Wikipedia talk:WikiProject Elements/Archive 53

This is an archive of past discussions about Wikipedia:WikiProject Elements. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 50

Archive 51

Archive 52

→

New periodic table: implementation

I have a proposal about what we're going to do once we've established what changes, if any, we want to see in our future periodic table. So far this includes the group 3 question and the question about categories. I'd like us to implement those changes, if they are approved, simultaneously, and on top of that, there is one more pending change which we could also implement at the same time.

We have established that the current color scheme has potential for improvement. I propose to change the coloring scheme once we have figured out what categories it should have and what their composition would be.

As soon as we have agreed what categories the periodic table should have and what their composition would be, we should not change the periodic table instantly. Instead, we note that a conclusion has been reached and we start a sufficiently long period of time, say, two months, in which all editors who would like to propose a new coloring scheme for the periodic table can craft one. When the period ends, we discuss the submitted schemes and choose the best coloring option. Once we have chosen the next coloring scheme, all changes we have agreed upon go live.

I submit this idea to the project and I would like to hear what other editors think.

@Double sharp, Sandbh, DePiep, YBG, ComplexRational, and Droog Andrey: comments are welcome. Sorry if I forgot anyone.--R8R (talk) 14:46, 18 October 2020 (UTC)

@R8R: I'm not sure I understand you correctly. Do you mean that even after the group 3 and category things are decided, we should wait to decide on the colouring scheme? Personally I think that once we have a consensus on the group 3 options and categories we should let them go live first and then think about changing the colours. Otherwise it gets into the situation again where we have a strong consensus on the Project but things never get anywhere. I found myself, in fact, very frustrated by that situation when we reached the point in the earlier group 3 discussion that we had a massive swing towards changing back to Sc-Y-Lu, but IIRC you told me that we needed an WP:RFC despite that massive consensus here because the previous time it was decided in an RFC. Personally I think that we should be able to solve this as a project because these articles are of interest first and foremost to us among editors; the biggest participants at that RFC were from this project. Changing the colouring scheme requires no less discussion than changing the group 3 thing, so why should one need an RFC and the other not? I found it myself somewhat muzzling along the lines of "we have a consensus, why can't we implement it"? Now I think it was correct in hindsight because that consensus was not so policy-based (it was more about the science argument than the sources one), but if we get the same level of support again, I see no reason why we could not just change it and act on our consensus.

In fact, if the group 3 thing gets a consensus before categories, I would support making the group 3 change live before talking about categories to keep us discussing one option at a time.

So I'd rather put it in the following order: put the questions like this:

Do each of us support / oppose the change back to Sc-Y-Lu? (Binary choice. Whatever is decided, the other option must be footnoted, I think.)
- If Q1 passes, implement general change back to Sc-Y-Lu with the footnotes; if it fails, just add the footnotes to our current Sc-Y-La arrangement. Then pass to Q2.
Do each of us support / oppose wiping out the category colouring in favour of blocks only? (Binary choice.)
- If Q2 passes, implement and finish (we have previously discussed the block colours and I don't think there's anything else that needs to be argued about it). If Q2 fails, proceed to Q3.
What changes if any should be made to the current category set?
- If any change gets a consensus, implement it and proceed to Q4. If not, retain status quo and proceed to Q4. Some sort of time limit should probably be present to prevent things from getting out of hand, like you suggest for the colourings.
What colour changes if any should be made now that the category set is decided?
- If any change gets a consensus, implement it and finish. If not, retain status quo and finish.

I think the bold "implements" are necessary to make sure that once a consensus is reached, it gets acted upon. But what do you think? Double sharp (talk) 15:00, 18 October 2020 (UTC)

I converted my immediately prior thread to become the consensus-gauging thread for Q1. So you and everyone else can !vote there. I kept the question neutral, I left my opinion to my !vote. Hopefully to remain open until all major participants of this project and this discussion have !voted. Double sharp (talk) 15:23, 18 October 2020 (UTC)

What I mean first and foremost is that we should agree on the categories and we use and that we should use the opportunity to change the colors since I recall there is an agreement the color scheme we have today is not the best scheme we could have. So when we roll out a new category set for the public to see, it would be great to introduce new color at the same time.

Many outside people watch our articles from time to time (I used to be a person like that before I joined, for instance) and it seems to me that it would be wiser not to have them see so many consecutive changes, but rather lay them all down at once. That is why it seems to me it'd be a good idea to implement all changes at the same time. In principle, the group 3 question could be decoupled from rediscussion of categories and recoloring them but I feel we could do a better by not doing that. This is, of course, up for debate.

The concern that things can end up not done is understandable. The way I propose to address it is to enforce a strong time limitation. Two months seems like a plenty of time if you want to submit anything. Whatever is submitted in those two months is what we choose from and there's that. If there is a preferable option, it should be obvious to anyone. If there is no agreement on what the preferred option is, we could do anything. We could find an uninvolved editor and ask for their opinion as a tiebreaker. It's going to take a little time but the fear of having the project stumble otherwise should be a stimulus enough for us to solve it collegially.

I also feel that the current coloring scheme has outlasted its welcome, so to speak, and the process I propose seems to be the strongest guaranty that we actually abandon it in the end. So any move from it is likely a good move, and the desire not to be the one to take the blame for not moving is, I believe, a good stimulus to get anywhere. However, we could, for instance, use EdChem's help if we don't manage to sort it out ourselves. We could even end up tossing a coin if we're stuck as far as I am concerned.--R8R (talk) 15:32, 18 October 2020 (UTC)

P.S. I am thinking about your group 3 proposal, not just ignoring it. I'll write something in a few days.

@R8R: Please, take your time. As long as it looks like something is going forward, I'm happy. I just want to see a process, among us project members, actually get to completion and result in a decision based on the consensus or lack thereof. And that's why I'm generally in favour of keeping each thing separate so that different issues are not conflated. If outsiders see things changing slowly, then, well, it's not much different from seeing entire articles change slowly when we rewrite them. ^_^ Double sharp (talk) 15:44, 18 October 2020 (UTC)

@R8R: (ec) (I like the current ping-all habit, since there are many long threads, evolving, to follow).

re "one more pending change which we could also implement at the same time". I don't think a 'while we are at it, let's change [implement] it-2 too' is useful. Instead, I am fine with: once a conclusion is explicitly reached, we can consistently roll it out however much and complicated edits are needed. Meanwhile, any other change can be fleshed out and reach a sound conclusion at its own time. Implementations are not simplified by mixing them -- and isn't that true for discussion equally? I want to call: first of all keep separate topics separated. "Separate" as in: scientifically and logically not connected. (For example: redesign & rewriting Periodic table#Overview can be kept separate from current Group 3 issues).

re improving the category colors (is not categorisation itself we understand): yes after the cat scheme has concluded, that is a challenge to solve (the advantage being: when the number of categories reduces, the requirements re accessability and legend-requirements from-legend-to-chart-or-back can more easily be met; with 7 or 8 or 9 colors this is a huge issue!).

However, your 'period of proposals' setup I don't think effective. First I'd like to establish the requirements: wrt accessability like contrasts and wrt legend like 'can one find the graph color in the legend?, and v.v.?'. (preliminary, for example, I think of getting rid of legend fontcolors because that adds reduces contrast options, so only use black and wl-blue text). -DePiep (talk) 15:33, 18 October 2020 (UTC)

Having read the Double sharp reply (after ec), I think can I agree in this detail. I seriously think we should not tie ourselves to voluntary deadlines. Current colorscheme can serve a new cat-scheme. -DePiep (talk) 15:38, 18 October 2020 (UTC)

@DePiep: I understand about the deadlines. I am willing to compromise and not have them. But I seriously, at this point, want something to get done. I want a process among project members to get to completion and result in a decision based on the consensus or lack thereof. Whatever can make this goal happen gets my support. Whatever the decision is, I want some sort of resolution. A consensus and a decision to implement it. Or no consensus and a decision to do nothing and wait for another year or two. If separate topics being kept separate helps that, which I think it does – then it has my support. ^_^ Double sharp (talk) 15:44, 18 October 2020 (UTC)

@R8R: I understand, including impatience creeping into this. I myself am unhappy all along with the grey and the brown, and the unevenness. Core of my approach would be: let's talk to agree about the requirements first (like contrast, other colors used, and good legend design). Could start right now btw, were it not for other time consuming talks here.

If we do not look critically at the requirements, we are sure to end up with bleak colors: good re contrast, bad re legend functioning. That is: an illustration, not a clarification. Enlarging degrees of freedom is crucial IMO. -DePiep (talk) 16:03, 18 October 2020 (UTC)

DePiep, you see, the last time we discussed a change to a coloring scheme in 2016, a proposal had enough support. We took our time then to see if we could craft a better version and four years after that, we still use the old scheme. That's why this time, I am calling for a time limit. I think there is no real problem if we implement one scheme and then, one month later, we come up with a better scheme. Making two changes is better than making zero changes awaiting for the perfect first one.

If two months is not enough, it can be three months or four months. That is negotiable. The point is, if it is four months, it is four months, because otherwise it risks to be four years and counting just like it is today. We should learn from our mistakes.

To be clear: whether or not we should feature alkali metals is not a scientific decision. It's an editorial decision.

Any timing for recoloring should start only when there is an established agreement on what a new table should display. That much is clear and I hoped I made that abundantly clear in my starting message. I don't expect there to be any precise coloring requirements like contrasts, though I believe, that whether or not there should be new borders or color for states of matter will be a part of it. I asked the project to abandon the state of matter colors a while ago but it was decided they should be kept. Again, everything is up for discussion.

We can, in fact, discuss the requirements right now, so that we're all set when we have decided what categories we should use.--R8R (talk) 16:09, 18 October 2020 (UTC)

@R8R: I agree that we can discuss requirements right now, but also think that we should also leave off more than general principles to when we decide what categories we use. Current scheme, tweaked current scheme, blocks alone, or whatever. Let's decide everything individually to keep things clear. ^_^ Double sharp (talk) 16:19, 18 October 2020 (UTC)

Sad to read that a 2016 discussion is causing distress now. To be clear: "2016" was not derailed or frustrated, it just did not work out well, because it was not viable. Not because I checked it but because from the same reasoning I can redo from memory: 2016 resulted in very flaw bleak colors that did not serve good-legend requirements. Full stop. -DePiep (talk) 21:50, 18 October 2020 (UTC)

The problem here is not the fact we have not adopted the 2016 colors. I think that however you look at it, the 2016 colors were better than what we had back then and what we have now, but it could very well be that even better colors could emerge. The problem is that there was a discussion that even better colors could be made, but we never saw them. I remember having high expectations myself but they were never met; there never was even a presented attempt that could try to meet them. That is the problem that I am aiming to solve.

Of course, the idea of a time limit is a double-edge sword, and it could even backfire if I fail to meet the deadline. If I fail to meet it, there's nobody to blame for that but myself. Fully cognizant of such a possibility, I still believe that a time limit is the way forward. Everybody has the same starting position.--R8R (talk) 17:30, 19 October 2020 (UTC)

(outdent)

I agree with this plan of #Progressive implementation. I note (1) I think the ELEM editors would be encouraged and energized by seeing our completed work go live. (2) IMO, this overrides concerns re successive waves of changes impacting those who copy us. (3) It may be helpful to take a day or two to list changes needed before making them, especially in template space. (4) This system generally follows the principles of separation of concerns. (5) We should start a discussion of color selection principles long before we need them. YBG (talk) 06:28, 19 October 2020 (UTC)

@YBG: Thank you! Double sharp (talk) 10:53, 19 October 2020 (UTC)

Since there has been some discussion on the exact form of the footnote, we may add a Q11⁄2 between Q1 and Q2 regarding how the footnote should read. Double sharp (talk) 20:01, 21 October 2020 (UTC)

Coloring principles

Here are the requirements that I can discern; everyone is welcome to add more ideas.

Bare minimum

The following principles must be fulfilled for a legitimate attempt to recolor the periodic table:

Each category has a separate color that is not used for anything else;
There are five text colors: black, unclicked wikilink, clicked wikilink, liquid, gas. Of these, only the last two are variable. The five colors must be sufficiently different from one another to allow to tell them apart easily.

More conditions

The following conditions do not constitute a barrier for proposed scheme, but it is nonetheless highly desirable to fulfill them:

Each pair of a category color and a font color used in the periodic table should have a contrast greater than or equal to 4.5;
Different types of borders should have colors that are easy to tell apart from both one another and the category colors used for cells that use this type of border.
Different category colors should be easy to tell apart for users with protanopia, deuteranopia, or tritanopia, and all pairs of a font color and a background color that can be found in the periodic table should be easy to tell apart.

Discussion

The only idea that I am not certain about is whether we want to have colors for predicted elements or not. We currently don't use them, but we might want to rethink on that.--R8R (talk) 17:30, 19 October 2020 (UTC)

@R8R: Let's maybe think about that last one once we've gotten to the phase of discussing whether or not we should have colour categories in the first place – because if anything my concerns about WP:OR apply even more strongly to predicted elements. Other than that, seems quite reasonable to me. ^_^ Double sharp (talk) 17:39, 19 October 2020 (UTC)

@Double sharp: once we've gotten to the phase of discussing whether or not we should have colour categories in the first place – sorry in advance if I missed something obvious (I haven't had time to read the whole thread), but I don't understand this: why would we not have color categories? I understand the logic behind perhaps not having predicted colors, but almost every PT I've seen has color categories. If I missed this, would you kindly point me to it or offer a condensed version?

