want the subject covered at the level of Cotton & Wilkinson, and still less want that at the level of the most specialised literature, because they won't understand it. Should we scare them off right at the beginning?
 * WP happens to be a hypertext encyclopaedia. We already have separate articles for pretty much all the topics whose treatment has been cut down to size by me. Periodic trends, Extended periodic table, History of the periodic table, Alternative periodic tables. A reader who wants more can easily open one of those new pages in a new tab and continue browsing. So the higher-level information does not have to be lost. But this way it does not get into the way of the reader who isn't ready for all that.
 * The cutting-edge literature contains many things that are true but not yet standard. Such as the Aufbau thing Scerri mentions. So does it include many things that are widely accepted by specialists but are not yet universally accepted to be true and are surely not standard in textbooks. Such as the group 3 thing: if you looked at specialist sources it looks like Lu completely dominates, but if you look at textbooks you'll see it's really not yet standard. (And that is why I'm convinced we should currently stick with La for now.) We also can hardly tell if a very recently published paper is cutting-edge truth that must quickly be reflected, or just plain wrong and worthy of being ignored, when it has had no time to get an impact. That is for secondary and tertiary sources written after the fact to do. In short: we can't tell for ourselves easily if textbook errors are really errors, and by WP:VNT I am not convinced we should try. Double sharp (talk) 03:54, 21 December 2020 (UTC)

, thanks. I’ve amended item 4 to read…


 * "In a WP article on the PT I do not want a regurgitation of what introductory level chemistry textbooks have to say on the matter. In addition to introductory level content, I expect an overview of what the specialist, more reliable, literature has to say."

…because I feel that offers a way head. "And now?", as asked.

Articles should not not regurgitate what introductory level chemistry textbooks have to say on the matter.

Nor should they represent an overview of what the specialist, more reliable, literature has to say.

Rather, they should do both. Hence Item #6:


 * "…the start of the article is at the introductory chemistry textbook level and as you go down the article more content from specialist literature appears. That changeover probably applies within each individual section, too."

Thoughts?

General comments The PT article is not written specifically for people starting to learn chemistry in middle school. All sorts of people read WP articles. WP articles even get cited in academic journals. I addressed the question of people starting to learn chemistry, in item #6.

Middle school folk only need to read as far down the article, and as far down within each section, as they need to. The better ones will read further down. Meanwhile, more advanced readers may choose to read the whole article.

Specialised literature can be written more at the general textbook level.

I know WP is a hypertext encyclopedia. I used hypertext links in my FA articles on metalloids, and heavy metals.

I addressed the diminished reliability of textbooks in my dagger footnote.

Lu in group 3 dominates the specialist literature because it is a minor controversy for most chemists and nobody, up to now, has therefore seen the need to "defend" the position of La in group 3, in a comprehensive manner. --- Sandbh (talk) 05:26, 23 December 2020 (UTC)

R8R's proposal. I believe the usual practice is for the editor concerned to draft a revised article in their sandbox, before launching it in the article space, as opposed to freezing the article space, while one editor works on the article. --- Sandbh (talk) 03:25, 21 December 2020 (UTC)


 * yes, what you're describing would indeed be the usual practice; at the very least, it would be more usual than what I'm suggesting. However, if we were in a usual situation, I wouldn't have put this idea forward in the first place and I'd likely still be entertaining myself with more pressing RL issues to return to Wiki at some later date. I am fairly certain that the usual practice won't work. Let's take it seriously and assume bad things may happen, as they have earlier this year.
 * Firstly, my time for Wiki is rather limited. I was going to add to the collaborative development of the article because I thought it's be a common project for our project, but this has been tried, and the project finds itself in a disagreement once again. I'm not talking about who's right and who's wrong; I'm talking about how collaborative editing has not enabled the editors to reconcile their concepts of what the article should look like and agree on a compromise. So that didn't work, and I thought I'd go out of my way to help resolve the issue in a way that more or less resembles a compromise. If I am to be entrusted with resolving the issue in the way I'm proposing, I'll take that seriously and I will spend my time, as limited as it is, to help resolve it because I'm the responsible one in that case. If I'm not given this responsibility, there is really little reason for me to go out of my way to spend as much time on it, as I'm not the responsible one anyway---I do have other issues to devote my time to, so that plan won't work, either. What would another plan be? We either demote the article (which is something I'm fine with, but this will be seen as a bad outcome overall) or in absence of input from Double sharp and myself, you do rewrite the article completely, and let's say it's good enough to have nobody trigger a FAR in the end. This outcome will still likely result in a grumpy Double sharp, who will probably be very disappointed with the way things played out and will, based on my extrapolation of current events, assume you don't listen to other editors... so the question really is, will a periodic table article written by yourself but a broken relationship with Double sharp be really worth it? I'm not suggesting that he'd justified to think that and I'm not suggesting that he wouldn't; what I'm saying is, all of this sounds not as unlikely as it should. Is it really worth it to take the chance?
 * Secondly and far more importantly, let's say my spare time wasn't an issue, nobody signed up to any self-limitations, and I wrote an article in my user space and released it later. Let's say then there's something you think I missed, or you want to pinpoint, and you add it---sounds realistic so far. Then Double sharp finds that you push an unimportant point and complains about it---not unlikely, either---and you remain confident you're right and keep going your way, making some changes along the way that you find to be a way to bridge different editors---I see that happening, too---but Double sharp thinks you refuse to listen again, and then rage quits. Given how the events have been this year, this sounds far, far less unlikely than it should, regardless of who was right and who was wrong. Will it, again, really be worth it?
 * If I thought there was a way for you two to find a common concept of what the article should be, then I wouldn't be writing this. I can be proven wrong, of course, and I'll welcome a demonstration of that---it just doesn't look like I will. The only realistic way out of this conflict, as I see it, is to agree to put the pen down on this article, knowing that another person, not affiliated with either of you, will do the work. Maybe there'll be something you don't like, that's a possibility---but the chance that the other one won't like something is just as big. And maybe having somebody else write the article will help rethink what its content could be... there may be things that either or even both of you didn't think of first but found great in the end once somebody else had written them. As I said, I won't insist we do this, and this is the last time I'm offering to agree that you both resign from editing the article, assuming that if the offer is not taken now, it will never be. It's up to you to respond however you will, but I'm also asking you to think about not what is right, but about what the consequences of your choice will be and if they really look any more promising than what I'm proposing.
 * I expect that this heat in your relations is temporary, I really do think that, but it's here for the time being. It would be a shame to do something now that you won't be able to undo later.
 * Also, to be clear on that, I'm not suggesting the article is frozen entirely... there's a sterk disagreement between you two and I'm suggesting a solution that only involves you two and myself.--R8R (talk) 19:47, 21 December 2020 (UTC)
 * re Considerations in moving forward. Something is wringing between the OP ("And now?") and you post. This thread,, has moved into the editing process topic (for example, early by your ). Then followed subthreads and : editing process. Now your post here changes to content discussion without any bridging explanation. I other words: your " items to consider" (re article Periodic table) do not make sense. You are clearly invited to reflect on the editing process issues here, not stating your article proposals (yeah we asked for it, yeah they are due, no not here, no not now). I think we are waiting for a sign of engagement from you, clarifying how you want to cooperate with fellow editors. That could include responses to disputed edits, building fruitful talkpage thread discussions, considering restrictions re undesired talk forms (see ), etc. I note that one ELEM editor has recently expressed frustrations with the editing process, pondering withdrawal from editing. BTW and for the record: the term "cooperative continuous improvement" is not clear, not binding and not a solution (unless it says: "Wikipedia editing process"). -DePiep (talk) 20:14, 21 December 2020 (UTC)

Inconsistent approach
I've been doing some work on the post-transition metal article aka Metals close to the border between metals and nonmetals.

I noticed entries in this article for the super-heavies: Cn, Nh, Fl, Mc, Lv, Ts, Og.

Why are these entries here? You won't find this stuff in general chemistry textbooks.

My answer is that they are here because they are interesting and sourced from reliable specialized sources. And that's a good thing. It's consistent with items 4, 4a, 6, and 12 above.

Yet, according to recent deletions from the periodic table article they shouldn't be here. What gives? Sandbh (talk) 05:29, 25 December 2020 (UTC)
 * Because this is a specialised article about a specific group of elements, and not a first entry-level article like periodic table with explicit subarticles. Double sharp (talk) 06:08, 25 December 2020 (UTC)
 * Also, you will find a cite to this stuff in G&E (referring to Fricke's predictions) even if not a mention. I should probably get around to citing that stuff, anyway (just was too lazy earlier to do the work of converting all the citations to your style). But, after the last few days, I feel like I don't have either the time or the energy to continue arguing this, even if I still have concerns. So, I cleared my watchlist like YBG and decided to return to lurking mostly. Not sure if I will be doing much anytime soon (apart from clearing a few loose ends like those missing cites). So, you can do what you want, and I'll let people who actually are going to be active comment on it if they so wish. Double sharp (talk) 06:37, 25 December 2020 (UTC)

, OK, thanks for that. Sandbh (talk) 08:29, 25 December 2020 (UTC)

The PTs in the PT article
Separated from, which is mostly about the text of the article, specifically, the recent expansion and subsequent contraction of Periodic table. This thread relates to the periodic tables to be displayed in the article. Feel free to revert if you think this separation is wrong. YBG (talk) 20:27, 19 December 2020 (UTC)

Two questions re graphic design
Two things:


 * I kind of think, that for clarity the first table should be 32 column. Since everyone agrees that somehow the f block really belongs in there. Then, very first subsection can be a mention of the 18 vs 32 issue, to say that theoretically the table looks as 32, but in practice mostly it is seen as 18. Do you think that makes sense?
 * First PT should be B&W, we agree. How much should be on it? Should it be the showcase-style one with everything? I am inclined to say no, this PT should just stick to very basics for same reason you mention. Symbols are a must of course. Probably Z as well. Should names be there? Maybe is actually best illustration... Of course, showcase-style one with everything may come late in section when the standard PT template is given. ;) Double sharp (talk) 10:27, 19 December 2020 (UTC)
 * reply to "First PT": b/w, 32-col, (earlier version).
 * 1. Would break top-image-convention (by its ratio)?; todo is check what is acceptible as page design.
 * 2. What in the cell? Z and symbol, I'd prefer written as "80Hg". Possibly name. Not atomic weight: not a main feature (sorry Mendeleev), relevance is difficult to grasp at first, for example has little RL connection. Clickable? No absolute requirement, we can see later whether this would an improvement. (BTW we can overlay an image with click-areas using mw:Extension:ImageMap). -DePiep (talk) 10:46, 19 December 2020 (UTC)
 * I think 32-col would be possible as Wikitext if we just went for Z and symbol and gave up on the name. After all, we are going to have a lot of PT's in this overview section to highlight different things. ;) Double sharp (talk) 10:49, 19 December 2020 (UTC)
 * I would like it to be 32-col, but that would only be possible if the cells are restricted to Z and symbol, as $120 Og$ or $120 Og$ or $120 Og$. If we were to use 120Og, it would consume about twice the horizontal real estate. YBG (talk) 20:18, 19 December 2020 (UTC)
 * Yep, I met this too. Currently Periodic table/black-and-white/32-column puts number mid-under symbol (my slight preference): $Og 120$ -DePiep (talk) 22:30, 19 December 2020 (UTC)
 * , although Og is actually E118. ;) Double sharp (talk) 09:19, 20 December 2020 (UTC)
 * 3. In the b/w demo, please take a look at the period number style: . We haver discussed this before, I'd like to reopen this proposal later on. Also in there: column headers are grey. Is that cheating in b/w? If we can deliver same meaningful & clarifying image without greys, would be great design. -DePiep (talk) 11:07, 19 December 2020 (UTC)
 * Solution to prevent greys: found, looks good, stresses column! See the b/w demo.


 * reply to "first subsection": yes, we best write: the same PT is commonly presented in 32- and 18 column form" (half those words, & keep neutral=best). Or show ~ Periodic Table overview (wide).svg  =  15LaAc with LaAc in group 3.jpg -DePiep (talk) 11:22, 19 December 2020 (UTC)


 * I think the first PT in the article (outside the lede) should look as similar as practical to the PTs that readers are likely to have seen. I also think it should contain only the most important aspects.  So, I support black and white and including only symbol and atomic number.  I am open to being convinced otherwise, but to me the most common arrangement has the atomic number above the symbol.  I suggest a larger font size for the symbol as that is the identifier for each cell.  Following through on familiarity, I am thinking that an 18 column PT is the default in most situations and so wonder if the 32 column version wouldn't be better left to later in the article?  EdChem (talk) 09:50, 20 December 2020 (UTC)
 * , I agree that atomic numbers usually come above the symbol, so I would support that.
 * As for 18 vs 32 as the first table, you have a point. Maybe 18 should be first indeed. I just think 32 should probably be dealt with immediately afterwards, because it is pretty much the most obvious way to explain to the reader what those asterisks in the 18 form mean. Double sharp (talk) 09:53, 20 December 2020 (UTC)
 * Yes, the asterisks need to be explained. My personal view is that the asterisks under Sc / Y and 15 element f block is confusing as it implies the entire f block plus either La / Ac or Lu / Lr belong in group 3, and it complicates the idea linking blocks to electronic structure.  I think it is inferior to either possible 14-electron f-block tables and should be deprecated... or, better yet, burned at the stake.  Unfortunately, my personal views and thoughts with an educator hat on carry no weight when textbooks and IUPAC use it.  Consequently, with my Wikipedian hat on, I can't object to an asterisk table on any policy ground.  The only thing I can do is support an explanation being provided as soon as possible.  If the overview gave an 18 column table with a 14 element f block, I think explaining the link could be left much later... but with a 15 element f block, I agree that the 32 column logically needs to appear early on.  EdChem (talk) 10:08, 20 December 2020 (UTC)
 * I think we're in complete agreement. This said, it seems the current WP situation is for tables to generally show a 14 element Ce-Lu / Th-Lr f block, and as far as I can see no one wants a 15 element f block. So perhaps we can explain the link a bit later indeed. ^_^ Double sharp (talk) 10:17, 20 December 2020 (UTC)
 * the diagram appearing a few posts above from showing a 32 and 18 column table has a 15 element f block in the latter.  EdChem (talk) 10:41, 20 December 2020 (UTC)
 * So it does. If that's going to be in there, then I agree, 32 needs to appear early. ^_^ Double sharp (talk) 10:43, 20 December 2020 (UTC)
 * Size of displaced f-block (14 or 15), in 18-col form, is free and given a single group 3 choice preferably would be 14. Does not change the principle of 18-col form. -DePiep (talk) 11:38, 20 December 2020 (UTC)
 * about main or first PT to show: 18- or 32- column?
 * I'd say we use 32-column. Our first and foremost reflex, here at WP, should be: use a "32-column PT" whenever possible. f-block displacement is a degradation and a compromise, while needless. Adding an f-block displacement is adding complications and distractions, and is the opposite of showing the essence of the PT&mdash;our encyclopaedic task is to convey the structure of the PT. For now, we take that that structure is Z, columns, rows, blocks (improvements possible). BTW, so far we are seeing that even black/white can do that, which is a great graphical point. Adding colors here is distracting from the essence too, be it used for blocks, categories or el.negativity.
 * When introducing the PT to first-time scholars, there is no gain in having to spend time on explaining the use of asterisks & dislocation. And from that introduction (of a nuisance) on, the scholar has to mentally make the relocation (f-block into main table), each and every time they see the PT, before finding or understanding whatever they look for. Is there any atlas or clockface or DNA-model or painting or other visual icon that uses this presentation trick?
 * Then you mention that 18-column is "the default in most situations", which likely and pitily is true but habit is not an strong argument IMO. Reasons are "bookpage ratio" (sound, but not applicable for webpage, and let's note that the classroom/lab wall has an excellent ratio for the 32-column PT). Anyway, using the book ratio is a compromise, moving away from good teching and describing. Also is used (not by you) that people are "used to/familiar with the 18-col form". That may be true for, by definition, experienced scholars & scientists (ELEM/CHEM editors included). But these people already *do* have learned to mentally position the f-block correctly, by spending learning time.
 * Another reason for an 18-col form is historical science: ca. 1930, before Seaborg et al discovered actinides and f-block, 18-column was the complete form; then those who could follow Seaborg et al could easily mentally adopt the extension. But that was 80 years ago. (We have also accepted the change from from Reihe-form to Period-form, ca. 1905?). In other words: our familiarity with 18-col presentation should not blind us for the best encyclopedic presentation for this concept.
 * As a detail: when you write "32 should probably be dealt with ...": sounds like a aversion . Hope you won't hit it too hard off-school. Whichever form is in top, we should make just one thing clear: 18-col and 32-col are presenting the same Periodic Table. And this without judgement or whatever background or wording. Or, as I illustrated above @11:22 : "icon1 = icon2".
 * TL;DR
 * 32-col is encyclopedically ideal, as it shows the PT essences in one figure. It has not that complicating displacement, always requiring a mental step, of re-insertion, before approaching the PT. Habits (both 'the form most often seen' and scholars/scientists being 'familiar' with) is opposite to the encyclopedic job of describing the PTs essence. We should clarify, as simple as possible, that the 18 and 32 col forms are presenting the same PT (they depict the same scientific statement). Between 18-and-32, content does not change. -DePiep (talk) 11:34, 20 December 2020 (UTC)

, why do need asterisks? --- Sandbh (talk) 03:40, 21 December 2020 (UTC)
 * What I said about displacement & having to reposition. "f-block displacement is a degradation and a compromise, while needless." The asterisks themselves are not the point. -DePiep (talk) 08:18, 21 December 2020 (UTC)

Re group 3 in the draft above
I notice that you show a Sc-Y-Lu-Lr group 3 in your template Periodic table/black-and-white/32-column.

Now, I personally decided that I'm happy waiting for IUPAC. And I had thought everyone was OK with that, even the Lu supporters (I personally am OK with it), but considering this reappearance and 's mentioning of the reasoning for his Lu support again at, perhaps I was presumptuous in presuming that. And indeed, if others want to continue the talk, it is their right even if I currently don't: I feel I can't and shouldn't decide that over their heads.