@R8R: Should we follow the same principles we discussed during an email thread pertaining to the color scheme of the SHE decay modes chart—namely attention to contrast of adjacent categories and colorblind-friendly combinations? In that case, we also should be careful about how many font colors we introduce, but specifics can come later. ComplexRational (talk) 21:01, 19 October 2020 (UTC)

@ComplexRational: I have a draft proposal ready to explain just why I don't think it's wise to have colour categories, but I wanted to keep the issues discussed one at a time, so I have not posted it yet. The brief reason is that although it is common for sources to give categories, the literature has no agreement on (1) what categories to use, (2) what names to use for those categories, and (3) what elements are included in each category. In other words: many PTs have colour categories (not all though, see Greenwood and Earnshaw), but exactly how they divide the elements into which colour categories is not something you will find any agreement on. That is why I am of the view that we are implicitly taking a side by colouring everything in as if it was all agreed and that that is against WP:NPOV and more specifically WP:DUE. And that is why I have been advocating scrapping the colouring of categories in favour of discussing element categorisation in article text, and colouring only blocks, because (1) that's not unknown in the literature as many Russian tables do it, and (2) that is something that apart from the group 3 issue people actually agree on. You can find disagreement in the literature about whether arsenic (or quite a few others in the p block really) is a metal, nonmetal, or metalloid, but you will find no disagreement that it is a p block element. But the full proposal and rationale will wait a bit I think to not muddle issues. It will appear once everyone has commented on the La vs Lu default issue and we put into action the result. Double sharp (talk) 21:12, 19 October 2020 (UTC)

@Double sharp, ComplexRational, and R8R: I think we can go full steam ahead here without regard to the result of any other discussion. Even in the most radical change envisioned, we would still have color categories, although there would only be four of them: s, p, d, and f. We should still apply the same color principles in that case. YBG (talk) 02:58, 20 October 2020 (UTC)

@Complex Rational: sorry for a late response. More or less, yes. This one is going to be more complicated, however, because there are more color combinations to consider. If you want me to, I could describe it to you later in greater detail, at least what I see. There shouldn't be considerations like "this category should be green," but for example, one thing that is rather obvious is that there is a number of gaseous elements in the top-right corner of the table, and the color of those categories shouldn't look too much like the color of the gas font color. Solid elements are present across the whole table, so it's no wonder the color for solid is likely to be black (not reddish, greenish, or blueish).--R8R (talk) 11:26, 4 November 2020 (UTC)

@Double sharp, ComplexRational, R8R, and YBG: re we could run ahead, save for time reasons right away. And save for: this is not a one-dimensional problem. A redesign should have a strong base (including science communication and webdesign, not strongly available in us, habituals). Also, other discussions on this page these months take loads of time to follow, evolve and to disentangle.

For starters, we need the User Requirements for all PT colors (and graphics alltogether), before flying head on into 'solutions'. Anyway, there is no deadline so we have time to build this part properly and fruitfull.-DePiep (talk) 20:15, 21 October 2020 (UTC)

@DePiep: well, since we're discussing criteria, it will be nice if you tell us what is missing. Because otherwise, it reads like this: we have to stay put for now until we formulate better criteria, I know there could be better criteria, but I will not expand upon this thought.

What kind of science communication are we talking about? You previously argued that there should be no dependance on cultural issues in a coloring scheme. Have you changed your position since then or are the two statements reconcilable; if so, how?

What is User Requirements? Googling does not return a satisfactory answer, so I'll gladly listen to what I may not know.

Other discussions on this page do not directly affect this one. The only thing that matters is the outcome of the group 3, nonmetal, and block discussions, for which we are waiting. The rest does not have anything to do with this one, so they are thankfully not a factor. If some other outcome pops up, it will matter too, but not before it does.

There is indeed no deadline per se, Wikipedia is always work in progress. Yet it doesn't mean we can't do anything now. If it indeed takes loads of time to create a perfect table, it doesn't mean we can't do anything in the meantime; if it did, that would be dawdling. Again, my end goal is replace the present coloring scheme for something better, and I hope you share that goal, so that you can help us formulate criteria for what is better and perhaps even create the scheme we'll end up picking. If you can't for whatever reasons devout time now to creating a new scheme, that's fine, too; after all, some scheme is passed before you can join in on the fun, you can always ask to consider your scheme, too, once you present one, and there is (and should be) no way of stopping you from doing that. No scheme will ever be final.

However, if you don't present one, will not aid the discussion of formulating criteria for a future scheme, and will ask that no change is made even though you previously agreed the current scheme was not particularly good, other editors will likely think this is counterproductive behavior, and will likely act upon it. You expressed interest in creating a new scheme in 2016 and said it was in your to do list in 2017; however, it's 2020 now and despite all the "time to build this part properly and fruitfull," we haven't been presented a scheme yet. I do want you to help us create a better scheme, possibly even create the one we'll go with; by all means, please go ahead. However, if, all things considered, you can't aid that, then I'll ask you to at least not obstruct creation of such a scheme by other editors, and I expect understanding in asking for that from other editors.(retracting this comment since this worry turned out to be unfounded)--R8R (talk) 14:29, 25 October 2020 (UTC)

@R8R:. First reply: I don't like the tone and approach of this post. It is trying to tie me to quotes (old, and out-of-context), edits not made (really, why did you not restart the 2016 discussion? Why do you keep suggesting that it was frustrated instead of not well-fleshed-out?). And no, contrary to what you say it's not affected by just one discussion here. By now, all discussions on this are contaminated with ANI-reports. See, before I can respond in content, I have to weed out aspersions and derailed discussions. -DePiep (talk) 17:12, 25 October 2020 (UTC)

I admit that I did think, when I first read R8R's comment, "oh no, this is going to start things again", because of the last sentence. I would probably have felt a bit annoyed if it were directed at me. It seems to me that User:DePiep is perfectly willing to start now; he just seems to want user requirements to be fleshed out before a solution is reached, but that seems to be part of a possible road towards a change. If I have read him wrongly, then I apologise and ask him to clarify what he meant. So I am unsure if the characterisation of obstruction is accurate, and I also wonder if it was appropriate. Double sharp (talk) 18:12, 25 October 2020 (UTC)

@R8R: Forgotten ping, sorry. Double sharp (talk) 18:17, 25 October 2020 (UTC)

About right, Double sharp. The "willing to start now" might need a pinch of salt, because other discussions need attention too.

Then, about "requirements". In the design process, there are two poles to recognise: the science & content we want to show versus the good-webdesign we want to apply. Example details of the first, content: (7-8-9)+1unk cats; express other properties (SoM, origin, today by fontcolor and borderstyle); other info (say make possible to clarify electron config with support colors). Example details of good (web)design: sure font contrast per w3c (black+wikilinkblue or more?); colorblindness (CB) awareness; what to hyperlink; and prime legend-functioning: ability to connect cell bg-color<-to->legend location two-way, and also between cells to distinct colors; understand what the Reader of this encyclopedia expects (subconscious too), and what helps them, and even invites them to discover more. These examples are just for starters. These poles in general have competing requirements, and each requirement (whichever pole) will reduce the degrees of freedom of design. Consider that, likely, "9+1" will make it impossible to serve all i.e. expect a need to compromise *somewhere, somehow*.

Now getting the requirements (desires, wishes, aims, options) clear & agreed upon is not a one-dimensional route, and so not suitable for a discussion/thread like this page uses (I'd think: per pole a WP-page+talkpage). Each set of requirements needs thought development, establishing principles, instruments to work with, check options. Over this, it requires a design attitude from those involved, willing to communicate and accept, to learn and understand, to explore new areas of knowledge. (For example, I am thinking about inviting WMF webdesign people, to help solve our quests; like a "zoom-in/zoom-out option" to go from cell<->PT, would be great ah!). So far for step 1: 'requirements'.

After this, the design can start. It is a pilgrimage with uncharted routes, guided by stars. In the end we will meet atop a hill, together with thousands of readers, to enjoy the many shining bright periodic tables above us—the ones we thought were the stars guiding us all along. -DePiep (talk) 20:27, 25 October 2020 (UTC)

First of all, Double sharp, thank you for stepping in when you sensed the discussion was getting heated. May I ask you to stick around this discussion until we've finalized the requirements?

When I proposed in the previous section we invent a new coloring scheme, DePiep said we should discuss the requirements first. So I set up this section to talk about requirements. I would normally expect to see here some actionable suggestions, ones we could write down and refer to later, but what I saw instead is some vague description of how the problem is difficult (at least, that's what it appears to me: I could not extract any proposition from it one could say yes or no to), but at the same time, I was reminded that we need user requirements. That is precisely what this section is for. I welcome what explanation DePiep has just provided. I sort of wish it was provided the first time I asked, but late is better than never.

DePiep, I have a question. You mention there are other things you are preoccupied with and you cannot devout all of your attention to this quest. We don't know how long it will take to have you turn all of your attention to this quest. We don't know how long it will take to formulate the perfect requirements. We don't know how long it will take to produce a perfect table based on these perfect requirements. This could take a lot of time, so the question is this. Would you agree we try to create a new coloring scheme in the meantime, before the perfect one has been produced, and not oppose adoption of such an interim scheme unless there are ways in which it is worse then what the current scheme? If the answer is yes, then of course all of my obstruction concerns will be clearly devoid of any substance, and I'll gladly take that word back.

Since there were two questions pointed at me, I feel obliged to answer them.

Why did I not restart the 2016 discussion? -- the simple answer is, I did, back in 2017. I even asked this question: "Could you explain to the amateur that I am what's wrong with them [the 2016 colors]?" I didn't get a sound answer for that one. What I did get is that saying they were not perfect, and no actionable comments. I was hyped up back by what you could produce then, but this hype has waned now that it's been three years since then and nothing has been exhibited, and it's still far from clear when it will.
Why do I keep suggesting the 2016 proposal was frustrated instead of admitting it was not ideal? -- the answer is simple, too. It was a clear improvement compared to the old scheme, and it was frustrated because it was not perfect even though it was an improvement. In fact, the fact that it was an improvement is something you mention in the 2017 discussion, too. But to quote, you said, "please do not propose that" (italics in original), because the 2016 proposal was "not perfect." I didn't press for adoption of the proposal at hand back in 2017 because I thought you'd create an even better scheme. Three years later, nothing has happened, and I begin to think we could've spent those years with a better scheme.--R8R (talk) 21:54, 25 October 2020 (UTC)

@R8R: Thanks for your kind words; I will have a time shortage, but I will try to at least stick around at the project page issues. I confess that I am not 100% sure about starting to create a new colour scheme in the meantime, when we have not even gotten to the stage in our plans where we arrive at deciding what categories there are, but for me it is simple: if that is what a majority of project members want, then so be it. Double sharp (talk) 22:18, 25 October 2020 (UTC)

Thank you very much! I'm reassured by that.

The point in having an interim scheme is simple. The way DePiep describes it now sounds impressive, but what description I heard in 2017 sounded impressive, too, but it didn't result in anything. If I were confident that a coloring scheme would actually be created this time, then this would not be necessary. I was confident back in 2016 and 2017 that a new scheme would be produced, and I didn't press for the proposal at hand precisely for that reason, but in the end, nothing was presented, so there's a chance this might happen again. (I am absolutely sure DePiep means it now, but I'm just as sure they meant it back then, and we didn't get anything.) I think there shouldn't be a problem recognizing an interim scheme in case the creation of the perfect scheme takes far too long again; after all, if the perfect scheme is created, it will likely supersede what interim scheme there will be.--R8R (talk) 22:39, 25 October 2020 (UTC)

DePiep, I have posted to user talk:R8R about the tone of his post, which I agree was inappropriate. Would you be able to look past the tone and comment on the questions that R8R raised that you see as pertinent to this discussion? Thank you. EdChem (talk) 04:44, 26 October 2020 (UTC)

Fair enough, R8R. This week I will publish a structured subpage to discuss and develop the whole issue of periodic table graphics (say, presentations). PT content topics (like which categories to cover? do blocks instead?) belong here at WT:ELEMENTS. From there too, we will look at the feasibility of an early new color scheme.

I will report here. Anything to clarify? -DePiep (talk) 19:15, 26 October 2020 (UTC)

Great; I'm happy to hear that. I am also happy to conclude that my worry about obstruction turned out to be unfounded, and as promised, I'll gladly retract it.--R8R (talk) 10:36, 1 November 2020 (UTC)

Due to distractions, I have not been able to meet my stated one-week goal. Personally, I am fine. -DePiep (talk) 22:02, 5 November 2020 (UTC)

Since I have decided that categories are actually OK, I feel I should answer R8R's question from the beginning: no, I don't feel that we should have predicted colours. Such an avoidance releases us from the burden of how to deal with conflicting predictions (as gets common for heavy enough elements) and elements for which current predictions means they fit poorly in any category scheme (Cn, I am looking at you). I also think that the amount of effort that would be needed to scour the literature and decide on a single colouring for some specific elements here (copernicium and oganesson) is not proportionate to how little the post-108 elements really matter for the average general beginning reader. Anyone who wants to know can click on the individual articles where these things are discussed in detail. Double sharp (talk) 23:05, 15 November 2020 (UTC)

Astatine

I suspect that there are many readers who vaguely recall their high school chemistry who would be surprised to see astatine classified as a metal. Am I correct in saying that astatine shows properties that are halogen-like and also that are metal-like? WP's article lede states that "most of [astatine's] compounds resemble those of iodine" and that it is "usually classified as either a nonmetal or a metalloid." Does the article need an update or is putting it as a metal another choice where multiple perspectives are reasonable? I note that we also state that "astatine is the least reactive of the halogens," citing an article from 1959. Is it predicted Ts would be less reactive still? Or not a halogen? Should the table or PT article cover the classification of elements like At and the more recent discoveries? I know the hassium article includes that the properties were consistent with its expected position in the PT, for example. Would a reader expect Og to be a noble gas? If we are talking about changes to the PT article and display table, I think it's worth looking at other potential topics / issues. Thoughts? EdChem (talk) 21:41, 18 October 2020 (UTC)

Should this not be a ==-level section? -DePiep (talk) 21:52, 18 October 2020 (UTC)

I was ambivalent about where to put this... I have no objection to it being changed to double-equal header rather than a triple-equal. :) EdChem (talk) 21:56, 18 October 2020 (UTC)

@EdChem: (edit conflict × 1) Thank you so much for keeping each topic in a separate section! ^_^ Trouble with astatine is that indeed it shows both halogen-like and metal-like properties. If we factor in predictions as well – condensed-phase astatine according to calculations would be metallic. I think the problem is that for most chemists At is a nonentity and so if it is thought about at all, it is thought of as "well, it would be the fifth halogen" and packed off to the nonmetals, so that's where you will normally see it. Which to some extent it is, but sources focusing on its chemistry (which understandably are not that many) tend to note that it is surprisingly more metallic than you might think and has similarities to Po also in that way. The low concentrations you have to work with no doubt make the interpretation worse, especially when you are inevitably using iodine as a tracer which is slightly different (it was thought for a while that AtO₃⁻ was being formed because astatine was being carried with iodate – actually it turned out to be an At⁺ iodate salt after all IIRC). So I think you'll find that multiple perspectives are common in the literature, and then we have the problem on which sources to weight more.

As for Ts: it is not even clear if Ts is a halogen in the first place. The problem is whether "halogen" means a category of similar elements that implies nonmetallicity (in which case probably metallic Ts with little inclination to form tennesside anions is a kind of lame inclusion), or the whole group starting with fluorine. The IUPAC Red Book isn't helpful here because it lists only F through At but was written at a time when Ts had not been discovered. Similar issue with whether Og is a noble gas or not: judging by some articles, it might not be.