That said, I also feel that there is little point asking for a reiteration of our opinions when we just had one two short months ago (Wikipedia talk:WikiProject Elements/Archive 52), and I do not think anyone has switched camps. So, I guess the question is: does anyone want a formal closure (that would assess the consensus and/or lack thereof) of the turnout at Wikipedia talk:WikiProject Elements/Archive 52 by an uninvolved party, or is everybody happy with just dropping it for now and waiting for IUPAC (like I am)? If anyone wants one, then we need to find someone uninvolved who is willing to determine consensus for such a long discussion with long supplements.

Double sharp (talk) 09:47, 20 December 2020 (UTC)


 * Changed into g3=Sc-Y-La-Ac, as current enwiki standard (I just started with some old frame). No urgence to establish/reopen this issue *here*. Once we find a (another) stable outcome for group 3, by WP-consensus or by IUPAC, we will apply that to all out PTs consistently.-DePiep (talk) 10:07, 20 December 2020 (UTC)
 * OK, thanks for the explanation, . ^_^ I just want to make sure everyone is also happy with this state of affairs and that I haven't acted over anyone's heads by deciding to wait for IUPAC. Double sharp (talk) 10:12, 20 December 2020 (UTC)

Since you !voted there, and apparently I forgot to ping you. Double sharp (talk) 10:13, 20 December 2020 (UTC)
 * (reply/add to self:) ... However, for encyclopedic reasons I'd reject any dual form "can be La, can be Lu" notation in the first PT. It's just one of those details to be fleshed out in a section or article (say, like hydrogen positioning). Any footnote explanation is distracting and complicating. -DePiep (talk) 10:14, 20 December 2020 (UTC)
 * agreed. My current view is that no footnote is generally needed. Double sharp (talk) 10:54, 20 December 2020 (UTC)

A direct comment about the above B-Al-Sc thread
I share the impression that the correlation coefficients thing is very much OR. I also have the general impression with this current conversation is that what is writing in favour of (B-Al-Sc) is, for the current era, so out there that it becomes actually difficult to find sources refuting it. B-Al-Ga is today common knowledge, and B-Al-Sc is mostly only done for comparison if done at all. Nobody takes B-Al-Sc as a vertical column seriously anymore! When it was Sc-Y-La vs Sc-Y-Lu, that was one thing, because that debate is a current one. I am not aware of the B-Al-Sc thing being held by any significant group of chemists for the last fifty years. Of course, the problem is that this makes it difficult to respond without oneself engaging in OR, synthesising common knowledge about periodicity in order to address an argument that no one for the last fifty years has considered worth making. No wonder didn't provide a source, because his argument is so standard about secondary periodicity in Russian sources. When something is really standard, you often forget exactly where you heard it from. Compounding the problem I guess is that most of the systematisers of the periodic law were Russian and wrote in Russian and are therefore hard to access for most of us.

This being said, I have luckily managed to find a recent article that tackles the matter as a historical one. This one. By Naum Imyanitov, published in 2018 in Foundations of Chemistry, the same journal Sandbh published his group 3 article in (so there should be no objections on unreliability on that front).

An erroneous belief, that change in properties along the group occurs in a monotonous manner, is also reflected in the study of the dependence between the formation heats of oxides and the atomic numbers of the elements (Roth and Becker 1932). Authors concluded about the monotony of the changes for the following rows.

B − Al − Sc − Y − La C − Si − Ti − Zr − Hf − Th N − P − V − Nb − Ta S − Cr − Mo − W

However, did not pay attention to existence of secondary periodicity in more representative rows within the same graphs

B − Al − Ga − In − Tl C − Si − Ge − Sn − Pb N − P − As − Sb S − Se − Te

Okay. So that basically is it. When B-Al-Sc was given as the right periodic trend (which is not even always in that era), it came because people thought that trends are monotonous and did not know about the secondary periodicity. Today we know better, case closed. And please, look at the dates of that misunderstanding: 1932, hardly the sort of understanding we want to reflect today. Secondary periodicity was already discovered in 1915 by Biron, you know: it just had to wait a while to get noticed because the date meant it was swallowed up by WWI and the ensuing Russian civil war. Imyanitov also explains it there as a combination of kainosymmetry, contractions, and relativistic effects. All three of those are well-known everywhere, although maybe not by that name in the case of kainosymmetry (for English speakers it is either the primogenic repulsion or the first-row anomaly). And that's exactly what DA and I have been talking about except that we couldn't remember the sources we learned this from quite some time ago.

I am tempted to ask if we could perhaps move on and not keep on talking about forgotten chemistry, especially when it has mostly been forgotten since before Queen Elizabeth II ascended, and more recent sources indicate that it has been forgotten for a reason. And I also wonder why this was brought here in the first place. Is it to argue that WP should show B-Al-Sc? But then there are no sources, just OR correlation coefficients, so why should we change based on OR? I'll take that back if it turns out this analysis has been previously published, in which case I'll instead say it's hardly DUE when almost nobody else for the last 50 years wants B-Al-Sc. Is it just to show us something cool? Then OK, we've already reacted, we've explained why B-Al-Sc misses the point of periodicity and referred to sources saying so and/or using similar arguments to those we got from remembering what we learned about basic periodicity. That should've been enough. So why are we still on this over a week later? Why has a periodic table with Al over Sc suddenly appeared in the periodic table article? And why is an OR diversion and DA and I explaining to Sandbh based on modern sources why his view is outdated and based on an incomplete understanding of general trends now taking up more of our limited WP time than something we can actually put into articles?

Please forgive any signs of bluntness here. I have tried to rein it in, but some frustration probably seeps through unconsciously. I just sense that this is starting to go in circles while having had very little relation to a typical WP article-improvement discussion from the very beginning. I contributed to it, OK, I somehow had the idea that referring to general trends as commonly known in sources (as seen above) could perhaps persuade Sandbh away from something his analysis that doesn't consider those convinces him of. Apparently it cannot. So, all right, this will not be productive. I went to an OR thread and used some element of OR (that at least is mostly supported by known generalities rather than new statistical analyses, but fine, it was still long and not cited) to attempt to explain why it doesn't work, so I contributed to the problem. So let's stop contributing to the problem and nip this in the bud then. And just ask that such historically interesting but superseded things like Al over Sc please please not be presented to the general reader as if they were common, modern things in any context. And in the discussion, for the vast landscape of RS in the world to please not be cherry-picked to pick out the tiny fraction arguing for B-Al-Sc that are recent in order to create some sort of dichotomy I have not previously heard of between chemical B-Al-Sc and electronic B-Al-Ga when just about everybody thinks both chemistry and physics support B-Al-Ga. And please not more stuff about context being the most important, when I still have yet to see any textbook writers follow that idea and actually change their PT layout in order to emphasise different things. Please? Double sharp (talk) 06:23, 15 December 2020 (UTC)
 * Double sharp (talk) 06:25, 15 December 2020 (UTC)


 * 1. I hope my What's important response to ComplexRational will overtake all of this. 2. More to follow. Sandbh (talk) 06:47, 15 December 2020 (UTC)
 * No, it does not. It simply contains reassertions of your idea of context being the most important, when no standard textbook "walks the talk" by actually changing its tables depending on context. I wonder where the DUE is here. It contains reassertions of the idea that B-Al-Ga is against the "damned chemistry", when essentially the whole literature today agrees on B-Al-Ga even for a chemical table. I wonder where the DUE is here. And it contains an idea to write about the four variations including B-Al-Sc, thus giving it an incredible amount of UNDUE given exactly how minuscule its support in the standard literature of today is. Wait, that already almost happened, an Al-above-Sc table is now in the periodic table article already. So, no, from my perspective it rather confirms it rather than overtakes it.
 * I will stop here because I am frustrated and do not want to say things I will regret later. Nonetheless, I strongly suspect from the above that my frustrations about OR, DUE, and not understanding commonly accepted modern chemistry are shared. I already pinged YBG, DA, and EdChem who, if I understand them correctly, have made similar complaints about the above thread. Given our previous 2020 history, perhaps it would be better to have them articulate it now that I have. Double sharp (talk) 06:55, 15 December 2020 (UTC)



OR. A simple XY plot of Z v. the value of property for, say, five elements, showing the R2 value is not what I would call OR. Thousands of authors comment about trends going down groups. G&E discuss this and draw the XY plots, including for groups 3 and 13 in the same chart, and comment on the difference in the curves. In any event WP:OR doesn't apply in the talk space.

Secondary periodicity is a non-issue. It is there in the modern form since B-Al were moved over Ga, with the acceptance of quantum theory, and that is fine. The smoother trend going down B-Al-Sc-Y-Ac can still be seen in the n + 10 rule, which dates back to DIM's time. Neither option is superior, although B-Al over Ga is the popular form. Looking at G&E’s chart one wonders why this is the case (because of quantum theory, of course).

B-Al over Sc: This was not a proposal, as ComplexRational noted.

There are no sources “refuting” it since its validity never went away, even though it was displaced by quantum theory.