You may begin to see why I eventually started thinking that maybe it's best not to colour categories at all if so much of the p block has different sources saying different things about metallicity, no one is sure if "halogen" and "noble gas" mean the group or just part of it, no one is sure if "transition metal" includes group 12, no one is sure whether to include metalloids or just to use metals and nonmetals, no one is sure exactly how the nonmetals should be divided or even if they should be divided and what they should be called as a whole when the noble gases are excluded, and no one is sure what to call the metals in the p block. Most tables do go beyond blocks alone, but a significant minority stops there, and at least that's one thing people are relatively sure about. But maybe you have another way to fix this problem and retain colourings, in which case I'm ready to hear it! ^_^ Double sharp (talk) 22:02, 18 October 2020 (UTC)

@EdChem: Immediately following its production in 1940, early investigators considered astatine to be a metal. No surprise there considering iodine looks like a metal under white light and is a semiconductor with a band gap of about 1.3 eV. Batsanov in 1971 gave a non-relativistic calculated band gap for astatine of 0.7 eV i.e. a semiconductor with a metallic appearance. Relativistic calculations in 2013 predicted astatine would be a full-blown fcc metal. This article has been cited 35 times without dissent.

As far as the p-block is concerned, there is no drama and no fuss. We know from RS that the elements commonly recognised as metalloids are B, Si, Ge, As, Sb, and Te. To the left are the post-transition metals. In the literature, group 12 are treated as TM or PTM on about a 50-50 basis. Since group 12 have a predominately main group chemistry, not to mention the dramatic weakening of physical properties going from group 11 to 12, we count the latter as PTM. To the right of the metalloids are the noble gases, and the halogen nonmetals. The residual nonmetals, H, C, N , O, P, S, Se exhibit a rich array of shared attributes and some or all of them have been referred to by as many collective literature-based category names, some of which are not unique to the nonmetals involved, including:

biogens;
CHONPS nonmetals;
intermediate nonmetals;
light nonmetals;
moderately active nonmetals;
organogens; and
other nonmetals.

Some other terms that are about to appear in RS are:

coactive nonmetals;
moderately active nonmetals (again, and as also suggested to me by another chemist);
pre-halogen nonmetals;
rail nonmetals (refractory and interstitial light-life); and
Goldilocks nonmetals.

The last of these arises since the seven nonmetals concerned have properties that are neither too extreme, nor too weak, but just right to support life as we know it. Further, each of them nonmetals have their own WP biogeochemical cycle article. Sandbh (talk) 11:45, 19 October 2020 (UTC)

@EdChem: As a chemist, how many of these names given by User:Sandbh would you recognise or understand if presented without an explanation? And would you mentally read them as meaning exactly the set {H, C, N, O, P, S, Se} that Sandbh is referring to, or as some other set, or just as a set with some fuzzy boundaries? I ask this just as a rough gauge of the prevalence of these names. Double sharp (talk) 17:48, 19 October 2020 (UTC)

I'm curious about this, too. I said before that I didn't think there was an easily recognizable name that would point to this set of elements, and I still have this feeling, so I'd greatly appreciate another opinion about this that could point out whether I am right or wrong.--R8R (talk) 19:49, 19 October 2020 (UTC)

(P.S. Second question naturally contingent on whether those names are recognised/understood.) Double sharp (talk) 20:05, 19 October 2020 (UTC)

@Sandbh: Which of those twelve names do you support Wikipedia showing for {H, C, N, O, P, S, Se}? (Possibly more than one.) And are there significant differences in how much support there is in the literature for each of them? I ask to get a clearer idea of your position. Double sharp (talk) 20:59, 19 October 2020 (UTC)

A better quote about astatine from C&EN News rather than just a chemist's blog: "is astatine more like a halogen or like a metal?" So there's proof that it is not clear cut if "halogen" is a group name or a category name, and that it is not even clear cut if At is a halogen in the first place. Double sharp (talk) 20:52, 19 October 2020 (UTC)

Some answers, which are my opinion / experience and thus may differ from the literature (though likely not from common usage):

I don't recall having seen most of these names and I would stop and wonder what most of the modifiers of "nonmetal" actually meant.
I certainly am familiar with CHONPS but think of it being about the non-metals that are essential for life and also as a way for students to learn about the composition of major biomolecules. Carbohydrates require CHO, though of course some also have N (like glucosamine). DNA / RNA require CHONP, proteins require CHONS (though there are some with Se too, so I am not surprised to see it appearing). Like most such learning tools, it is a guideline to use with flexibility. After all, there are biomolecules that contain I, and chloride ions are common in intracellular and extracellular fluids – so I can see a reason for it to be taken as a reference to those two halogens as well.
On At being excluded from the halogens on grounds of chemical properties but included in the term "group 17 elements," I can see the appeal, and it becomes stronger when extended to Ts which is likely to be much more metallic in properties (assuming we ever can investigate them)... but on the other hand, the terms chalcogen and pnictogen already cover both metals and non-metals (as I understand them) so is there a reason for a different approach to halogens? But, going back to the first hand, is a term like "noble gas" appropriate for Og if it turns out not to be gaseous? Further, we already have a significant known chemistry of xenon and similar chemistry can reasonably be anticipated for Rn or Og.
Sandbh, I accept that the chemistry of astatine has significant metallic character, but I am not sure that you recognised my points:
- Even if At should be categorised as a metal on the WP PT, will it be a surprise to readers and thus warrant some comment in explanation?
- Classifying At as a metal is inconsistent with some statements that I quoted from our astatine article, and so some changes there or to our PT article or both may be warranted?
As a general observation, when I see comments that address situations where there is ambiguity or disagreement or the like and yet the comments read like there is no alternative view possible, I tend to become more doubtful and questioning. When !voting is an exception (at least somewhat) because advocacy for a perspective is understandable there, we are trying to develop appropriate encyclopaedic content and that requires consideration of the breadth of views from RS, etc. Now, someone well across the literature (and in this area, that is not me) may well have already considered issues like DUE and RS etc and so can jump direct to a balanced summary. Just a thought worth bearing in mind, IMO, for whatever that is worth.

EdChem (talk) 00:10, 20 October 2020 (UTC)

@EdChem: Will respond to the rest later, but according to predictions, it seems Og is probably neither noble nor a gas. (Two links; it seems to be a reactive semiconductor with common +2 and +4 states. I don't think either raised the comparison with known elements, so this is just OR for the talk page, but that sounds to me way more like germanium or tin than xenon. ^_^) Not that I think we'll get to experimentally test it anytime soon. As for chalcogens, I recall that Droog Andrey mentioned before that in Russian texts, polonium's position as a chalcogen is somewhat ambiguous and so is astatine's as a halogen, which may come from the "chemical properties" urge (don't know about pnictogens). I think that there are a lot of alternative views in the literature, and although I certainly prefer some to others, I think that WP should instead clearly reflect the situation that the literature is not decided. And I am concerned that any explicit colouring of categories makes it harder to show that undecided situation, particularly since you would need footnotes for nearly every p block element. Double sharp (talk) 00:25, 20 October 2020 (UTC)

@EdChem, R8R, and Double sharp:

The general reader. I'm reminded of EdChem's comment, somewhere, that we're writing for the general reader. In this context it doesn't matter what a chemist thinks.

What does science say? EdChem, I support your approach of considering what the science says, rather than what our personal preferences are.

Fuzzy thinking. On the science front, fuzzy concepts are, as I understand it, a part of chemistry and explain why, to some extent, chemistry cannot (not yet, anyway) be fully reduced to physics. That said, sharpish and fuzzy categories have always played a part in the development of science, generally. In chemistry there are fuzzy concepts like acidic-amphoteric-basic; metal-metalloid-nonmetal; or ionic-polymeric-covalent. Despite their fuzziness, these basic groupings include some of the most powerful ideas in chemistry.

The breadth of a discipline. I cannot speak for chemistry but in my own area of professional expertise (strategic people management; learning and development), I know a lot and I don't know it all. And I was never worried about unfamiliar jargon; there are too many models, theories, luminaries, and sub-disciplines, to be across them all. The good thing was I never stopped learning. In strategic people management you can always fall back on something called the people management life cycle (capability > raise > train > sustain > capability > repeat). Anything else you either learnt doing your post-grad, or you pick it up along the way, as required. In chemistry, as far as I have read, the PT serves as the organising or learning icon, plus whatever you learnt at university, and you pick up the rest along the way, as required.

H, C, N, O, P, S, Se. The WP situation strikes me as being akin to post-transition metal territory. There are several nomenclature possibilities found in the literature. Scientifically, my pick would be moderately active nonmetals. It’s an ugly, clumsy and relative term. But it’s generic; not tied to any particular sub-discipline, like biochemistry (important as that is); doesn’t have significant overlap issues; and is conceptually anchored in the left-right activity progression seen across the period table, as widely recognised in the literature. For the nonmetals this of course refers to the chemically weak metalloids to the left and the highly active halogen nonmetals to the right.

I no longer support other nonmetals. The great majority of Ngram hits are to vague and irrelevant expressions like “plastics and other nonmetals”.

Ambiguity. When it comes to situations where there is ambiguity or disagreement or the like I strive to make reference to all the possibilities. See, for example, metal; nonmetal; post-transition metal; metalloid; heavy metals; and lists of metalloids. Sandbh (talk) 10:55, 20 October 2020 (UTC)

A WikiGnome writes

One interest of a fellow gnome, Certes, is fixing bad links to WP:PTOPICs. I lunged with krypton, lithium and neon, and he fixed 150-odd bad links. He parried with chromium, lead and palladium. I riposted with copper, gold, radium and silver. (Discussion here.)

Are there any other elements which may have collected bad links? Narky Blert (talk) 20:37, 28 October 2020 (UTC)

@Narky Blert: None very readily come to my mind unfortunately – but it looks like some elements have associated disambig pages like [[Helium (disambiguation) or Tungsten (disambiguation), so there might be a few bad links. Double sharp (talk) 20:58, 28 October 2020 (UTC)

As I understand the question, I think it refers to resolving links unintendenly linking to the main topic Krypton while Krypton (TV series) should be the target [1]. So ELEM people are invited to look for WP:DAB-misguided incoming links.

If I am correct in this, you and your friend are doing a great job! From our point of working (elements), we see this: Krypton has 1000–1500 incoming links [2]. Shoud I check and fix them? Then think about the uranium or hydrogen numbers. So: please continue! (I remember re Congo: the nice part was, researching the intended articles. nice. And skip one if its a headache). -DePiep (talk) 21:25, 28 October 2020 (UTC)

Yes, the aim is to make many minor fixes such as this. (The access-date genfix is incidental.) Working through Krypton and others mentioned here may no longer be very productive as we already fixed the easily found errors but other elements may have problems to find. I didn't have the patience to check every link; I've been searching with filters both positive (e.g. linksto:krypton superman) and negative (linksto:krypton -noble -element -argon). I've had at least a cursory glance at the first 30 elements (H–Zn). Certes (talk) 21:45, 28 October 2020 (UTC)

You mean fortunately... There's little point in checking every incoming link when the vast majority will be perfectly good. The most productive method is to look at articles where bad links are both likely and easy to find. (London is in the "too difficult" pile.) I have a "sore thumb" style whereas Certes has a more structured one. Tetrahedron/Tetrahedron is a well-known problem; I started thinking about other possible ones relating to chemistry when I came across an article with a short list which included a DABlink (which was how I found it), Fiat (which was how I solved that problem) and Proton (which jumped out at me). Narky Blert (talk) 05:22, 29 October 2020 (UTC)

RFC proposed: Nonmetal categories

Colleagues, I intend to put the following RFC, at Periodic Table talk:

Should the WP periodic table show 2 or 3 nonmetal categories(?):

2	3
• Noble gases: He to Rn • Reactive nonmetals: H,C,N,O,F,P,S,Cl,Se,Br,I	• Noble gases: He to Rn • Halogen nonmetals: F,Cl,Br,I • CHONPS nonmetals: H,C,N,O,P,S,Se

I’ll draft some accompanying notes.

Comments please. Thank you. Sandbh (talk) 08:14, 29 October 2020 (UTC)

(I understand you ask for early comments here).

I hope you also spend arguments on the strengths of the two 'new' categories. As in: strong internal, common properties (the category characteristics), and distinction re other categories (otherness). This has been mentioned before on this page.

Minor idea: maybe you can 'name' the second category more completely 'CHONPSS', we should consider the name might stick ;-) and so be more to the point. (Or SCHONPS? CHNOPSS?). As said: minor -- and a bit funny too. -DePiep (talk) 18:30, 29 October 2020 (UTC)

Is there a WP:RS which says 'CHONPS'? or any variant thereof? Unless there is, it would be WP:OR/WP:SYNTH. Narky Blert (talk) 20:43, 29 October 2020 (UTC)

Chalcogen is a long-established name for the non-metals in what used to be Group VI. Narky Blert (talk) 20:55, 29 October 2020 (UTC)

@Narky Blert: re 'is there a RS for "CHONPS"?'. If it is descriptive, we don't need a source nor OR. However, then we'd have to write 'CHONPSSe' in full. Despite what Sandbh already wrote, below, I think we should use this correct full list. -DePiep (talk) 23:48, 30 October 2020 (UTC)

CHON, SPONCH, CHNOPS & CHONPS in the literature
"We may have on one side of a line, life, C H O N, and on the other side the same elements, CHON, but in different proportions, representing the absence of life, which is death, and between them, circumstances* which determine the conditions of these elements, whether they shall exist in one combination or an other. *Thus, for instance, flesh or blood is composed of C⁴⁸H³⁹N⁶O¹⁵, and when these decay and putrefy, Carbonic Acid (CO²), Water (HO), and Ammonia (NH³), are the result."	Clark 1865
"A survey of thermodynamic properties of the compounds of the elements CHNOPS Progress Report, 1 Jan. - 30 Jun. 1969"	Armstong 1969
“…for instance, the atomic numbers of the six SPONCH elements are 16, 15, 8, 7, 6, and 1, respectively. (Notice that the acronym SPONCH lists the six elements in order of decreasing atomic number.)”	Sanders 1986
"Phosphorus is unusual in the CHNOPS list…"	Keddy 2007
"CHNOPS constitute the building stones from which the matter of life is built"	Per Enghag 2008
"This is accurate enough to distinguish relativistic mass differences between expected isotopes of CHONPS elements…"	Moseley et al. 2010
"Morowitz has, therefore, described Earth-based life forms as CHNOPS organisms."	Keddy 2011
"To assess the astrobiological potential of these worlds we need to know whether they can supply, in biologically available forms, the essential elements for life (CHONPS)…"	Longstaff 2014
"This is a tiny subset of the molecular species of CHNOPS with 100,000 atoms per molecule."	Kaufmann 2016

@DePiep and Narky Blert: Thank you.

1. Yes, I'll add some notes about the two new categories, including along the lines you requested.

2. A funny thing: I looked at the 2 column in the above table and wondered how so many nonmetals could be "lumped" together in one category, given their diverse attributes.