I explained why Al appeared over Sc in the isodiagonalty part of our PT article in my response to EdChem:
 * ”I posted those changes to our periodic table article since I saw how clean the article structure looked, subsequent to our trial of continuous cooperative improvement editing, and noticed the periodic trends and patterns section could be improved with some discussion about isodiagonality and oxidation number trends. In this context I don't understand 's comment about not paying enough attention to patterns and general trends. Those changes had nothing to do with my advocacy for La in group 3, in a chemistry based table. I also advocate that Lu in group 3, and He over Be, is an excellent electronic based table. More important than either of these is that the context for the choice of particular periodic table is explained. It so happens that the La table happens to be currently more popular in the corpus of chemistry.


 * Aluminium in group 3 is convenient way to show the diagonal relationship between Be and Al and between Al and Ti, as has been reported in the literature. The Inorganic chemist's periodic table, for example, shows Al over Sc, with a diagonal relationship. Rayner Canham's corresponding article on isodiagonality has a PT with Sc over Al, showing the same kinds of relationships, with a diagonal line running through Li-Mg-Al-Ti-Nb-W. I see no controversy here.”

Textbooks suffer from various issues in terms of reliability including the well-known and documented text-book error phenomenon. As you noted, higher-level sources are more reliable in these areas.

Context: The importance of this is self-evident:


 * The so-called IUPAC Table is more of a chemistry focused pragmatic table.
 * The form with lutetium instead of lanthanum in group 3 is more of an idealised table, instead of a pragmatic table with "no need to lose sleep".
 * In a Solid-state physicist's table, both lanthanum and lutetium, as 5d metals, go under yttrium ((Vosko and Chevary 1993).
 * In the electronegativity-focused Pauling Table, group 3 is boron, aluminium, scandium, yttrium, lanthanum, and actinium.
 * Aluminium over scandium is more of a Metallurgist's table (Habashi 2009).
 * Geochemical tables (McSween and Huss 2010; Railsback 2018) emphasise property trends important for the earth scientist, i.e., they give up the beauty of symmetrised arrangements in favour of irregular chemical facts. Some tables define the carbon-silicon group as containing titanium, zirconium, hafnium rather than the standard set of germanium, tin, lead.
 * In the Astronomer’s tables (Esteban et al. 2004; McSween and Huss 2010; Yamamoto 2017), hydrogen and helium are the only non-metals and all the other elements are labeled as metals.
 * In a Superconductivity periodic table, group 2 is split into: barium and radium; calcium, strontium, and ytterbium; group 12 is beryllium, magnesium, zinc, cadmium, and mercury (Wittig 1973).
 * A periodic table with hydrogen over boron makes for a nice designer table (Luchinskii and Trifonov 1981).

Scerri raised the notion of a continuum of periodic tables, with physics and chemistry ends, and with the currently popular La form somewhere near the middle. Schwarz discussed didactic, designer and chemical periodic tables:. Note the acknowledgement in the latter article to Naum Imyanitov.

History: Our PT article discusses Meyer’s table; Newland’s table; DIM’s tables; the modern short form; the LSPT; the 32-column form; the IUPAC form; the La form; the Lu form; and Benfey’s table. The historical B-Al-Sc-Y-La form was referred to by such luminaries as Rydberg, Seaborg, and Pauling. —- Sandbh (talk) 11:48, 15 December 2020 (UTC)


 * , you wrote that A simple XY plot of Z v. the value of property for, say, five elements, showing the R2 value is not what I would call OR. Thousands of authors comment about trends going down groups. G&E discuss this and draw the XY plots, including for groups 3 and 13 in the same chart, and comment on the difference in the curves. In any event WP:OR doesn't apply in the talk space.


 * You might not call it OR, but it is definitely OR. WP:CALC talks about routine calculations and I am sure that determining a trendline or curve and calculating R2 is routine for you (and for many scientists) but you are not taking a source's work and performing basic arithmetic (by changing units, say); you are taking raw data and performing a statistical analysis.  In so doing, you are choosing a quadratic model, a choice that you have not justified based on theory, and I for one do not accept that the results are correct (in the sense of justified, I am not disputing that the calculation yields the results you state).  To engage in a brief OR diversion to illustrate this,  and others have stated that the trends should deviate when an additional filled subshell appears.  On that basis, a multiple linear regression approach might be considered.  Perhaps you have tried that &ndash or not, it doesn't really matter.  The point is that such considerations are explored in a journal article and considered by peer review before they are published.  Even then, inclusion in WP depends on the work being used by others.  Your argument for restructuring the PT to align with long-ago discontinued practices based on your analysis of trends in data you selected and performed your own statistical analysis on is absolutely positively OR.


 * Many authors discuss trends, that is not in question, so if you find one that presents the trends that you are showing, and uses them to support similar conclusions, present them. The OR problem will disappear and we can look to how those publications have been regarded in secondary are tertiary sources, so as to examine coverage with DUE weight.


 * Regarding talk pages, please note that WP:TALKPOV states that Talk pages are for discussing the article, not for general conversation about the article's subject (much less other subjects). Keep discussions focused on how to improve the article and that Talk pages are not a place for editors to argue their personal point of view about a controversial issue. They are a place to discuss how the points of view of reliable sources should be included in the article, so that the end result is neutral (emphases in original). You are correct that WP:OR, a policy that prohibits original research, does not apply to talk pages as OR is not prohibited on talk pages.  However, the purpose of discussion pages is article and project improvement.  Neither policy nor practice sees unrestricted or even extensive OR as encouraged or permitted.  EdChem (talk) 03:20, 19 December 2020 (UTC)


 * I discern a regrettable lapse into "you did this and that" behaviour. I hope this will not continue.

WP:CALC. These days, determining a trend line is no harder than entering a few numbers into Excel, clicking on "insert chart", clicking on "XY", clicking on the preferred kind of trend line, and clicking on "add R2 value". And it's done.

I'm looking up data easily found on WP, drawn from RS. That's not what I would call "raw".

I'm choosing a quadratic, since such a trend line consistently yields the best goodness of fit values. I know this because I clicked on all the other trend lines to see what kind of R2 value they gave. I learnt about quadratics in year 8 school. These days, a 10 year old could do all this.

Mapping trends down groups goes back to the days of DIM. Have a look in G&E and check all of their trends lines for multiple groups, Yes, the trend lines can deviate when an additional filled subshell appears but they don't always do so. I just plot the values and turn them into XY charts, using a consistent approach. I don't know enough mathematics to comment on the need for a multiple linear regression. In what way would this be required or better?

Restructuring the PT: Contrary to your assertion, I made no argument to restructure the PT based on my work. What happened was that I looked in the literature to see what that said. And I found plenty of examples of B-Al- over Sc. And I found a lot of things that I did not know that are relevant to, and will improve, our periodic table article.

WP:TALKPOV: That policy you're referring to applies to article talk pages. Our project page is not an article talk page. It is a page for discussing matters pertinent to WP:ELEM goals. One of the goals of our project is, "7. Other side projects of articles that project members see fit to pursue." --- Sandbh (talk) 07:16, 19 December 2020 (UTC)


 * To clarify, when I added that as a goal, in 2016, I thought it was fairly obvious that it meant: if you feel like improving any article, that isn't already on the goal list, go ahead and do it. And, indeed, I added a parenthetical example about my lanthanide campaign (fruits of which you can see are GAs of lanthanum, cerium, praseodymium before I got tired of writing about the same chemistry over and over again). It was about picking articles to improve (to GA, to FA, maybe even just to presentable aka B), not chatting about stuff with no possibility of ever getting into articles due to OR-ness. Double sharp (talk) 07:25, 19 December 2020 (UTC)


 * Also I think you'll find EdChem's point being corroborated by What SYNTH is not:

Treatment of numeric data is an encyclopedic issue: summarization by sum, average, etc. are necessary expedients, and should not be confused with original research. As an example, if a source shows (without any total following it) " ", a Wikipedia article can express the same data with summarization " ". (Whether to express only the result, if it is not explicitly given by any source, could be a point for discussion, but in any case it is not SYNTH.)

Summarizations based on statistical methods, however, are original research by synthesis, as they involve the reinterpretation of data, and decisions about which statistical methods and significance levels are appropriate. It is common to confuse the arithmetic mean (summarization) with the expected value (mean and standard deviation calculated by a reliable source). If in any doubt about summarization vs. statistical reinterpretation, discuss it first.
 * Double sharp (talk) 07:29, 19 December 2020 (UTC)

I'm not referring to your goal. I'm referring to WP:ELEM goal #7: "Other side projects of articles that project members see fit to pursue."