3. I feel there's no need to e.g. add "Se" after "CHONPS", since S can be read as S and Se. Not forgetting S and Se are in the same group, and share an appreciable number of properties. For example, selenium is found in metal sulfide ores, where it partially replaces sulfur; both elements are photoconductors—their electrical conductivities increase by up to six orders of magnitude when exposed to light. The two nonmetals form about a dozen chain and ring entities of composition S₍₁₋₇₎Se₍₁₋₆₎.

As well, CHNOPSSe doesn't sound or look quite as good.

4. There are some citations in the table. Note the three variations: CHON (this one is very old); CHNOPS and CHONPS.

5. We have an article on the CHON elements. This article, sans attribution, notes "The acronym "S.P. Cohn" was also used in high school biology classes to represent the six chemical elements."

6. Here is the link to the Wiktionary entry.

7. The WP PT is a metallicity based PT, showing the L−R progression in metallic to non-metallic character, so a group name such as chalcogen is not so relevant here.

8. I've asked some chemistry teachers if they use CHONPS. One said, "Yes, I have used CHNOPS for years." @EdChem:, a chemist, is familiar with it too. --- Sandbh (talk) 03:39, 30 October 2020 (UTC)

Relocated to a subpage of mine, in order to reduce the edit load here. Sandbh (talk) 01:40, 20 November 2020 (UTC)

Comments on draft RFC

Narky Blert

CHNOPS looks to me like a wholly anthropocentric, and therefore arbitrary, selection, based on biology not chemistry. 99.9% or better of biologically important compounds (on Earth, at any rate) contain nothing else. Most of the remainder include a metal cation. The number of compounds which contain any other element covalently bound is tiny. All the ones I can think of (other than the I-containing thyroxine) are peculiar specialist defences against predation, or are manmade pharmaceuticals or things like nerve agents. How many biological molecules can you think of which contain covalently-bound B, Si, Ge, As, Sb, Se or Te? Narky Blert (talk) 19:56, 30 October 2020 (UTC)

@Narky Blert: Thank you.

There are a couple of interesting considerations here.

1. The biogeochemical or toxicological (e.g. nerve agent) properties are but one facet of the CHONPS or SPONCH bunch, where S includes S and Se.

2. While the noble gases, as elemental substances, can be characterised by their invisibility and torpidity, and the halogens by their variegated appearance and acridity, the Se-S-P-O-N-C-H set exhibit the following characteristics:

Sub-metallic (C, P,^ Se), coloured (S) or colourless appearance (H, N, O), and a brittle comportment if solid.
Being sandwiched between the strongly electronegative halogen nonmetals and the weakly nonmetallic metalloids, their net physical and chemical character is moderately non-metallic. See Table 2, here.
Pronounced vertical, horizontal and diagonal relationships.
An overall tendency to form covalent compounds featuring localized and catenated bonds as chains, rings, and layers.^^
Prominent organocatalytic, and energetic (explosives and combustion) aspects.
In light of their relatively small atomic radii and sufficiently low ionization energy values, a capacity to form interstitial and refractory compounds.

^ Black P, the most stable form in ambient conditions, is these days easily prepared.

3. I don't make up the nomenclature. I can only go by what's used in the literature.

I used to think that e.g. biogen nonmetals, was too biologically focussed. But I finally got that this was just a personal preference, based on what I like or dislike, rather than a scientific decision based on scientific terminology, as per the advice User:EdChem has given us. The first time I saw a 32-column PT with a split d-block, I thought it was ugly and loathsome, and unacceptable. Now I understand that this was on account of "values→beliefs→rules" perceptual filters operating on auto-pilot. Now I know how to turn off my auto-pilot, and set aside my values→beliefs→rules, and just look at and accept what the science says.

4. Our periodic table is not just a chemistry-based table, so to speak, although chemistry is certainly a biggie. Other periodic-table fields of relevance include e.g. physics; biology; metallurgy; cosmology; geology; materials science, and so on.

5. I'm all ears as to a "better" name, based on the 11 options available to us.

6. Off hand, I know some sea sponges have endoskeletons comprised of SiO₂. Among V series nerve agents, VX is the most studied: C₁₁H₂₆NO₂PS. The selenium analogue of VE is selenophos C₁₀H₂₄NO₂PSe. This is said to be more potent than VX. The article "Why nature chose selenium"[3] gives a history of the important discoveries of the biological processes that selenium participates in, and a point-by-point comparison of the chemistry of selenium with the atom it replaces in biology, sulfur. This article [4] discusses the role of boron in metabolism, "that render it necessary for plant, animal, and human health." As I understand it, the chemistry of B, Si, and Se tends to be characterised towards the covalent patch of the field of bonding possibilities.

--- Sandbh (talk) 01:44, 31 October 2020 (UTC)

@Sandbh: Some terrestrial plants incorporate SiO
₂ - two reviewish articles 1 and 2. My guess would be that all the chemistry involves compounds with Si(IV)-O bonds. Narky Blert (talk) 06:58, 31 October 2020 (UTC)

DePiep

Note #DePiep-01: using "category" in articles.

The word category has no specific meaning in (PT-)chemistry, unlike group (ouch, twice), period, and block. So far, until ca. mid-2020, we have used it internally only on talkpages etc. But recently it has been added in FA-article TOC-level: see Periodic table § Categories. This might be too much of a stretch (OR?). Hey, there is not even an article Category (periodic table).

I add that this categorisation used to be by 'the YBG set of requirements': "all elements, once, and there be coherence" (my short take). In the current (2019) form, this leaves out sets like Rare earth metals and Coinage metals. Later on in this discussion we might meet this issue: random, non-YBG categories to deal with. (Classifications is a good word b/c it has reason; set is really ok to make any list).

My long term question: do we keep using category in articles for this classification-by-YBG-terms? -DePiep (talk) 00:28, 31 October 2020 (UTC)

@DePiep: The word category is found three times in the IUPAC Red Book, including:

"In particular, alcoholates, thiolates, phenolates, carboxylates, partially dehydronated amines, phosphanes, etc. are in this category." (p. 151)

Fowler's modern English usage (1998) says:

"category. To begin with, restricted to its original philosophical meaning, ‘one of a possibly exhaustive set of classes among which all things might be distributed’ (COD); later (attributed to Kant) ‘one of the a priori conceptions applied by the mind to sense-impressions’ (COD). These philosophical senses made their way imprecisely into general use from the 17C. onward, and settled down with the broad meaning, 'a class, or division, in any general scheme of classification’. This more general sense, though objected to by Fowler (1926) and others, is now uncancellably established. Examples: She placed them in two categories: the honest imbeciles and the intelligent delinquents—O. Manning, 1960; He had had the whole of creation divided into two great categories, the things he was for and the things he was against—M. Frayn, 1965; ‘I hope you aren't a clergyman.' She said it with real vehemence, as if it was the one category of person she was not prepared to have in the house-A. N. Wilson, 1981"

Here in WP, we refer to "Chemistry" as a category, [5] with 72 subcategories. --- Sandbh (talk) 03:02, 31 October 2020 (UTC)

Yes, yessss, YES! As you say: "category" in chemistry literature is different and more wide than the current, internal enwiki meaning. -DePiep (talk) 23:53, 7 November 2020 (UTC)

R8R

There is no comment on the content of the RfC proposal I could give (I think it's fine), but I am wondering what Double sharp has to say. My understanding is that Double sharp has effectively suggested there should only be four categories, namely, the s-, f-, d-, and p-blocks, and expressed his desire to have an RfC on that topic, too. That proposal directly conflicts with this one, and my thinking is, there should ideally be an attempt to coordinate the two given that they are mutually exclusive.--R8R (talk) 12:08, 1 November 2020 (UTC)

@R8R: The two proposals aren’t necessarily mutually exclusive. There are two ways of achieving both solutions: categories and blocks; or blocks and categories, as I suggested, here. Sandbh (talk) 12:43, 1 November 2020 (UTC)

I have my doubts about that. Double sharp's point was not that blocks should be introduced, his point was that blocks should be introduced instead of the current categories. There is no way the two can go together as far as I can see, which is why I'm asking you to coordinate your proposal with his.--R8R (talk) 11:33, 4 November 2020 (UTC)

I confirm that User:R8R is exactly right. For me, the categories should not be used to colour the table at all. In my view, they can be mentioned in the article, mentioned in the exact infoboxes, along the lines of "Selenium is a chalcogen. It is most often considered a nonmetal, but has sometimes been referred to as a metalloid or more rarely as a metal." But I am wholly opposed to any single colouring of categories because to me it fails to reflect the real source situation, which is more along the lines of the quote I just said. That is what will be seen if we take a sweep across many sources, at least as long as we don't just cherry-pick just the ones that look like they support the particular category du jour. As far as I can see, this proposal of mine is exactly mutually exclusive with the two detailed here. I personally would not have wanted to go for this now when the group 3 discussion is not quite finished yet (we already have majority for changing back to Lu but discussion with R8R is still going), but if it is to go forth, I agree with R8R that the "blocks alone" proposal should absolutely be included.

As well, the statement Anything beyond a simple !vote for either two categories; or three categories (+ your Table 2 pick/s), should be placed in the "Discussion" sub-thread, following in the proposal looks to me quite like an attempt to sideline other proposals. As far as I am aware, this is not usual practice on WP, where it is reasonable and possible to !vote for neither option presented in a survey and suggest one's own. I hope this is not the case. Double sharp (talk) 11:42, 4 November 2020 (UTC)

@Sandbh: I am seriously concerned about the changes introduced to the draft since I commented on it. The idea that a vote for a different change the one being implemented can be still counted as a vote for a change is very disturbing. Votes may not be counted the way they were intended to be counted. That's a serious problem.

There are better ways to conduct a multiple-choice poll. For example, when New Zealand was considering a flag change, citizens first voted on which new flag should be adopted if there was to be a change at all, and then they voted on whether there should be a change to that selected flag. This seems much fairer to me; however, I am concerned that participating a two-stage RfC is a bit much to ask from outsider participants, which is why I think the first round (selecting the favored option) should happen within our project, and then the agreed upon name should be offered to outsider participants. It was my impression this was what was going to be proposed the first time.--R8R (talk) 12:28, 4 November 2020 (UTC)

@R8R: a bit much to ask from outsider participants: Thank you for phrasing this eloquently, because it's precisely why I'm generally not too enthusiastic about opening the question to a wide RFC. After all: the general community probably is not at all keenly aware of the situation. Even among chemists some of these issues are pretty obscure. Double sharp (talk) 15:57, 4 November 2020 (UTC)

EdChem

Fidelity to what?
Presentation is unbalanced / biased – a comment that reactive nonmetals is suspect (IIRC) is clearly advocating for one position over another.
Since both options include noble gases, the question is really about the non-metals excluding the noble gases.
I understand that halogen non-metals is offered to exclude astatine (and possibly tennessine) but how many who use the PT are going to continue to simply refer to halogens?
why is the option of halogens, chalcogens, pnictogens not offered?
The question presupposes that none of the metalloids are non-metals. I suspect many readers will have learned the PT with a metal / non-metal boundary and metalloids around it, some as metals with non-metal characteristics, some as non-metals with metallic characteristics. Is there an established consensus for the metalloids to be treated separately from metals and non-metals? If so, I would note this in the preamble or background.

An RfC is meant to offer a simple question for outsiders to consider and offer a comment. The background should neutrally present the arguments / evidence for others to consider in forming a view. Advocacy belongs with comments / views / !votes, etc. The above list points to potential issues that an outsider might wonder about and which thus are worth considering in constructing an RfC. Also, an RfC cannot override what the consensus of RS is... and so should not offer such an option. For example, we wouldn't hold an RfC to decide on metalloids as a separate collection / set if the consensus of literature is that they are; rather, we'd document the RS consensus and (as needed) discuss its implementation. We may hold an RfC, for example, if RS given DUE consideration support two possibilities and we choose which to present and which to discuss as the alternative.

Some of the issues are ones I haven't reached a conclusion with my Wikipedian hat on. As an educator and thinking pedagogically, I can't see the benefit that comes with "halogen non-metals" over "halogens" given the insignificance of astatine. As a chemist, I can see the benefit of precision but wonder if chemists will simply understand when astatine is included / excluded by implication. As Wikipedian, however, as RS are drawing such distinctions I have to wonder about the appropriate inclusion of such terms... but at the potential cost of over-specialising encyclopaedic content and in so doing rendering WP less accessible for non-specialist readers. Not an easy one, either to decide or even to construct an RfC about... but the draft needs some re-drafting, IMO. EdChem (talk) 22:20, 1 November 2020 (UTC)

@EdChem: Thank you very much for your fresh perspective.

In drafting the RFC I aimed to be as concise as possible. There is so much more I could include in the RFC or write in response to your observations but I choose not to, in light of guidance provided to me by other editors to keep one's RFC "short".

1. "Fidelity" means better fidelity in classification science, I suppose. Or there is historical momentum. Since it's inception in 2002, the WP PT has had three nonmetal colour categories over 18 of its 20 years. We changed to two categories in 2018 as almost nobody "liked" other nonmetals.

2. NPOV is very important, and I was wondering how to achieve that. I take your point about saying reactive nonmetals is "problematic" and I added an explanation about why it is regarded as such. Do you have a suggestion for better wording? I'll look at this further.

3. I included the NG so as to put the question into context, and that it was about classifying the nonmetals generally.

4. Yes, many folks who use the PT will continue to simply refer to the halogens just as they refer to the pnictogens and the chalcogens, never mind the cross-cutting metalloids. Bear in mind the WP PT is more of PT showing the left to right progression in metallic to non-metallic character, rather than a groupic table. Also, that the main WP PT shows the group names as well as the colour categories. There is room to show the group names on our WP graphic, but other WP:ELEM editors seem not keen on this option and have argued this will add too much detail. YMMV.

5. The option of chalcogens, halogens, pnictogens is not offered, (a) in order to keep the focus simple and targeted; and (b) since the WP:PT, being a metallicity table, has never shown these as colour categories (except when At was coded as a halogen).

6. It has been known for over 120 years that metalloids have a predominately nonmetallic chemistry (including those sometimes regarded as metals i.e. Ge, As, Sb). Ontologically, anything that is not a metal, is a non-metal. That said, with the emergence of the semiconductor industry in the 1950s, and the development of solid-state electronics from the early 1960s, metalloids came to be regarded as “in-between elements. They are still widely regarded as such. I'm inclined not to add this to the RFC on word count grounds.

Category relationships
Metallic	Nonmetallic
AE-AEM-Ln-An	Halogen nonmetals
Transition metals (most of them)	"SPONCH" nonmetals
PTM ("poor" metals)	Metalloids ("poor" nonmetals)
Noble metals	Noble gases

A traditional aspect of teaching the periodic table is to contrast the alkali metals with the halogens. This cannot be done with the lede WP table.