That link you provided me is not one of Wikipedia's policies or guidelines. Even so, it says at the top: "This page in a nutshell: Although avoiding original research is an important part of ensuring that Wikipedia content is verifiable, use some common sense about it… [italics added]. The trends I have referred to are well known, and go back to DIM's days. What happened to common sense here, and considering the principle behind WP policies, rather than the black letter law? Not forgetting WP:IAR, if a policy is getting in the way of improving WP. I know what those other policies are for, to stop genuinely original thought being posted in the article space. That's not the situation here. Sandbh (talk) 11:47, 19 December 2020 (UTC)


 * I think you'll find, if you step through the history, that that is precisely where that goal on ELEM came from. My lanthanide campaign was just an example I posted. With the symbols e.g.. That's why I could later add another example (my halogen campaign), and finally generalise it to all article projects. All my edits, and I like to think I know what I meant.
 * If we're back again at the idea that policy is getting in the way of improving WP – then I think it's best for me to stop responding (as we've gone through this already) and let continue this conversation with you if he wants. Reason, again, being so that I don't out of frustration say something I regret later. Double sharp (talk) 12:00, 19 December 2020 (UTC)

The earliest reference I can find to the side projects goal is 17 Nov 2007, by Nergaal. --- Sandbh (talk) 23:08, 19 December 2020 (UTC)

In my opinion, the critical point here that we (as a project) need to address is whether statistical analysis such as that you have presented is OR or not. My view, supported by the content at What SYNTH is not, is that such analyses are unambiguously OR. You have written above that These days, determining a trend line is no harder than entering a few numbers into Excel, clicking on "insert chart", clicking on "XY", clicking on the preferred kind of trend line, and clicking on "add R2 value". And it's done. On the process, you are obviously correct. However, "clicking on the preferred kind of trend line" is making an analytical choice, one that should be grounded in an understanding of the theory that is being automated by the software. Whether or not the editor making this choice is making it knowingly does not matter to whether or not the outcome is OR in the WP sense.

Presenting material from older sources as a base for potential edits in article space is fine, as is bringing together multiple sources on the topic for a talk page discussion. I know many trends are discussed in the literature. If you can make a case without using OR, its strength can be considered for implementation. If your view can only be supported with OR, then it does not belong in article space. When the discussion is addressed at article space change, it needs to be conducted with WP:TALKPOV in mind. EdChem (talk) 23:48, 19 December 2020 (UTC)


 * The basis for clicking on the preferred kind of trend line was explained. Namely the polynomial option consistently gave the best option. For example, 8 of 12 chemical properties for B-Al-Sc-Y-Lu have better than 0.90 goodness of fit values. Such an analysis is trivial.


 * In this context, I posted elsewhere from WP:COMMON:
 * "Wikipedia has many policies or what many consider "rules". Instead of following every rule, it is acceptable to use common sense as you go about editing. Being too wrapped up in rules can cause loss of perspective, so there are times when it is better to ignore a rule. Even if a contribution "violates" the precise wording of a rule, it might still be a good contribution. Similarly, just because something is not forbidden in a written document, or is even explicitly permitted, doesn't mean it's a good idea in the given situation. Our goal is to improve Wikipedia so that it better informs readers. Being able to articulate "common sense" reasons why a change helps the encyclopedia is good, and editors should not ignore those reasons because they don't reference a bunch of shortcut links to official policies. The principle of the rules—to make Wikipedia and its sister projects thrive—is more important than the letter. Editors must use their best judgment.


 * Why isn't "use common sense" an official policy? It doesn't need to be; as a fundamental principle, it is above any policy."


 * WP:COMMON is not a WP policy or a guideline. That said, it captures how I feel.


 * On the trends in question I posted them here out of wonderment that the trends going down B-Al-Sc-Y appeared to look better than those going down B-Al-Ga-In-Tl.


 * As it turned our there is whole branch of periodic table history on such a configuration spanning 1893 to 1989+, that I did not know anything about, and which is missing from the historical content of our periodic table article.


 * So now we know why we are where we are today. The popular form is an accident of history combined with acceptance by the chemistry community of placing the noble gases into group 0 because of their seeming inertness; La made it into group 3 because it was discovered before Lu, and when it was discovered the f-shell closed at Yb rather than Lu, nothing changed about the chemistry involved, so La stayed where it was; B-Al were moved out of group 3 and into group 13, on electronic grounds.


 * That's what I call a good outcome. Sandbh (talk) 00:47, 20 December 2020 (UTC)
 * , I understand what you did in Excel to get the results. What I am saying is that selecting a model (linear, quadratic, exponential, etc) is making an analytical choice.  It should follow from consideration of which is justified in the circumstances from a theoretical perspective.  Trying various models and choosing the best fit is not good practice in statistical analysis.  Any editor presenting data with trendlines that is not already reported in RS is making a choice that renders the analysis OR whether the choice of model is well-grounded theoretically or not.  The analysis is not trivial in the sense of the calculations that are permitted under policy.  Neither WP:COMMON nor WP:IAR nor any other policy will change the analysis from being OR which is impermissible in article space and unhelpful in supporting the inclusion of the material that you are describing.  My view could be found to be wrong by consensus (not a local consensus) at a well-attended wiki-venue, but I will be astonished if my understanding of policy in this area proves to be in error.  I recognise that you are contributing in good faith and genuinely believe that this type of statistical analysis is not OR.  Unfortunately, I believe that you are incorrect and that it is important that you come to understand why such an analysis falls unquestionably within the WP meaning of the term "OR".  EdChem (talk) 10:28, 20 December 2020 (UTC)


 * , I will agree to disagree with you on what counts as OR. I suspect most editors know what WP:OR was really designed for i.e. unsubstantiated statements added to the article space such as e in fact = mc2.0001. The trend lines under discussion don't represent such a case, they are consistent with known trends and trend lines (such as those in G&E), and they wouldn't have made it into the article space in any event.


 * What is missing in this case is: (a) an understanding of WP:COMMON; (b) an appreciation and application of the higher-level principle/s behind OR rather than black-letter OR law; and (c) a dash of WP:IAR.


 * In any event, as noted, my focus is on the good outcome i.e. the trove of forgotten historical sources dealing with B-Al- over Sc, and Mellor's observation about why B-Al- came to be moved over Ga. --- Sandbh (talk) 11:40, 20 December 2020 (UTC)
 * , I am not sure that this is an "agree to disagree" situation because one of us has to be wrong about our interpretation of OR. I think the input of outsiders is needed here, so that we can all be sure on whether this type of analysis constitutes OR.  I am not suggesting that sanctions are warranted whatever the outcome as the issue is not including the material in article space.  Whether to give it weight in considering the historical approach where B and Al were sometimes placed is worth resolving, IMO.  Thoughts?  EdChem (talk) 23:47, 21 December 2020 (UTC)


 * , I think it is moot since I was not going to draw on it in the article space. B-Al-Sc-Y-La stands on its own given the 17 examples I cited from the literature, including Rydberg, Seaborg, Deming and Pauling; and Mellor's reference to the controversy in that it is was supposedly resolved in favour of B-Al- in group 13. On concerns as to the rigour of the methodology, I used this in my peer reviewed article on the location and composition of Group 3, and it attracted no interest from any of the referees, nor the editor. The reference to sanctions is peculiar. Where did that come from? Feel free to raise the "OR or not?" question elsewhere. Sandbh (talk) 06:43, 22 December 2020 (UTC)


 * , I know you are not planning to put it into article space, but the analysis itself was either OR or not and I think that is worth clarifying. As far as methodological rigour goes, and noting this is a different issue for the literature than WP, all I will say is that it would have attracted comment from me if I was a reviewer.  I have read many papers where the conclusions may be correct but the statistical work supporting them is deficient, so I am disappointed but not surprised when such issues are missed by referees.  As for my mention of sanctions, it was meant to reassure you that my intent is not to post anything like "Sandbh has violated OR and that X, Y, or Z should happen", but rather to seek opinions in hopes of resolving which of our interpretations of policy more closely matches community views.  I would certainly not raise the question at ANI, for example, which is a venue far more focussed on wrongdoing than on understanding.  EdChem (talk) 11:05, 22 December 2020 (UTC)

Periodic table family tree
After the stormy weather, here's some sunshine in the form of a rough family tree I drew to help me untangle things.

There are some missing links in the tree:
 * 1) A date for the IUPAC table—and did they publish a table before then, and what type was it?
 * 2) In the type II column, the first published table with He-Ne, and Lu in group 3. Bury might be it.
 * 3) In the type III column, the first published table with He-Ne, and La in group 3 table
 * 4) There is a sidetrack I’d like to add about the placement of Be-Mg in group 2 or group 12.

Links:
 * ADOMAH
 * Bury
 * Deming 1923
 * Gardner & Mazzucchelli
 * Pauling [1960 version ]
 * Rang
 * Romanoff

--- Sandbh (talk) 01:17, 19 December 2020 (UTC)
 * First Lu table could be Bassett 1892 (though that is also K-Rb-Tm), see Thyssen and Binnemans' Accommodation of Rare Earths in the Periodic Table. Though there could be an earlier one they don't mention, and the tree still seems like OR to me. Double sharp (talk) 04:09, 19 December 2020 (UTC)
 * , Bassett is B-Al-Ga-In-Tl. I looked in Thyssen and Binnemans. They have nothing to say on B-Al-Sc-Y-La, aside from showing a table by Meyer (1918) in which group 3 is B-La-Ac, with the Ln branching off La, and 4 is C-Th, with the Ln finishing under Zr. Groups 1, 2, 15-18 are as we understand them now. Sandbh (talk) 06:09, 23 December 2020 (UTC)

OR how? Sandbh (talk) 12:10, 19 December 2020 (UTC)
 * As I think everyone else will answer you: it's OR because this analysis and classification isn't in any other source. But, as is becoming increasingly apparent from 's informal poll (a brilliant idea on his part, IMHO), it seems that there's a strong divide on this project regarding what counts as OR, how much use of 1ARY sources we should make, plus what counts as DUE: your views on one side, and everybody else's clustered closely with each other, but far away from yours. So I doubt this will sway you.
 * I do have some thoughts on how this situation might be resolved, but I think I would like to wait and see what EdChem and others have to say about this. Double sharp (talk) 12:17, 19 December 2020 (UTC)

I'll fall back to the article link provided earlier: What SYNTH is not.

This article says at the top: "This page in a nutshell: Although avoiding original research is an important part of ensuring that Wikipedia content is verifiable, use some common sense about it… [italics added].

The family tree I drew is a simple record of historical fact. What happened to common sense here, and considering the principle behind WP policies, rather than the black letter law? Not forgetting WP:IAR, if a policy is getting in the way of improving WP. I know what those other policies are for, to stop genuinely original thought being posted in the article space. That's not the situation here.

I could have as easily written the family tree as words. Thus Janet first published the He-Be, Lu form in 1928. Various other authors have supported this form (including Scerri) or its equivalent. Jensen (1983) noted tables with Lu in group 3 first appeared in the 20's and 30's and was a strong supporter of this form. Deming (1923, 1940) popularised the currently popular 18-column form with La in Group 3. Rang (1893), Rydberg (1913), Seaborg (1945) and Pauling (1988) published (B)-Al-Sc-Y-La tables, which are no longer popular.

That's the kind of interesting, value-added, citation supported information I'd expect would be included in an FA periodic table article. --- Sandbh (talk) 23:35, 19 December 2020 (UTC)


 * Early replies: What are the "Family" relations exactly? At first glance, I see outer appearances (look-alike/differces) as criteria. Not structural distictions. For example, 18-col vs. 32-col is irrelevant for any structure OR 1RY published. Next: adding a column (noble gases) or a full block (Seaborg, actinides, ca. 1935) is a major change in structure. Group 3 constitution, and H-placement alike, is relevant but not a new family branch. -DePiep (talk) 23:54, 19 December 2020 (UTC)

I think the family tree, from right to left, shows four lines of descent: (1) chemistry (He-Ne, B-Al-Sc-Y-La); (2) chemistry + electronics (He-Ne, B-Al-Ga-In-Tl, Sc-Y-La-Ac); (3) electronics + chemistry (He-Ne, B-Al-Ga-In-Tl, Sc-Y-Lu-Lr; and (4) electronics (He-Be, B-Al-Ga-In-Tl, Sc-Y-Lu-Lr).

In other words a four-step progression from chemistry to electronics.

The first step looks to have been Rang (1893) with B-Al-Sc-Y-La. Rang was followed by others: Rydberg, Deming, Seaborg, Pauling.

The second step away from a chemistry table appears to have been Romanoff, who showed B-Al over Ga! He was followed by Deming in 1940. I guess Seaborg, with his actinide hypothesis, cemented this branch of the family tree.

The third step away from a chemistry table was running in parallel, with some tables in the 1920s and 30s showing Lu in group 3. This line bloomed with Jensen (1983).

The final step, to a fully electronic table, with He over Be, and Lu in group 3, also ran in parallel: the LST by Charles Janet (1928). At the time he didn't know he had published a fully electronic table but he twigged latter on. And many other authours have expressed their admiration for this form, including ADOMAH, and Scerri, or its equaivalent.

There were some minor things happening around the edges. Rang (1893) looks to have been the first to place H in group 13. The noble gases moved switched back and forth between group 0 and group 18, but either position was of no great consequence. H, as always, moved around, depending on the author. 18-column v the rarely seen 32-column didn't mean much. The table used within IUPAC appeared at some time, as a fence sitter. Sandbh (talk) 03:42, 20 December 2020 (UTC)


 * PS: There is no OR here. I found the smoking gun in the literature talking about the switch from B-Al over Sc to B-Al over Ga. The current popular form has a ton of RS, including the impact of the Seaborg hypothesis. Jensen traced the history of the Lu form. The pure electronic form, starting with the LSPT, has been written upon at length. It's a story worth telling, in encyclopædic form, of course. Sandbh (talk) 03:57, 20 December 2020 (UTC)

Periodic table analogues
A Xmas curio for when things have calmed down.

On the left is a literature extract from 1962, categorising the metals. Items of passing interest:


 * Al over Sc;
 * the split between groups 3 and 4; and
 * the inclusion of Pt with the soft metals.

The nonmetals are on the right, plus extracts from the literature speaking to the analogies between the four metal and four nonmetal categories. The image is mine. An earlier version appeared in my article Organising the metals and nonmetals.

The PT article is currently the poorer for not mentioning something about these analogies, as it used to.Sandbh (talk) 01:08, 25 December 2020 (UTC)

History
I've added a short paragraph to our periodic table article, filling in the historical gap between the popularity of Deming's 1923 18-column table and the modern form of the table (see paragraph 8, "Deming…"). Thoughts? Sandbh (talk) 03:08, 27 December 2020 (UTC)

WP:ANI (26-12-2020)
There is currently a discussion at Administrators' noticeboard/Incidents regarding an issue with which you may have been involved. Thank you.
 * -DePiep (talk) 21:55, 27 December 2020 (UTC)
 * Initial post 26-12-2020; ANI archive full thread: /ANI Archive 1055 (7 Jan 2021) -DePiep (talk) 21:53, 8 January 2021 (UTC)


 * There was a prior WT:ELEM thread about this ANI discussion. The thread included content that belonged at ANI rather than on this page and closing it as non-neutral and off-topic for this page would have been appropriate.  I support 's decision to start this neutrally-worded pointer to the ANI discussion as a replacement.  The original thread has been preserved and can be found in here in Archive 55 of this talk page.  I request that discussion of the ANI thread be kept at ANI.  EdChem (talk) 22:40, 28 December 2020 (UTC)


 * To all: please take note . -DePiep (talk) 01:27, 3 January 2021 (UTC)

ANI thread closed

 * As you are probably all aware, the ANI thread has been closed and Sandbh has been topic banned from "chemical elements, broadly construed" for 6 months. That will certainly prevent his participation at this page for that duration.  Speaking only for myself, this is a sad outcome but one that had become necessary.  EdChem (talk) 02:43, 5 January 2021 (UTC)

Archiving old threads
I've been worried recently about how some sections are archived too quickly (in my opinion, prematurely). The last instance was just the day before yesterday, when a new section on a topic that was already being discussed on this page was created, and the existing discussion was archived away, and the two occurred within the same minute. There was no prior discussion on why a part of the discussion should be archived so quickly when the discussion continues. Could we agree that this is a project page as opposed to a user page and as such, it should be up to the project when a section is archived, not to any particular user upon their instinct to do so? There should either be explicit agreement to archive a page, or the automatic archiving timer must run out.--R8R (talk) 11:15, 28 December 2020 (UTC)
 * The section I archived was an off-ANI discussion of the ANI thread. There is no good reason to support this. Even when considered acceptable, it complicates that discussion, creating a situation of incomplete information. Also, it might be bordering WP:CANVASSING issues. I replaced it with a neutral notice + working link. -DePiep (talk) 11:38, 28 December 2020 (UTC)
 * Stopping a discussion is one thing and hiding it in the archives is another. There was no need or agreement to do the latter just yet. I repeat the point this is a community page, not a personal page; you can manage your talk page without asking anyone in a way you see fit and the same goes for me and anybody else. But anyway, it isn't the only incident when a section is archived early, I had noticed that long ago, and I'd like it that it doesn't happen again without prior consent by the wider community which uses this page.--R8R (talk) 13:52, 28 December 2020 (UTC)

Yes, agree with R8R. In the midst of ANI I have had to look in our archives, for content that is still live. Premature archiving. That said, in general, I welcome housekeeping endeavours. I guess it is a question of getting the balance right. Live, controversial material should not be archived prematurely. Sandbh (talk) 22:04, 28 December 2020 (UTC)


 * After : I see the issues of reorganising / reordering of threads and archiving as aspects of the same issue: well-intentioned management of this page that has also been disorienting / confusing for readers and participants.  I am also uncomfortable with some threads being labelled as "Major" as it is a subjective descriptor and readers / contributors can decide for themselves what is important to them.   and, would you both please be less proactive in this area going forward?  I am not sure that a strict requirement for consensus for early archiving is needed, but I do agree with the sentiment that actions have occurred prematurely at times.  Equally, this page is huge and some unilateral efforts at management are understandable.  Speaking personally, I appreciated 's approach a while ago when a request to move comments that I and others had made preceded action – though it was not required under policy, I felt respected by the request and was grateful to not be taken by surprise when the move was implemented.
 * On the specific ANI-related thread, I was uncomfortable with the initial post as it declared actions as disruptive and so was not a neutral pointer to the ANI discussion. I would have objected except that the comment made was the name of the ANI thread.  Sandbh's subsequent comments were properly located at ANI, which is why I posted requesting the discussion stay at ANI.  DePiep's replacement thread is neutrally worded, as he noted in his edit summary – he sensibly chose to use the ANI-notice template and include a simple link to the ANI thread, in line with the recommendations for avoiding canvassing and I support his action in creating the new thread.  Now, I recognise that it may be argued that closing the section was preferable to archiving it, but I don't believe that DePiep's intent was to hide the original thread.  DePiep closing the thread was open to criticism of involvement given the ANI thread calls him disruptive and I would have objected if he had replaced a non-neutral thread pointing to the ANI discussion of him with a non-neutral thread point to the ANI discussion of Sandbh – but that's precisely what he did not do.  R8R, if you want to unarchive the thread and then close it as non-neutral,  I won't object... but I question if it is really necessary in the circumstances when the thread in question should not be continued?
 * I do agree with R8R that there have been actions in page management that have been sub-optimal, and thank you to R8R for bringing this up for discussion. I too would like to have fewer surprises in coming to WT:ELEM.  Can we come to a broad consensus on this process issue?  Thank you.  EdChem (talk) 22:31, 28 December 2020 (UTC)
 * My reasoning for archiving the "ANI-notice" discussion thread, as I wrote, is completely different from archiving other threads here. It was problematic, and I'd expect responders here to weigh that. -DePiep (talk) 23:13, 28 December 2020 (UTC)
 * I visit this WikiProjcet maybe once a week and normally just watch diffs (and hopes all the colour/groups discussions are gone :-). But I'm not in favour of all the archiving - it makes it more difficult to follow the project (from the sideline). Christian75 (talk) 23:45, 28 December 2020 (UTC)