As an educator, and pedagogically, the significance of halogen nonmetals is that the halogens are as much subject to the increase in metallic character going down the group, as other groups. Already, iodine is a semiconductor with a band gap of ~1.3 eV, and has a metallic appearance under white light. Indeed, astatine (based on its chemical behaviour) was thought to be a metal when it first synthesised in 1940. There is much on WP that is less accessible for non-specialist readers. I doubt simple periodic trends and a division of the elements into simple categories is one these, including the resulting symmetry (see the Category relationships table). This symmetry can facilitate learning since fewer observations are required to describe the applicable system.

--- Sandbh (talk) 00:42, 2 November 2020 (UTC)

@EdChem: I believe I've now addressed all of your concerns. Sandbh (talk) 08:33, 2 November 2020 (UTC)

Sandbh

@Narky Blert, DePiep, R8R, EdChem, Double sharp, and ComplexRational: I intend to go live with the RFC on Friday 6 Nov, my time (Eastern Australia). Sandbh (talk) 03:42, 3 November 2020 (UTC)

@Sandbh: I hope the issue of the collusion between your proposal and Double sharp's will have been resolved by then, and a way to coordinate the two proposals will have been found.--R8R (talk) 13:45, 4 November 2020 (UTC)

@Sandbh: (1) I presume the RfC text to be centralized will just be Q1 and Q2. (2) Do you want to explicitly request that people not respond to !votes in threaded discussions (3) You said I've asked some chemistry teachers if they use CHONPS. If you'd asked me, I would have said, "I've been using them continuously ever since sperm fertilized egg nine months before I was born. YBG (talk) 08:50, 6 November 2020 (UTC)

@YBG: (1) Yes. To be clear, I intend that the RFC will include everything between the two pairs of double lines. (2) I believe this is covered by the "Survey instructions", namely "Anything beyond a simple !vote for either two categories; or three categories (+ your Table 2 pick/s), should be placed in the "Discussion" sub-thread, following." (3) Yes, quite so. The full question was, "Have any chemistry teachers heard or used this term?" To which one responded, "Yes, I have used CHNOPS for years." Sandbh (talk) 22:14, 6 November 2020 (UTC)

@Sandbh: (2) I agree it is covered, but I'm not sure sufficiently that people will "get" it. But your choice. (3) Right, my joke only worked by eliminating the term "term". YBG (talk) 23:00, 6 November 2020 (UTC)

@YBG: Roger that re (2). I'll look at it. Brevity is everything in an RFC, but. Sandbh (talk) 23:41, 6 November 2020 (UTC)

DS

I guess people may be surprised to hear this since I have been arguing strongly against categories, but I actually think Sandbh's current version of this RFC is very good. I have absolutely no objections to this being raised. Double sharp (talk) 15:17, 13 November 2020 (UTC)

Sandbh, if I may make a suggestion: have you considered first RFC-ing a change to three nonmetal categories, and then RFC-ing separately a name for the extra one? Reason being that with this many options, I am afraid you will get no consensus for any one even if the majority are in favour of changing to three nonmetal categories. Double sharp (talk) 10:53, 15 November 2020 (UTC)

Double sharp, I logged on this am and there were 13 messages for me (roses rather than brickbats, I hope). I had thought about an RFC for three to two. Then I thought about "fear of the unknown" and why someone would vote for three, unless they knew what they were voting for. Hence the current RFC. If three gets up, then the name with the most ticks, will be the name. I've explained this in the draft RFC but if you missed it I may need to bring this out more, so that it's clearer, Sandbh (talk) 22:33, 15 November 2020 (UTC)

Sandbh, don't worry. There's no brickbats. I just wonder why you feel that we need to RFC these things when we didn't RFC the old colouring changes like Po to PTMs and polyatomic/diatomic nonmetals, to be honest. And neither did we for changing astatine to a PTM just this year: I suggested it, you didn't object, so I took it as an invitation to go ahead. Now here we are and it seems 100% not controversial. I gather you have some concerns when it comes to our project setting standards for everyone, but who said that? It doesn't seem consistent with how we edit articles. If most of us are OK with something, then can't we just WP:BRD it through like for new periodic table pictures?

I've managed to get somehow within a day at the bottom of this talk page (you're pinged) a situation where a change to 3 categories seems to be generally getting a not-too-bad reception. One "rather not", but still "we could do it", and one ambivalent. So it's just us now, and I'm for it and if I understand right so are you. The price seems to have been to stick with "other nonmetals" for now, but it's better than nothing. I worry that going on to the RFC is not only needless when there is general agreement here but may result in nothing getting done. It may be a better option to just make the 3-category change now and then, once people are used to 3 categories again, ask for an RFC to change it away from "other nonmetals" into something you would rather have. I worry that too much focus on preparing RFC's will make the project stagnate into bureaucracy, basically, as compared with how the recent recolourings of At and Cn proceeded so quickly and without a hitch, just like that of Po in 2012. Not to mention take time away from the priority of the article. Double sharp (talk) 22:45, 15 November 2020 (UTC)

Double sharp, Indeed, no brick bats! ^_^

On trifurcation, noting I haven't caught up with all my pings, the last I heard was that R8R was not supporting this.

I recall you were supportive reasonable early on, and that was welcome. It was a good example of the fact that we still spoke to one another, and we do agree on some things, never mind our philosophical differences. And the funny thing is that that's all you can do with philosophy i.e. argue about it, since their are no absolute answers. Not that R8R's view was not welcome; it's always good to hear from him.

Your idea about other nonmetals is a good one. I can see where you're going with that one.

Personally, I don't think much about other nonmetals. OTOH I now appreciate that what I like or dislike doesn't matter. If that is the state of the literature, then we should reflect that.

Similarly, I'm not too happy about e.g. biogens, given the involvement of F, Cl, Br, and I in biology. But again, I'm not responsible for the state of the literature, and my likes and dislikes don't count. And it seems that, according to literature, the halogens were late arrivals on the biological involvement scene, after the SeSPONCH non-metals. My impression is that is why the organogen name was introduced, with biogens as a subset of same.

The other thing I don't like about biogens is that it's biology specific. Then again all the Noble prizes in chemistry these days go to the biochemists. And it once again doesn't matter what I like or dislike.

A misgiving is that "other nonmetals" is not a fair(?) reflection of the literature given the panoply of other names.

I'll think about this some more. Sandbh (talk) 23:24, 15 November 2020 (UTC)

@Sandbh: Maybe it's not fair. I haven't formed a full opinion yet based on what you've given. But if it is not fair, then at least it can be an intermediate step that no one can complain is wrong (just awkward) before we move to something better; and if it is fair, then we're fine in the first place, right? The point is just to be better than what we have now. If it means reaching the gold in multiple steps rather than one, what does it matter as long as we eventually get there? ^_^

The last I heard from R8R was that his problem was more about "halogen nonmetals" than anything else. But he seems to concede that we could do that, even if he prefers we don't. He may correct me if I'm mistaken. Double sharp (talk) 23:41, 15 November 2020 (UTC)

@Double sharp and R8R: We may have a way ahead then i.e. other nonmetals and halogen nonmetals, without the need for an RFC. R8R: your thoughts will be pivotal. Sandbh (talk) 04:39, 16 November 2020 (UTC)

@Double sharp and R8R: I read R8R's good thoughts in the Nonmetal categories section. I'll go ahead with the RFC, after some more agonising about its format. Scratch one log jam, whichever way it goes. ^_^ Sandbh (talk) 05:15, 16 November 2020 (UTC)

@Sandbh: I'm not sure I understand, so let me seek some clarification. I guess based on your 04:39 post this means that the RFC will just be about changing "other nonmetals" to something else and that I should go ahead and implement the 3-way split now as a first step forward? My logic was that since we both seem to agree on that and the only objection (R8R's) was tempered, we could go ahead and do the relatively uncontroversial thing first without additional fuss, and then do the RFC for the more controversial thing (renaming "other nonmetals"). Double sharp (talk) 10:14, 16 November 2020 (UTC)

@Double sharp and R8R: Well, from reading of R8R's most recent comments he does not support trifurcation. I didn't see anything about tempering his objection. So my intention is to shortly go ahead with the RFC as is. Sandbh (talk) 22:20, 16 November 2020 (UTC)

@Sandbh: I did read in R8R's comment indeed I am rather opposed to the idea, but I also read in the same comment We could do it, and it wouldn't be the end of the world: astatine is a relatively marginal element, and we can talk ourselves into thinking it's okay. But I'd prefer we didn't create such ambiguous categorizations when there is an option not to. So that seems to temper it unless I read him wrongly. I've pinged him here, so he may clarify if he prefers.

In any case, a consensus need not be unanimous. As a general guide, I'd say that since there are approximately around five to six of us at any point, a reasonable guide may be: if it's all against one, then one shouldn't be able to block and hold "veto power". History shows about how that usually ends. But if we have two in opposition, then it's a good sign not to do something, because it means that the opposing side can convince somebody else in addition to its proponent. By that link I am not saying that R8R is trying to block things for no good reason; to the contrary, I think he has a good reason for what he says and that his perspective is worth listening to. But I also think that asking for unanimity in every decision is going to make it impossible to do anything even if the intentions are good (and indeed that even did historically happen), and that it will result in an undue amount of time taken on these issues compared to working on the article text. That is also why I am trying to see if we can avoid the RFC process. If we should be able to deal with behavioural things between ourselves and not bother outsiders with everything, then I do not see why we should not similarly be able to do so for content things. Double sharp (talk) 22:31, 16 November 2020 (UTC)

Since I am worried that my intentions here may be misconstrued: I want to clarify once again that this is not intended as a behavioural complaint and I apologise if it is seen that way. It is only intended to explain why I feel unanimity should not be necessary for a consensus, although I understand that in these more trying times at the moment it is desirable to avoid acrimony. Double sharp (talk) 23:33, 16 November 2020 (UTC)

@Double sharp and R8R: Out of respect for the two of you, I'd like to hear from R8R. Sandbh (talk) 23:07, 16 November 2020 (UTC)

@Sandbh: I understand and accept. I have tried coming up with a second compromise at the bottom of the talk page that I want your and R8R's views on as well: astatine as a halogen with a deliberate policy of not saying whether or not it is a metal, following Britannica, LANL, ACS, and RSC. I'd like your comments on that before we decide whether or not to pass to an RFC. If the two of you are both OK with either this or the previous compromise, then I guess it can be a go; if not, then an RFC seems the answer (in which case I would like to coordinate my proposals here somehow with yours). Double sharp (talk) 23:10, 16 November 2020 (UTC)

Somehow I forgot to give the link. It is WT:ELEM#Compromise idea 2: other nonmetals and halogens with At as a halogen?. Double sharp (talk) 23:11, 16 November 2020 (UTC)

WP:ELEM Protocol (very draft)

As mentioned at WP:ANI here’s a suggested protocol setting out aspirations for the way we do things around here. I've strived to keep the word count as low as possible. Less is more, in my view.

In no particular order, after item 1( = no change):

Original	Current
1. GF. Always presume GF.	1. GF. Always presume GF.
2. Bold. Be bold in making sure that our articles exhibit the best article standards and follow our guidelines. But please do not follow the rules and guidelines too strictly, keeping in mind that the purpose of our rules and guidelines is to make the best encyclopedia possible. Source: WP:ELEM on collaboration.	2. Bold. Be bold in making sure that our articles exhibit the best article standards and follow our guidelines. But please do not follow the rules and guidelines too strictly, keeping in mind that the purpose of our rules and guidelines is to make the best encyclopedia possible. Source: WP:ELEM on collaboration.
3. Civility. Polite, civil discourse is expected. Being sharp, at times, does not constitute incivility, nor a failure to assume good faith. Expressions like “you must be kidding me!” or laughter intended with ill-feeling are unwelcome, subject to sentence 2.	3. Civility. Polite, civil discourse is expected. Being sharp, at times, does not constitute incivility, nor a failure to assume good faith. Expressions like “you must be kidding me!” or laughter intended with ill-feeling are unwelcome, subject to sentence 2.
4. OR. In accordance with WP:OR there is no prohibition on OR on a talk page.	4. OR. Our talk space is directed towards improving article content, and OR in aid of that goal is fine, noting, subject to Protocol #2, WP articles must not contain OR.
5. Own. Nobody owns any article, template, table, content or domain of knowledge.	5. WP:OWN. Nobody owns any article, template, table, content or domain of knowledge.
6. Revert. Anybody can revert any edit at any time. That said, it is courteous for members to notify one another of their intention to revert another member’s edit. A lack of WP:CONSENSUS is not a valid reason to revert.	6. Revert. Anybody can revert any edit at any time. That said, it is courteous for members to notify one another of their intention to revert another member’s edit. A lack of WP:CONSENSUS is not a valid reason to revert.
7. Content. Focus on content rather than conduct, unless the conduct concerned is in breach of one or more elements of this protocol. Any issues of content contention can be put to an RFC at any time.	7. Content. Focus on content rather than conduct, unless the conduct concerned is in breach of one or more elements of this protocol. Any issues of content contention can be put to an RFC at any time.
8. Science & chemistry. Members are expected to focus on ~~what~~ science ~~says~~ rather than personal likes or dislikes: "I don’t like it" is not science. Discussions on aesthetics are permitted provided members note the subjective, culturally-informed nature of the latter. Members are asked to bear in mind the fuzzy nature of chemistry and that associated terminology and nomenclature are not necessarily precise nor consistent.	8. Science & chemistry. Members are expected to focus on science (as set out in RS, having regard to e.g. DUE, and NPOV, and to Protocol #4) rather than personal likes or dislikes: "I don’t like it" is not science. Discussions on aesthetics will necessarily be subjective, given the culturally-informed nature of the subject matter. Members are asked to bear in mind the fuzzy nature of chemistry[^[6]][^[7]] and that associated terminology and nomenclature are not necessarily precise nor consistent.
9. WP Policy. Our talk page is not the place for ~~a festival of~~ WP:POLICY citing. WP:5P5 applies here: "Wikipedia has policies and guidelines, but they are not carved in stone; their content and interpretation can evolve over time. The principles and spirit matter more than literal wording, and sometimes improving Wikipedia requires making exceptions. Be bold, but not reckless, in updating articles…"	9. WP Policy. Our talk page is not the place for undue WP:POLICY citing.

These items have been informed (and biased) by my recent experience here and at WP:ANI. Feel free to shoot, salute, amend, suggest or add your own items. --- Sandbh (talk) 01:14, 1 November 2020 (UTC)

@Double sharp, YBG, EdChem, R8R, ComplexRational, and DePiep: per above. Sandbh (talk) 01:21, 1 November 2020 (UTC)

@Sandbh: I'll write my initial comments; I hope to expand and act on them next week.

It is strange that the OR part focuses on talk pages whereas this sort of thing is clearly secondary to what happens in the article space.

The ownership part is very vague and leaves interpretation to the reader.