 * I support housekeeping and archiving by anybody, including DePiep. I need no notice; I find it easy enough to work out what happened. IGF I posted the thread (that DePiep archived) as a courtesy to let others know what was going on. IMO, archiving that particular thread, without prior discussion was inappropriate.


 * I won’t mind posting a notice of my intention to engage in major or possibly contentious housekeeping. But not minor housekeeping, which everybody should be doing in order to keep our page coherent. I agree with DePiep as to his sentiments about disorderly threads, hence my support for semi-regular housekeeping. Sandbh (talk) 00:37, 29 December 2020 (UTC)


 * Of course, archiving is common practice when done by bot (&gt;60d), and is not a problem when done manually for obviously finished discussions. The latter I only did (manually i.e. by clicking the botlink by section) when & because this page is huge and discussions are improductively long/intertwined. Even then, I aimed to only archive threads that appeared stale (won't call this 'houekeeping'). This said, I am surprised that you find it is this bot-like archiving that makes it "difficult to follow the project". If you check by once/week, aren't the huge numbers & sizes of edits, the TOC reorganisations, and the interacting threads way more complicating ;-) ? -DePiep (talk) 21:21, 29 December 2020 (UTC)
 * EdChem: I've removed the Major labels and retained the bolding, so that it is easier to find the sujets du jour, so to speak. Sandbh (talk) 23:13, 29 December 2020 (UTC)
 * Sandbh, thanks for removing the "Major" headings. I have moved your reply as it was inserted into the middle of my post and such insertions are (at least) strongly discouraged under the TPG.  EdChem (talk) 00:08, 30 December 2020 (UTC)
 * Ah, the bolding is kept -- I noticed this in the TOC first. Also applicable in subthreads. But hey, what would be the criterium? "du jour" may be too subtle/sensible/subjective. And better no threads-talks per sub/section I prefer. How to use this option? -DePiep (talk) 01:41, 30 December 2020 (UTC)


 * After having read the YBG clarification of their Wikibreak, I understand that refactoring this talkpage (I did some myself) is much more problematic than I perceived. So I will stop doing so without proposing & agreement. Some management might be needed though, as the talks can be long and interweaved. From the same background, I suggest is selfrestrictive too in this matter. -DePiep (talk) 13:26, 2 January 2021 (UTC)

On creating a compromise colouring
There's something Andrew D. said at the ArbCom case page that I think makes a lot of sense, too: "perfect is the enemy of good".

Now, I've stated a way above in which one can think of the current colour scheme as policy-compliant. It reflects the situation in the sources adequately in the sense that most sources show some kind of colouring, and that it is usually not too far from this one. It seems to me that as long as this variation exists in the sources one cannot do better, but one is not doing too bad.

Now, I'll also note that we've had almost this colour scheme since literally 2002. And it served us pretty well up to 2013: the categories there would correspond basically to

alkali metal | alkaline earth metal | lanthanide | actinide | transition metal | post-transition metal | metalloid | other nonmetal | halogen | noble gas

So let's ask: what's changed since then? Not very much, honestly.
 * Group 3. Yeah, let's not touch that right now. It is one of the things that changed, let's note that and move on. On moratorium until we hear about the IUPAC project progress, if you will.
 * Group 12 moved from transition metals to post-transition metals. Well...the 2017 RFC was basically a snow close, so no need to revisit that: let's leave well enough alone.
 * Elements beyond Hs. Nothing to say here. They were coloured in at first, but then it was pointed out that there had been no chemical studies, so they were uncoloured. Cn and Fl got chemical studies that at first looked conclusive but then turned out not to be so, so they got coloured in and then uncoloured back again. That seems to be about how it should work.
 * Polonium. Well, no one seems to have disputed it since I brought it up back in 2012. Most people who focus on it seem to agree it is a metal, so there it went.

Then there is the one element that caused all the trouble.


 * Astatine. Problem is, this is where halogens leave the nonmetals. Like I said in 2012, I would agree with colouring Po as a post-transition metal and At as a metalloid, but what do we do with the halogens category? And thus began the quest to get a better category name than "other nonmetals", deal with astatine effectively, and try to have something similar.

I think that, looking at Sandbh's RFC with so many plausible options, as a contrast to that linked 2017 RFC that dealt with group 12, this may be a case of "perfect being the enemy of good". It seems to me that the literature generally agrees that if a PT is coloured, halogens should be coloured, but it doesn't seem to have any agreement on what to call the other nonmetals. That would certainly explain why Sandbh was able to marshal some level of support for multiple options, but I don't think there is total dominance for any one of them.

So I'd like to turn our eyes back to 2002 and think: suppose at that point, just having built a good enough PT, we had just heard that At turned out from early chemical studies and modern theoretical corroboration to be a metal. What would we do, now that it's not clear in the literature even whether halogens extend past iodine?

I think there's a relatively simple solution. Just replace "halogen" with "halogen nonmetal" as has been suggested above. Then that actually manages to please all sides. If you agree with the part of the literature that thinks a halogen has to be a nonmetal: then it's just a pleonasm, but not wrong. If you agree with the part of the literature that thinks a halogen is anything in group 17: then it's actually necessary to be right. So both parties will agree it is right, and the general idea that F-Cl-Br-I are distinct enough to deserve being a category is there. For astatine, we invoke WP:CONTEXTMATTERS and say: OK, well, it's an element most people will never encounter, so let's look at the few sources that cared about it. In fact, that's going to create exactly Sandbh's 3-category solution, so I guess he may be happy. Although he may be less happy about what I propose to call the new-old category. ^_^

Sure, it's not "perfect". But it seems to me that many of us are actually looking favourably at restoring halogens just because that's what most sources have. I worry that in Sandbh's proposed RFC, because there are so many possible options for the remaining nonmetals, none will get a clear consensus. I also worry that the fact that there are so many possible options rather strongly suggests that the literature isn't unified and that it will scotch any possible consensus (in fact, I still can't bring myself to really strongly support any option that isn't "other nonmetals"). So I'd ask: why let the perfect get in the way of the good? Let's just do a small improvement that keeps things still OK, and it'll be better than nothing. If we try to go for the gold from the beginning, we'll get nothing.

Therefore I would like to ask: could everyone accept this compromise? It doesn't matter if you think it's not the best we could do, just whether you think it could be better than what we have now. I'd rather we have one modest improvement than try for a huge and great improvement that ultimately is too radical to pass. This way, we at least get halogens back on the table like they usually are, even if we don't have the name for the other category.

I edited this from a 2011 revision; obviously it'd be updated to current styles. We can change the colour scheme later, but that's a different topic. Thank you User:DePiep and User:R8R for your work there. ^_^

In that spirit, I would like to suggest that if this is acceptable, we "lock in" the categorisation and the layout for a while. No sense in arguing about it again when there's an entire article to improve and any further improvements run the risk of getting no consensus. That's inspired by what User:Jehochman said at User:EdChem's talk page: Can we have this discussion once, come to a decision, and then not have to revisit it too often? Obviously only for the first 108 elements (we need to be able to reflect scientific advances if elements 109 and up get chemical studies of course).

If need be, the name of the category "other nonmetals" can be revisited later, and that particular thing need not be locked in. But remember, the change from group 12 to leave the transition metals was tried twice. It failed the first time because then the category name was "poor metals" and its situation in the literature was more iffy than "post-transition metals". The second time, the new category name was "post-transition metals", and it resulted in a WP:SNOW success. Basically: if generally there did seem to be agreement between Sandbh and me at least in October IIRC that splitting out the halogens was justified (and that maybe counts for something because we generally post here more stuff than anyone else about all this categorisation stuff), I would prefer that that less controversial step not be jeopardised by mixing it with something more controversial.

Do we need disclaimers? I guess on the main periodic table article we do, and for the infoboxes we probably do need them just as we do for something like or. But the literature situation is inconsistent, so I think more than that is the perfect being the enemy of the good again. So I'd support this compromise in the form of just changing all the PT images and templates to this, standard style everywhere, contract to 18-column consistently with small asterisks, etc. etc. etc.