I wouldn't expect us to "focus on what science says." I'd expect us to focus on what reliable sources say. There's a difference between the two, namely in that the science part is less discouraging from OR on what is truly scientific and source interpretation. I wouldn't expect fuzziness being mentioned in the rule set; this is more specific that what rules suggest.

I fear that "festival of WP:POLICY citing" is too vague a phrase to have any concrete meaning; people will interpret it differently, and some may question it altogether. I don't quite understand what kind of behavior I'd be expected to display upon reading such an instruction.--R8R (talk) 12:39, 1 November 2020 (UTC)

Sandbh, the idea of the instructions (for want of a better word) at the top is to assist in applying the talk page guidelines within the context of the WikiProject. These talk page guidelines apply across WP are cannot be overridden by a local consensus. One area of concerns that I have is that your interpretation of the existing text is inconsistent with my understanding of the intent of the overall guidelines. This could, of course, point to a misinterpretation on either or both of our parts or even simply point to situations where reasonable alternative interpretations are possible. As an example, you appear to see the use of OR to be unrestricted in talk space. I see talk space as directed towards improving article content and OR in aid of that goal is fine... but that this is not a free pass for any OR. In particular, OR that reads as a reason for article space to disregard RS or DUE is problematic in my view. I hope that you would agree with this point in the abstract, putting aside whether this has actually been a problem in any WT:ELEM discussions.

Similarly, when I refer to article content being in line with Science, I mean science as known in the literature based on RS and in line with DUE, presented in an NPOV way, etc... I didn't / don't mean whatever the cutting edge of science would be. I recognise that the presentation of the periodic table as a single object is a difficulty given the variety of tables viewable in books, the small but complicated issues that arise with new elements, the issues of La / Lu, etc. We would have a much easier time writing an article on periodic tableS with pros and cons for each without taking a position on which is the periodic table. I think this is part of the reason that discussions have become complicated. If there was truly a single periodic table about which we were debating superficial presentation, we could just hold an RfC and get a decision where the science does not point at any single answer... but that's not the case. The areas of disagreement reflect scientific issues where reasonable cases can and have been made for different positions.

To me, this implies that relying on the guidance of policy and talk page guidelines that have evolved to address many difficult circumstances in the past. It seems unlikely that the issues here are unprecedented across the 6+ million existing en.WP articles.

It is worth remembering that some of these instructions point at article space behaviour and some at talk page behaviour. WP:BOLD provides advice on article space actions, reverting in article space and talk space has very different meanings, and OR is about an absolute prohibition in the former while allowing some usage in the latter. EdChem (talk) 02:27, 2 November 2020 (UTC)

@EdChem: Thank you; I'll certainly be taking on, and considering your carefully put thoughts. A quick comment. Consistent with WP:IAR, there is no "absolute" prohibition on anything at WP. Sandbh (talk) 04:45, 2 November 2020 (UTC)

Sandbh... in theory, you are correct. In practice, if some hypothetical editor tried to put unambiguous OR into an article, justified by an edit summary acknowledging that it was OR but citing IAR as justification, it would not survive long. The same fate would await IAR-justified edits that clearly ignored policies like NPOV or RS or BLP. An editor repeatedly taking this approach would likely be blocked or banned. I contend that the prohibition of OR in article space, as a matter of policy and practice, is at least close to absolute. If you doubt that, perhaps asking at WT:OR will provide further opinions. EdChem (talk) 11:10, 2 November 2020 (UTC)

Seconded. The essence of WP:IAR is, Do not rigidly apply formal rules if they impede readers' ease of understanding. It does not in any way override our duty as editors to try to ensure that WP contains nothing but accurate well-documented well-established information. Narky Blert (talk) 20:33, 2 November 2020 (UTC)

@EdChem and Narky Blert: I should've quoted WP:IAR, which is derived from WP:5P5, in full: "If a rule prevents you from improving or maintaining Wikipedia, ignore it." So it is not the case that anything goes at WP per WP:IAR, per se. It is more subtle than that. So, if e.g. OR, NPOV, RS, or BLP prevents you from improving WP, ignore them, per WP:IAR. Of course, what happens after that is subject to the views of WP editors generally, as is the case with any edit. Sandbh (talk) 21:58, 2 November 2020 (UTC)

Sandbh, perhaps think of the "you" in IAR as less you (singular), a single editor, and more as a collective... that is, IAR applies when the change prevented by a rule is one that a consensus of uninvolved editors would agree is an improvement, justifying ignoring the rule. In my experience, IAR does cause individuals to get into difficulties when what a single person sees as an improvement is seen by many others as unhelpful / detrimental. This topic might be more effectively explored with concrete examples rather than in a theoretical manner and at a meta level, where each perspective has positives and negatives. EdChem (talk) 02:40, 3 November 2020 (UTC)

@EdChem: Thanks. As WP:IAR, as written, refers to "you" (rather than a collective) I'll continue to read it that way. As I understand it, WP:IAR does not mention the need for a "consensus of uninvolved editors" before it can be acted upon (per WP:BOLD). Of course, WP:BRD can be applied thereafter. Sandbh (talk) 03:00, 3 November 2020 (UTC)

@EdChem: Oh my. I just looked at Wikipedia:Talk page guidelines and ran head long into a 6,400+ word count extravaganza. Talk about a wall of text.
--- :( Sandbh (talk) 22:20, 2 November 2020 (UTC)

@EdChem and R8R: I feel I may have now addressed and accommodated your comments. How do see the protocol now? Sandbh (talk) 03:32, 3 November 2020 (UTC)

@Sandbh: As long as #8 is there this will never get my support. The whole point is to focus on what the reliable sources say, even if you or anyone else thinks they are all scientifically wrong. Judging by your comments on the Lu sources, calling them unscientific just because they are based on single arguments (which oddly enough never stopped scientists from citing them and oddly enough is not a criticism that has appeared elsewhere), it seems you cannot do that. Oh, not to mention that #9 is also an immediate dealbreaker. We are talking on WP, we should abide by its rules when it comes to not just article-writing, but the necessary discussion that builds consensus for what goes into the articles. Yes, in general policy citing is undue because everyone involved usually has a good grasp on policy. Judging by how EdChem has to keep explaining to you things about WP:IAR which you don't listen to, it seems you don't, which makes it not undue anymore. Oh, not to mention #4 which also is an immediate dealbreaker for me for the exact same reason. And also #6, which rather makes things good only for people who shoot things into articles without trying to get a consensus: a normal policy-abiding Wikipedian who thinks his change may be controversial has to go through discussions, and someone who decides to just put his text into an article has suddenly immunity to reversion because lack of consensus is apparently not a reason to revert. No, I will not accept any of this, because I would like Wikipedia to remain Wikipedia. Double sharp (talk) 11:53, 4 November 2020 (UTC)

@Sandbh: I do not find that this protocol will aid the development of the project as it stands. I do appreciate the effort put into it but I think that many points need a greater level of precision before we could discuss them because they are very open for different interpretations.

2. "keeping in mind that the purpose of our rules and guidelines is to make the best encyclopedia possible" -- people will inevitably disagree on what this means. Say, there has been disagreement about that between you and me. This phrase will hardly help us resolve the differing views because it's rather vague, and you and I will fill it in differently.

3. "Being sharp, at times, does not constitute incivility" -- I, again, fear that this phrase lacks sufficient substance to make sure it means the same thing to all editors.

5. "Nobody owns any article, template, table, content or domain of knowledge." -- the same comment. What does it mean exactly? Is WP:OWN not enough?

8. I feel my previous concern has not been resolved. We are not supposed to focus on science, we are supposed to focus on reliable sources. Wikipedia:Reliable sources is a core policy of the English Wikipedia. We can't adopt a guideline that would contravene that. We shouldn't have such a guideline.

9. My concern has not been resolved, either. What is "undue WP:POLICY citing"? Again, you and I differ on what is undue policy citing and what isn't. How is this guideline going to help us?

One thing I agree with is explicitly stating that lack of beforehand consensus is not a sufficient reason to revert someone's bold edit unless there is an established consensus behind the current standing that such a bold edit would contradict.--R8R (talk) 14:12, 4 November 2020 (UTC)

While I applaud the efforts here, I find my self agreeing with R8R's concerns. IMO, many (if not most) of these proposals are problematic. However, I am not sure I agree with the protocol re reverting and R8R's comment on it. I am concerned that #6 could be read as an explicit rejection of WP:BRD, which though not a policy, I have found to be a very helpful guideline. It might be better to scrap this thread entirely. I think the site-wide guidelines are sufficient. YBG (talk) 15:42, 4 November 2020 (UTC)

@YBG: I think the site-wide guidelines are sufficient. +1, thank you. They should be sufficient indeed IMHO, as befits their being site-wide. The problem is that people have to respect them. With the general community understanding trumping whatever one thinks they mean when they disagree, befitting how WP is built on consensus. Double sharp (talk) 15:48, 4 November 2020 (UTC)

@YBG and Double sharp: There is generally no problem in restating policy as long as we don't contradict it. We can have it as a local guideline, too (whether we should, however, is another issue), to put some emphasis on a given policy/guideline if there is a desire to do that. As for reverting without expressing a particular concern with an edit and merely noting lack of beforehand approval, this is actually discouraged, too. WP:BRD: "BRD is not a valid excuse for reverting good-faith efforts to improve a page simply because you don't like the changes. BRD is never a reason for reverting. Unless the reversion is supported by policies, guidelines or common sense, the reversion is not part of BRD cycle."--R8R (talk) 16:08, 4 November 2020 (UTC)

When a GF bold edit replaces long-standing text with alternate text that is controversial, I think a revert followed a discussion is the right approach under these circumstances. I am concerned that #6 could appear to contradict this. But even if it does not, I still think it would be best to stick to site-wide norms and augment them only when there is an overwhelming agreement among project participants. By overwhelming, I do not mean unanimous. If everyone in the project but me agrees, that is clearly overwhelming. IMO, the most likely (and most fruitful) area for project-specific guidelines would be if some project-specific concern is unmentioned or ambiguous in the WP guidelines. But even that can be overcome, see, for example, WP:ALUM. YBG (talk) 16:47, 4 November 2020 (UTC)

@YBG: if an edit is controversial and you can identify in what specific ways it is controversial (other than simply being not what it used to be), then it's fine to call for a discussion as I see it and I don't see a problem with that. Each individual editor should recognize their opinion is not paramount, both the one who sees a potential problem (therefore admitting other editors may not share the concern) and the original editor (admitting the stated concern is probably worth raising).

The whole reason why I think we could have our own guidelines (that would supplement or even simply restate, rather than contradict, the site-wide guidelines) is this. While the site-wide guidelines are supposed to work and there's that, as we have seen throughout this year, they sometimes don't. Do we want to do anything about it? One response could be, we should do nothing, the site-wide guarantees are already perfect or as close to that as it gets. Unfortunately, this response would do nothing to help solve the problem. Another response could be, we could bring the behavior of an editor who violates the guidelines under the spotlight of ANI. I believe there is no real desire to have this as the final ending, either. So what else could it be? The thing that comes to my mind is having our own "gentlemen's agreement," or call it whatever you will. If we were to discuss what kind of behavior we wanted to see from other editors and what standard of behavior we were willing to stand by, this could have a positive impact on our future relations within the project. Even if the agreements themselves don't end up being all that fancy and instead add precisely nothing to the existing guidelines, the very process of negotiating what we particularly want other editors to do could help us a lot, because such a process could help point out how grievances are born and what could be done in order for this not to happen. The possibility to point out the results of such a discussion in the future if someone does violate the agreement, willingly or not, is also important.

@YBG, Double sharp, DePiep, Sandbh, and EdChem: if there are other ideas that could help us not end up in conflicts in the future, I'd like to hear it very much, and I'm sure others would like to hear it, too. There has been a number of conflicts this year. Do we want to help ourselves to stop the count for this year, and ideally forever, at whatever value it is right now, or should we do nothing and see if the problem resolves by itself? If we do want to be more active and help ourselves, what could such help be like?--R8R (talk) 13:21, 8 November 2020 (UTC)

@R8R: Another response could be, we could bring the behavior of an editor who violates the guidelines under the spotlight of ANI. I believe there is no real desire to have this as the final ending, either – indeed, I do not like having to go to this ending, but I always felt it should be on the table. If the general community interpretations of the guidelines are perpetually rejected in favour of one's own idiosyncratic ones, and despite lots of suggestions nothing changes – then yes, it becomes my desired ending because nothing else will work. If someone wants to stay on this site – follow the site guidelines. Anyone may write whatever he or she wants elsewhere where the guidelines are different. Double sharp (talk) 13:58, 8 November 2020 (UTC)

@Double sharp: I agree with what you wrote, but it doesn't really answer my question. I am looking first and foremost for a way to facilitate abidance by the rules, rather than for a way to alter them. I'm not looking for a way to get ANI off the table entirely, either, though I take it we should try our best not to end up there; only if we must. This was the sort of suggestion I got when I started an ANI, and I've internalized it, it makes perfect sense to me. What can we do in order to be able to say, "we've tried everything, and as much pain as this brings, there is no other way, because everything else has been tried, and I commend the latter statement"? "Follow the site guidelines": how do we stimulate that? I've proposed a suggestion: we talk it all through and all once again explicitly commit not to do what might disappoint others provided the agreement does not contradict in the general framework of Wikipedia policies and guidelines. To me, it looks like a good thing however you look at it: it should allow you to state your concerns with other editors, and allow other editors to hear you and tell you what they don't like about your behavior; it doesn't take ANI entirely off the table but also discourages you from pushing for it unless you absolutely must. Don't you think this would be the most collaborative and reasonable thing we could do, and that it would be better to try it? If you think it wouldn't help, why not?--R8R (talk) 14:51, 8 November 2020 (UTC)

@R8R:. Best step would be to improve discussion quality. When contributing, engage & respond, not soapboxing one-way talking. Stay on topic; one-topic-one-place; and be brief (long posts indicate immature thoughts). Keep threads & sideroutes organised. All as a collective (common) effort. Also, discussions are intended to develop into a common result. That implies: digest opposing arguments.

Whether we work by wiki-wide guidelines or local ones, this talkpage attitude is required. IOW, if such attitude is absent, any discussion will run into the sand or in the long grass, and makes frustrations easier to enter. Personally, I can say that a well-performed thread is easy to be loyal to. All this was very common before 2020, so it's not new to us. -DePiep (talk) 16:06, 8 November 2020 (UTC)

@Sandbh: Not to mention that even if we were supposed to focus on science, that huge argument from the early part of the year should have made it abundantly clear that what you think is correct science is assuredly not guaranteed to match what other editors think is correct science. Double sharp (talk) 15:21, 4 November 2020 (UTC)

I have another concern about adding project-specific guidelines in a small project such as ours. Whenever you have guidelines, there is the possibility of differing interpretations. If a dispute over project guidelines arises, our ability to find uninvolved project members to mediate is severely limited. But if we have a project disagreement on the interpretation of some site-wide guideline, it is much more likely that we can find uninvolved mediators who can resolve the issue. Even with the ugliness of our project's recent experience at ANI, I still prefer that forum to a free-for-all just amongst ourselves. YBG (talk) 16:47, 4 November 2020 (UTC)

@YBG: +1 Double sharp (talk) 17:13, 4 November 2020 (UTC)

DePiep says: No way. No reason or need to make separate guidelines for WP:ELEMENTS. Why should an editor have to sign up to any extra set? The issue is: the already available guidelines are not being respected. And why redefining "assume GF", "no PA", "no OWN"? duh.

I am not very familiar with the source issues at ELEM, but I am sure e.g. EdChem (skipping Double sharp here) recently has added good guidance on handling sources (again, without requirement for new guidelines).

TL;DR: No, no separate guidelines. -DePiep (talk) 23:15, 7 November 2020 (UTC)

@DePiep: +1. If it is consistent with pan-WP guidelines, there is no need to say it again, and if it is not, then one should not create an enclave on WP where pan-WP guidelines don't apply. Double sharp (talk) 23:28, 7 November 2020 (UTC)

Sit rep comment by Sandbh: I haven't recently commented or edited here pending the outcome of ANI #3. I may do so if I have some spare time. Sandbh (talk) 05:03, 9 November 2020 (UTC)

Straw poll

@Sandbh, Double sharp, YBG, EdChem, R8R, ComplexRational, and DePiep: Do you think we should spend more time on this subject? Please add your signature below without additional comment. YBG (talk) 08:26, 13 November 2020 (UTC)

Yes, I think WT:ELEM should spend time developing project expectations/guidelines

No, I think WT:ELEM should NOT spend time developing project expectations/guidelines

Comments re straw poll

YBG thoughts: If most of us think we should spend time on this, my recommendation would be that we have another straw poll to decide when we should do it, eg, immediately (eg to avoid ArbCom) or later (eg after ArbCom is complete). If on the other hand, most of us do not think we should spend time on this, then we should archive this thread and agree not to bring it up again for a reasonable period of time. YBG (talk) 08:26, 13 November 2020 (UTC)
ComplexRational comment: Clear expectations and guidelines (of course not forming a WP:LOCALCONSENSUS) would probably do well to clear ambiguity and misunderstandings, and enable more constructive content collaborations. IMO this talk page has gotten way out of hand, both in length and the nature of some discussions, and I'm astonished by the headbutting and lack of progress on perennial issues. ComplexRational (talk) 14:42, 13 November 2020 (UTC)
DS comment: I was originally of the POV that this was unnecessary because we already had global guidelines that should be applicable. However, since much of this seems to really be about how to understand global guidelines in the context of our situation, I have been persuaded by CR's rationale to support the idea. I agree with CR's characterisation of how things have gotten here and apologise on my part for having contributed to it. Double sharp (talk) 14:53, 13 November 2020 (UTC)
@Double sharp:, please add your signature to the count above. YBG (talk) 17:56, 13 November 2020 (UTC)
More from YBG I think this thread becanme a train wreck because it tried to bite off too much. If the majority of us decides we should spend more time on this, then I suggest that we tackle soemthing small, perhaps by (1) editors submitting potential guideline areas (2) a simple !!vote to decide which small topic area to take up (3) a discussion aimed at reaching consensus on that simple topic area, fiercely limiting scope. I have more to say, but I will wait until there is a majority of Yes !!votes in the straw poll. YBG (talk) 17:56, 13 November 2020 (UTC)

On the inevitable misunderstandings that arise from giving individual elements single category colourings

Textbook example. Double sharp (talk) 19:10, 9 November 2020 (UTC)

Isn't the misunderstanding in the "typical" there? (is 'misunderstanding' the right word here btw?). DePiep (talk) 19:30, 9 November 2020 (UTC)

@DePiep: Ah, maybe I should have made it clearer. In my view the misunderstanding was that the IP editor thought the statement that boron was a typical "non-metal" was an "error" and replaced "non-metal" with "metalloid". In actual fact, this is hardly true; there is hardly a consensus in the literature as to which elements are nonmetals, or even as to whether metalloids as a category should exist at all. So it is not an "error": boron can certainly be called a nonmetal and some sources even do. But when you see boron, coloured in on Wikipedia as just a "metalloid", called only that in {{infobox boron}}, called only that in Period 2 element, called only that in List of chemical elements; then I would be massively surprised if people did not start thinking that that was somehow a correct classification and that anything else is an "error". It is the most natural conclusion to get from what we show and it is also just wrong. That is my quarrel with colouring categories and it will remain forever unless the literature suddenly starts having a consensus on such things. (I do not feel that it should in the first place, but the important thing is whether it does or not, not whether we feel it should or not.)

Such a problem impacts pretty much all categories. Is it "lanthanides" or "lanthanoids"? Do actinides/actinoids include Ac, or not? Do transition metals include group 12, or not? Are they called post-transition metals or poor metals or something else? Are the metals even split the WP way, or perhaps by class A vs class B behaviour or perhaps by electronegativity? Are we splitting nonmetals by groups or something else? Do halogens necessarily include At? Ts? Is Og a noble gas by default, or does its chemistry have to be confirmed first? Are Be and Mg alkaline earth metals? When 119 and 120 are discovered, do they automatically become alkali and alkaline earth metals, or not? Are rare earth metals used? Can categories overlap? No consensus exists for any of this in the literature, so why are we trying to decide ourselves and fly in the face of WP:NPOV?

I am of the opinion now that the colouring of our PT in the first place by our much-admired pioneers at this project was the "original sin" here. If it had not been instituted so early, then I suspect it would never have gotten up due to WP:OR concerns. Now instead the uphill battle is to dislodge it. And this is coming from someone who actually does feel personally that it's an easy task to define metals and nonmetals and make it clear which elements are which, and only does not support doing it for WP because there's no agreement in the literature for how to do it.

Well, hopefully that explains things better. ^_^ Double sharp (talk) 20:31, 9 November 2020 (UTC)

TL;DRall. 1. Not just the infobox boron; the infobox facts 'must' be pulled from the article body text. You say boron is not considered metalloid "at all"? 2. Also in Lists_of_metalloids (166/194=86%) i.e., ~not enwiki/OR. 3. If we were to drop the nine categories, there still would be the super-three categories metal-metalloid-nonmetal. Little attention here on enwiki, but major -- while retaining catergory-border-elements issues I agree. 4. There is also the ??wiki option, graphically nice too: (todo: find two-color-thatching PT, I saw recently). 5. Please don't say "coloring metalloid", it's "classifying metalloid" -- colors just follow. 6. I see your campaign to get rid of cat/colors, but in this we still could get the (metalloid) facts right. -DePiep (talk) 21:07, 9 November 2020 (UTC)

@DePiep: The lists of metalloids article only lists sources that consider "metalloids" to be a separate class from metals and nonmetals. Not all do that. I recall User:Droog Andrey saying something about most Russian textbooks not doing that. Those who consider there to be only metals and nonmetals disagree as to which elements are which. Those who consider there to be metals, nonmetals, and metalloids also degree as to which elements are which. So those super-three categories are actually just as as ill-defined as the nine. To mention them? Sure, I'm fine with that and would do it myself. To specifically colour things in so that it looks like their boundaries are clear in the literature, not fuzzy as they really are? I'm not fine with that.

I guess you mean the dewiki option. The problem is that for many elements in the p block you would need not just two but at least three colours among what we currently have (is Se a metal, a nonmetal, or metalloid? you can find sources saying each). Not to mention that our categories do not at all exhaust the categories that have been in the literature. Like I keep saying: where oh where are the pnictogens and chalcogens, IUPAC-approved, even (whereas metalloids are not IUPAC-approved)? Add that and we instantly bump Se up to four. Add in all the splittings of the nonmetals Sandbh has been collecting for his proposed RFC and we can bump it up probably all the way to a dozen. I do not think this is in any way a sustainable solution.

Meanwhile the astronomy editors have come up with an easy solution. Look for example at 90482 Orcus. Well, it is another minor planet, it belongs to many categories. The astronomers have not sought a taxonomy such that each minor planet only appears in one category, just like the chemists have not. So the astronomy editors have, sensibly enough, not tried to create something like that. Under "Minor planet category" we read "TNO [1] · plutino [4][5] possible dwarf planet [6]" with all possible categories in sources listed. And all three are mentioned in the article body text (indeed, in the lede even).

That's what I would like. Categories (which will quite possibly have to be renamed also; where is the source for "category" with the WP meaning?) – as many as you like as in the literature, they can go in the infobox. Antimony can be listed as "metalloid (most common), sometimes metal or nonmetal · pnictogen · post-transition metal (if considered a metal)" and so on and so on. Gold can be listed as "transition metal OR (rarely) post-transition metal · noble metal · platinum group metal · coinage metal". Lutetium can be listed as "transition metal OR inner transition metal · rare earth metal · lanthanoid (lanthanide)". Astatine can be listed as "metal OR metalloid OR nonmetal · halogen? · post-transition metal (if considered a metal)". And all of these different categories can be mentioned in the article. I just do not want this rich situation to be distorted into one colour for each element and just one category. Out of a cherry-picked nine when there are many more in the literature. Especially when the nine are trying to be jointly exhaustive and mutually exclusive and even then fail at doing it (what are the heavier members of group 3? they're TMs but the tables don't make it clear). And the nine being grouped into three supercategories (metal/metalloid/nonmetal), when many sources in the literature (particularly the Russian ones) give only two (metal/nonmetal). In my opinion, our current PT colouring is an OR/SYNTH/UNDUE festival that should stop. Double sharp (talk) 22:47, 9 November 2020 (UTC)

@DePiep and Double sharp: We've talked previously about showing multiple categories for each element in the info boxes. So, scandium e.g., as well as being a transition metal, would also be listed as a light metal, and a rare earth metal. Sandbh (talk) 00:47, 12 November 2020 (UTC)

@Sandbh: And how exactly one could combine three colours that way is about as clear as mud. Added to the point that not all books consider Sc a transition metal, that "light metal" and similarly "heavy metal" as you well known have no completely-agreed on definition, and rare earth metal does not actually always include Sc. Double sharp (talk) 00:52, 12 November 2020 (UTC)

@DePiep and Double sharp: There's no need to combine three colours. One shows the colour category box, plus the extra category memberships in text only. I suggest most books consider Sc to be a transition metal; it is more common that group 12 are not considered to be TMs. For light metals and heavy metals we could, for example, say "light metal(?)" to denote ambiguity or light metal to denote a "sometimes" status. It's not something I've given much thought to. Sandbh (talk) 02:02, 12 November 2020 (UTC)

This thread merits further consideration.

On the status of B there is this article (by me), Which elements are metalloids? [8] with 29 GS citations. It appeared in the Journal of Chemical Education and concluded that B is one of the six elements that are commonly recognised in the literature as metalloids. There has been no dissent across the 29 citations, as to my conclusion.

More to follow. Sandbh (talk) 22:28, 9 November 2020 (UTC)

We note in our periodic table article that, "Different authors may use different categorisation schema depending on the properties of interest." To this end. I've added an other categories section with a w/link to our article on names for sets of elements.

I know of no RS supporting the premise of this thread, i.e. "the inevitable misunderstandings that arise from giving individual elements single category colourings".

None of the authors of the following colour category PTs appear to think this is a problem:

the Encyclopaedia Britannic PT, which predates the WP PT, features nine colour categories
the most popular Google PT, which postdates the WP PT, features nine colour categories
Gonick & Criddle, in The Cartoon Guide to Chemistry (2005, p. 38) refer the reader to the LANL PT which they describe as "a wonderfully information-rich" PT. This has ten colour categories.
The ACS PT has ten colour categories.[9]
The RSC PT has eight main group colour categories; one TM colour category; and one each for the Ln and An.[10]
The World Book encyclopaedia of science: Chemistry today (1986, p. 23) has a PT with 20 colour categories. In the text, they "group: the following elements: H; AM, AEM; coinage metals; AEM; group 12; group 13; C; Si–Pb; Ti–Ni; Zr–Tc and Hf–Re; PGM; N; P–Bi; O; S–Po; halogens; rare gases; Sc group and the Ln; An.

Since the days of the classical elements of the ancients, humanity has a proclivity to categorise things. Categories demand explanations which may lead to new or refined categories which demand explanations etc—that sort of iterative process is at the heart of science [11]

We strive to cover significant viewpoints in the literature; that would encompass: AE,* AEM,* Ln, An; TM; metalloids, halogens,* noble gases* (* = IUPAC approved). That leaves the group 12–17 metals and the group 1 to 16 nonmetals. WP:ELEM chose PTM for the group 12−17 metals; WP:ELEM chose other nonmetals (2003-2013), and then poly/diatomic metals (2013-2018), and reactive nonmetals in 2018. WP:OR has been around, in one form or another, since December 2003; I can't recall WP:OR concerns having been raised during this time.

Categories are rarely sharp (Jones 2010, p. 170; Atkins 2013, p. 3). There are all sorts of fuzzy boundaries in chemistry https://www.newscientist.com/article/mg21128234-100-redefining-hydrogen-bonds-the-givers-of-life/]. Needless disputes arise wrt fuzzy boundaries [12]

--- Sandbh (talk) 03:55, 10 November 2020 (UTC)

@Sandbh: Of course they do not think it is a problem. They are giving their own viewpoints. We on Wikipedia, on the other hand, are supposed to reflect the situation in the literature, in which no single viewpoint predominates. The disagreement in categorisation between the colour-category PTs you show speaks for itself. And I see there's more synthesis in your rationale with yet another reference to Jones, which as I mentioned is a source primarily about the classification of Pluto and does not mention classifying chemical elements at all. Double sharp (talk) 08:38, 10 November 2020 (UTC)

@Double sharp: Wikipedia is no different than any other encyclopedia. Those other encyclopaedias are not giving their own viewpoints; rather, as encyclopaedias, they are giving their viewpoints of what the literature says i.e. an encyclopaedic overview of each topic. This is what we do, too. Of course there will be differences in coverage given the involvement of different editors.

Jones was discussing a principle of classification science. Such a principle applies across domains of science, including chemistry and e.g. biology. I cited Jones in my peer-reviewed articles on organising the metals and nonmetals [13] and the location and composition of group 3 [14]. Jones will be cited again in a forthcoming peer reviewed article: Cao C, Vernon RE, Schwarz WEH, & Li J, Chemistry in periodic tables. Frontiers in Chemistry. [15]. Sandbh (talk) 11:18, 10 November 2020 (UTC)

@Sandbh: So why is it that four of the six sources you give are not encyclopaedias? And what does it matter when it's quite easy to demonstrate by looking at other sources that what even the two encyclopaedia sources give is not a good reflexion of what the literature as a whole agrees on (because there is no such thing)? Double sharp (talk) 11:26, 10 November 2020 (UTC)

@Double sharp: I guess, I don’t know for sure, that we strive to reflect the consensus in the literature, noting consensus does not require unanimity. With the metalloids for example, the consensus in the literature is that B, Si, Ge, As, Sb, Te are the six elements most commonly recognised as metalloids. That said, we note the variations at the boundaries in the metalloid article.