No, this is not my personal favourite ideal colouring across the whole spectrum of colourings, but I think that probably speaks well for it being a good compromise: no one is completely satisfied, but people should feel it's better than nothing. And within the constraint of being close to the centre of what sources really give, I cannot think of any way to do much better, which makes me think that trying to better it is trying too hard.

Why do I think this might be a plausible compromise? Because such a split of the nonmetals is something Sandbh has proposed, so it is presumably not going to cause an argument; because the name "other nonmetals" is one of the options Sandbh mentions in his RFC, so it is presumably not completely unacceptable to him; because it is my preferred option from his RFC and seems the most obviously non-controversial one; because IIRC most supported some sort of restoration of halogens; and because it is based on something that lasted for years uncontroversially before new information about astatine threw a spanner in the works. But most of all; because I think that these reasons give it a chance to stop the fight completely, as something like this was around for over a decade and no fighting resulted, so I feel like tweaking it the minimal amount possible now to correctly reflect new information about astatine is a plausible way to go. But we will see; the more important thing is to have some compromise stop the fight, rather than for it to be this one specifically. Anything that can solve this issue swiftly, take our minds completely off it, and let us work together on actually improving the parts of the periodic table article without controversy would be a good thing in my eyes even if it doesn't end up being this. Double sharp (talk) 20:23, 13 November 2020 (UTC)

Could we perhaps agree to freeze this issue for a while too? It seems to me that so far, we have a couple of views on the current colouring, but it seems to have worked fine since 2018, and I don't think anyone's current idea is that radically far from it. So, I'd like to suggest we stop talking about it at least temporarily and get to work on the periodic table article itself as EdChem is suggesting to us. We may perhaps return to it when the memory of the difficult previous situation is not so close. Double sharp (talk) 22:30, 14 November 2020 (UTC)
 * Yes please. Our today's discussion flow is not effective. (Is there something about 2020?). -DePiep (talk) 23:33, 14 November 2020 (UTC)
 * Although if wishes to go ahead with his proposed nonmetal-splitting RFC, I think it may not be too much of a problem provided we just !vote, explain ourselves, and don't argue with each other. Double sharp (talk) 00:39, 15 November 2020 (UTC)

Category colors
Here's a table comparing contrast ratios between the colors we've used, plus black: Symbols are readable against all the colors used, though the Fr-alkalis are markedly less so, but none of the colors are AA-compliant with each other. In particular, only 3 pairs exceed a contrast ratio of 2.000 (Fr-alkalis against nonmetals, halogens, or noble gases). The Li-alkali do not contrast as well as the Fr-alkalis against the other colors, especially for visually impaired readers, and are especially hard to distinguish from the actinides. Aditionally, the transitions contrast poorly against the lanthanides and the post-transitions; the post-transitions are very hard to distinguish from the metalloids; and the most similar pair was nonmetals vs. unknown. The most important colors to distinguish are the ones that appear adjacent on the periodic table, such as the transitions and the post-transitions; these pairs are bolded in the table above. The nonmetals could be darkened a bit, but we'll have to come up with a completely new color scheme for the metals. –LaundryPizza03 ( d c̄ ) 12:54, 5 December 2020 (UTC)
 * Legend
 * : Black (#000000)
 * : wikilink blue font (#0645AD)
 * : Alkali metals, dark version (#ff6666)
 * : Alkali metals, light version (#ff9d9d)
 * : Alkaline earth metals (#ffdead)
 * : Lanthanides (#ffbfff)
 * : Actinides (#ff99cc)
 * : Transition metals (#ffc0c0)
 * : Other metals (#cccccc)
 * : Metalloids (#cccc99)
 * : Other nonmetals (#a0ffa0)
 * : Halogens (#ffff99)
 * : Noble gases (#c0ffff)
 * : Superactinides, as used in Extended periodic table (#d1ddff)
 * : Unknown properties (#e8e8e8)
 * I agree. I understand and  have made some progress on this front (see e.g. User:R8R/PT search for colors). Double sharp (talk) 16:06, 5 December 2020 (UTC)


 * First quick remarks: the Color contrast ratio applies to fontcolor vs background color, not neighbouring bg colors. So the red bg's in the table are not that meaningfull. That said, there is an other reason to strive for between-categories-color-distinction: the Reader being able to recognise category bg color from PT-cell-to-legend ('which category is this cell?' and v.v. ('which cells are metalloids?'). With this, as a third distinction, backgrounds should not look alike (confusion as noted: 'is this AM or PTM?'). -DePiep (talk) 22:06, 5 December 2020 (UTC)
 * I have added row #2, "wl": the wikilink blue font color (as in Main page). Its contrast with background is meaningfull. Because: wl-blue is a fontcolor, replacing black fontcolor. -DePiep (talk) 22:20, 5 December 2020 (UTC)
 * Why checking black as a bg color, and white fontcolor? All your contrast checks are black-font on color-bg, right? (Is why I added wl-blue this way). -DePiep (talk) 22:31, 5 December 2020 (UTC)
 * , thanks for your contribution on colours that is looking from the perspective of accessibility / readability. I agree that this is an important consideration and is one that I feel needs advice from those with expertise in the area.  I know that  has been thinking about these issues as well. I can only speak for myself (though I suspect I am not alone) in agreeing with the importance of making a choice of colouring scheme that supports visually-impaired readers in principle but having little knowledge of how this is evaluated in practice.  Looking at the above table, for example, I get that larger contrast ratios are desirable for readability and I infer that a goal of greater than 2 is an appropriate guideline, but I don't know how DePiep's comments fit.  Is this an area where those lacking expertise (like me) should just stand back, or is there some material that I can read to help make a useful contribution?  Are there contributions that would be useful from a chemical perspective?  For example, is it better for metal categories to be based in a related group of colours (say, shades from purple / pink / red / orange) and then different groups for non-metals (blue / green) and metalloids (yellow)?  Or, is it better for the metal colours to be unconnected (say, green, yellow, red, brown) and then blue metalloids and pink and orange non-metals (just picking different colours, many alternatives possible)?  Alternatively, is there a WikiProject or group that deals with accessibility / readability and colours that could offer help?  EdChem (talk) 23:56, 5 December 2020 (UTC)
 * The table is not exactly to the point.. W3C contrast checks are about font-vs-background colors, not neighbouring bg colors. (There is no AEM-bg vs LN-bg check needed anywhere). The table should be removed. Except for row 1 and my wl-blue addition. -DePiep (talk) 00:01, 6 December 2020 (UTC)
 * in other words: the table does not address WP:ACCESSABILITY. I also wrote in my 1st reply above. -DePiep (talk) 00:08, 6 December 2020 (UTC)
 * Thanks for the explanation. H:Colorblind suggests using only 5 colours (including black and white) as backgrounds... would that approach work, picking a background (blue, say) for the non-metals and use stripes (horizontal, vertical, and none), then another color for metalloids, then two colours for metals with stripes (including diagonal if needed)?  I imagine it would be messier than just with differing colours, but potentially more accessible – or even link to such a PT as an alternative for the colour one for those struggling with readability?  EdChem (talk) 00:41, 6 December 2020 (UTC)
 * re, short: we must provide ten (10) category colors. Difficult, if even possible (too manymuch restrictions). Only good news: it is a PT-pattern, so a l-to-r flow. Also we can gain by dropping font-colors for phase (red H, green Hg).
 * And: this table is not the optimal guide, as noted.. -DePiep (talk) 01:21, 6 December 2020 (UTC)
 * As for coloring the categories, EdChem, we are sailing between Scylla and Charybdis. Satisfying all contrast and CB (colorblind) needs, will give us bleak (not even pastel) colors. That are failing the "recognise a category" check.


 * Some self-redactions into more helpful wording. -DePiep (talk) 17:43, 6 December 2020 (UTC)

I still think the post-transition metals and metalloids are too similar in color. And if the real issue is the colors not contrasting well enough against the wikilinks, then I can accept that and come up with something more pastel than current. As the alkali metals have the same hue as the transition metals, I would also change the latter's color to purple, especially since the darkest AAA-compliant red is already very light-colored: Here's what the navbox would look like:

–LaundryPizza03 ( d c̄ ) 06:31, 28 December 2020 (UTC)

two comments on this:

first, we have established not too long ago (I may look for the specific discussion later if you're interested) that we would recolor the existing category set, and that there would be a contest, for which everyone could send their commissions. Changing the coloring scheme now outside of the agreed path forward would not be a good thing for the project. I'll initiate it in the coming days, and you'll be welcome to join it.

second, I saw your colors on a tablet first, and then on a big screen. In the latter case, they were more or less distinguishable; in the former, they were harder to distinguish than the existing colors (particularly the Ln-An-TM pinks).--R8R (talk) 11:23, 28 December 2020 (UTC)

Ready for archive

 * This topic has been explored more in /PTG/categories.
 * Then, by the change from Category into Block coloring, 2021-01-17, the topic has become moot.
 * Also, any further reseach & development can be done in the new task force PT graphics.


 * For this, I propose to Archive this thread shortly. -DePiep (talk) 20:12, 23 January 2021 (UTC)