There is an essay WP:NOTSYNTH which touches on aspects of this. For example:

"This essay is intended to help explain the spirit of that policy. In particular, this essay is intended to oppose taking an excessively strict interpretation of the policy in many cases. After all, Wikipedia does not have firm rules."
"SYNTH is not a rigid rule. Wikipedia doesn't have them, supposedly. But if a policy gets enforced zealously, it can be hard to tell the difference. The solution is to not enforce policies zealously. Never use a policy in such a way that the net effect will be to stop people from improving an article."
"What matters is that all material in Wikipedia is verifiable, not that it's actually verified. By this we mean that it is important that a suitable reliable source that supports this material has been published in the real world, not that someone has gotten around to typing up a specific bibliographic citation in the article.
”Citations are not an end in themselves. If there's a statement for which no source is cited, that's normally ok, as with the example on Wikipedia:No original research: "Paris is the capital of France" needs no source because no one is likely to object to it, but we know that sources for it exist. Likewise with very many unsourced statements, regardless of whether they could be deduced from sourced statements in the same article, we know the sources exist."

While these extracts don’t directly address your concerns I suggest they illustrate the principle/s involved.

--- Sandbh (talk) 22:25, 10 November 2020 (UTC)

@Sandbh: (1) That's only among the sources that do consider metalloids to be separate from metals and nonmetals. (2) While maybe among those sources those six are common inclusions, already there is fuzziness with the next tier of Po and At with half saying yes and half saying no. So where exactly is the supposed consensus in the literature for the classification of those two? (3) The issue with application of policy and rigidity (that statement is already going to the WP:IAR) has been discussed endlessly both here and at the WP:ANI thread about you.

The problem with those words "improving an article" is that people will disagree endlessly about what is and what is not an improvement. Frankly I am astonished that I keep having to point out that different editors may have legitimate different perspectives on what constitutes an improvement to the article, when we spent the better part of the year arguing over which presentation of group 3 would improve WP. That's precisely why we have such a thing as WP:CONSENSUS and other policies. It takes consensus to determine if WP:IAR applies, not just one editor saying so. You've stated your case, I've stated mine, now the important thing is how many are persuaded and how policy-compliant the arguments are (of course, "colours because WP:ILIKEIT" or "no colours because WP:IDONTLIKEIT" will get nowhere, for example). And in practice, if WP:IAR is being applied against really core policies, such as WP:NPOV as I see here, then discussion with User:EdChem and others through all these ANI threads strongly suggests to me that it will get nowhere. Double sharp (talk) 22:36, 10 November 2020 (UTC)

@Double sharp: On the ubiquity of metalloids in the literature I turn to four sources. The first is IUPAC and its three unsuccessful attempts to recommend against the use of the term metalloid to refer to the in-between elements. All that happened is that the prevalence of the use of the term metalloid, in the literature, skyrocketed. The other wisdom is that of the big three, namely C&W, G&E, and Wiberg, all of whom recognise metalloids.

The consensus in the literature is that the elements most commonly recognised as metalloids are the usual six. For Po and At we know from the lists of metalloids, that they appear about half as often. I've been watching the recent literature, and it seems to me that their appearance frequencies as metalloids are reducing, following the publication of my article on Which elements are metalloids? in JChemEd, this very article arising out of the chaos that use to be our metalloid article, now TFA.

I agree with you wrt the rest of what you said, with one caveat. Consistent with WP:BOLD, an editor can edit and invoke WP:IAR. They're then subject to BRD, of course. This may or may not resolve the issue, and may subsequently result in the need to establish consensus on whether the invocation of WP:IAR was (in which the case edit stands) or was not reasonable (in which case the R stands). Sandbh (talk) 00:24, 11 November 2020 (UTC)

@Sandbh: So, please tell me exactly how consistent that big three (there are other books around of course) are on exactly which elements are metalloids. I am pretty sure G&E was not even internally consistent on it the last time I checked. And please tell me exactly what you think might be a reasonable consensus categorisation of polonium and astatine in this scheme. I really wonder what could possibly be one when there is literally pretty much a half-for and half-against for the question "are they metalloids". And please tell me how much you think non-English sources should count; I recall from conversation with User:Droog Andrey that in Russian sources it is not common to use metalloids. After all, IUPAC is international, isn't it? Incidentally if we take that seriously, alkaline earth metals becomes seriously more iffy because we then have to consider Japanese and Russian sources where Be and Mg are way more often excluded than they are in English! ^_^

P.S. Why is it that the recent literature is important when it comes to pushing Po and At out of the metalloids box, but when opposing a shift of group 3 to Sc-Y-Lu the 4:1:1 figure keeps getting trotted out based on textbooks all the way back to the 1970s? And exactly how much are those frequencies for Po and At reducing? And why is such a reduction that happened after one of your articles got published relevant here, but we heard nary a word from you at Wikipedia talk:WikiProject Elements#Details of Sandbh's opinion and ensuing thread about the reduction of frequency of Sc-Y-La from over a 3/4 supermajority all the way to just losing a majority since Jensen's 1982 article (and those of many others) was published? And how do you know it had something to do with your article when there was in 2010 also another one promoting the idea that Po and At are not metalloids? Did some source say so, like Scerri said about Jensen's article being an important one in the history of Sc-Y-Lu promotion? Double sharp (talk) 00:30, 11 November 2020 (UTC)

@Double sharp: We talked about G&E, and C&W if I recall right, about their views on metalloids. I don't know where that thread is. For polonium, as far as we know, it has the band structure of a fully fledged metal, and does not exist in a semiconducting form. We also know that it relatively easily forms a cation in acidic solution. Smells like a metal to me. Astatine has been predicted to be a fully fledged metal per Andreas Hermann, Roald Hoffmann, and N. W. Ashcroft,[16] now with 35 citations and no dissent. When it was synthesised in 1940 its discoverers considered it to be a metal. Smells like a metal to me.

This is the the English WP, so I'd ordinarily expect English sources to have priority.

The Russian PT includes metalloids.

My comment about more recent sources wrt to Po and At was an observation that, with 30 citations, my article on metalloids seems to have been noticed and recognised as somewhat authoritative. (I could be deluding myself of course, with my perceptual filters. I try to bear this in mind.) Compared to the elements commonly recognised as metalloids, Po and At never had a chance anyway. The Lu option remains well outnumbered by La, as R8R has noticed. In both cases we note these runners-up in the respective articles.

I corresponded with Stephen Hawkes, the author of that 2010 article on Po and At, before he died. He was a thorough gentlemen. Unfortunately his article garnered ten times fewer citations than mine, and had a very narrow scope. Whereas mine was comprehensive and relied on the lists of metalloids, with its sample size of 194; the "gold standard" as YBG put it. I cited Hawkes.

To my knowledge, no one has said anything about my article on metalloids being e.g. "important" aside from it being cited 30 times.

--- Sandbh (talk) 05:04, 11 November 2020 (UTC)

@Sandbh: No one is saying that Russian tables never include metalloids, just that they often do not.

@Double sharp: If that is so, why does the Russian WP table include metalloids?

@Sandbh: You'd better ask Russian Wikipedia editors about that, not me. I do wonder how many WP's simply copied our original colour scheme. Double sharp (talk) 00:56, 12 November 2020 (UTC)

Yes, polonium smells like a metal to me too, but the question is whether or not the literature is really unified behind that! Double sharp (talk) 10:59, 11 November 2020 (UTC)

Unified? No. Majority opinion? That's my impression. Sandbh (talk) 00:39, 12 November 2020 (UTC)

@Sandbh: Then exactly how do you suppose colouring Po as only a PTM and not also somehow as a metalloid is a valid reflexion of the sources? Not to mention that it also takes part in other common categories that WP somehow decided not to include like chalcogens and maybe heavy metals depending on who is defining that one? Double sharp (talk) 00:54, 12 November 2020 (UTC)

Such concerns are usually addressed in the legend. Sandbh (talk) 02:07, 12 November 2020 (UTC)

An article on the use of colour in periodic tables: [17]. Sandbh (talk) 10:41, 10 November 2020 (UTC)

I have never disputed that colours are often used, and often for categories. It is just that there is no standardisation whatsoever in the literature as to what colours are used and what categories they are used for. Double sharp (talk) 11:19, 10 November 2020 (UTC)

@Double sharp: I don't understand. Who said that you disputed "that colours are often used, and often for categories"? Sandbh (talk) 00:39, 12 November 2020 (UTC)

@Sandbh: I don't mean you did. I mean that your articles are not addressing my point. Double sharp (talk) 00:51, 12 November 2020 (UTC)

@Double sharp: I didn't interpret your comment to mean "I" did, since I knew I hadn't, so to speak. I presumed you were referring to someone else. Which articles of mine did you have in mind, and what was it about them that didn't address your point(?), not that I wrote them with that end in mind. Sandbh (talk) 02:14, 12 November 2020 (UTC)

A related article: Looking for an order of things: Textbooks and chemical classifications in nineteenth century France [18]. Sandbh (talk) 04:37, 11 November 2020 (UTC)

@Double sharp: I want to address some questions you raised earlier:

1. Is it "lanthanides" or "lanthanoids"?
IUPAC recommends the latter; my impression is that lanthanoids is more popular in the literature including as informed by the ngrams.

2. Do actinides/actinoids include Ac, or not?
When the Ln were originally dubbed thus, it did not include La, since the intended meaning was "like" La. Over time the terms lanthanide and lanthanoid seem to have suffered scope creep, so that they know include La. Certainly that is the way IUPAC interprets it. By analogy therefore Ac is included; ditto IUPAC.

3. Do transition metals include group 12, or not?
It seems to be about 50/50. Even IUPAC recognises something along this line.

4. Are they called post-transition metals or poor metals or something else?
PTM seems to be the most popular term, judging from the literature and ngrams.

5. Are the metals even split the WP way, or perhaps by class A vs class B behaviour or perhaps by electronegativity?
The WP way seems to be popular, as a general way of splitting the metals. The stand out spit seems to be between the inner and outer transition metals. All that's left after that is the AM, and AEM, and frontier metals.

6. Are we splitting nonmetals by groups or something else?
Historically, both i.e. other nonmetals; halogens; noble gases; then polyatomic/diatomic/noble (monatomic); then reactive nonmetals/noble gases. We do the same with the metals.

7. Do halogens necessarily include At? Ts?
IUPAC recognises At as a halogen. It has made no ruling about Ts. The halogens are comprised of four nonmetals and one metal; so to are e.g. the chalcogens comprised of two nonmetals, two metalloids, and one metal.

8. Is Og a noble gas by default, or does its chemistry have to be confirmed first?
IUPAC has made no ruling. The feeling in the literature, as we mention in out organneson article, is that Og is not expected to be a noble gas. It's too soon to call.

9. Are Be and Mg alkaline earth metals?
The consensus in the literature is that they are, even though Be is amphoteric rather than alkaline, and Mg was able to be isolated from its oxide in impure form, via the heating technology of the day, even though 'earth' referred to oxides from which the metal could not be isolated.

10. When 119 and 120 are discovered, do they automatically become alkali and alkaline earth metals, or not?
By default, they probably will be referred to as such, at least at first.

11. Are rare earth metals used?
We don't although EB do. REM seem to have a much greater profile in popular media so WP is arguably out of touch in that regard.

12. Can categories overlap?
They usually do, especially given the fuzzy nature of chemistry.

13. No consensus exists for any of this in the literature, so why are we trying to decide ourselves and fly in the face of WP:NPOV?

Consensus (majority opinion, not unanimity) appears to exist for all of these things in the literature. On Og we tread cautiously and show it as unknown properties; presumably we will do the same with 119, 120. Encyclopedia Britannica, the World Book Encyclopedia, RSC, ACS, and LANL take similar approaches, allowing for the fuzziness of chemistry, which we acknowledge in our periodic table article. So too does the most popular online PT [19] itself modelled after WP.

@Sandbh: Consensus is not unanimity, but for it to be the only view presented it must be clear dominance, which by your sources in your previous comments clearly does not exist. Even the table you call "modelled after WP" doesn't exactly match WP: for one thing, when you expand it to a wide table it shows Sc-Y-Lu, and for another thing, it considers group 12 to be transition metals. Otherwise, you've got to reflect all major POV to avoid WP:UNDUE. Which a single colouring will never be able to do. And of course Q6 fails to answer the question, which is about the literature, not what WP made the call to do as it shouldn't have. Oh, and if you think categories can overlap (Q12), then why do you not propose and not point to a single category scheme that reflects this?

P.S. And now "frontier metals" is presented without comment in these answers. Looking at Post-transition metal reveals that this is a term from your own journal article. I really wonder who else will understand it without further explanation. I am glad that it is just being mentioned and not promoted, but do we have to cloud the issue when the primary thing is what goes into the articles, and not everyone is aware of the zillion category names for this approximate set of elements? Double sharp (talk) 00:21, 11 November 2020 (UTC)

@Double sharp: Our PT represents a form of words in graphical form. If a reader, cookie-cutter stye, chooses to focus on those words alone, absent of their context, that is their choice. We note the otherwise missing context i.e. the fuzziness of the categories in the periodic table article. I've suggested adding a small note to the graphic to this end and have not done so, due to opposition expressed here. I'll draft an RFC about this, and post it to the periodic table talk page. On Q6, WP:ELEM chose those spilts, IGF, via consensus, on the basis of our understanding of the literature, as did the founders of the WP colour category table.

I've proposed overlapping categories for the WP PT, only to meet staunch resistance from colleague here. I recall you even pointed out one of these tables on another of the WP's, the German one if I recall right?

I looked at the 149, by my count, non-English WPs. I found three showing just metal-metalloid-nonmetal; and two block-only tables. So 96%+ show colour categories, such as we do.

Cool tables:

Arabian(?), with its inclusion of block-only icon.
French table with something similar;
German, with its hybrid colourings of Ge, Sb, Se, and At.

I casually mentioned frontier metals in passing, as a way of referring to the metals that occupy "frontier territory: abutting the metalloids. The term itself comes from a peer-reviewed open-access article in the literature (that I wrote), now with 5,400+ access. As the PTM article notes, the name frontier metal is adapted from Russell and Lee, who wrote that, "…bismuth and group 16 element polonium are generally considered to be metals, although they occupy 'frontier territory' on the periodic table, adjacent to the nonmetals."[20]. --- Sandbh (talk) 03:55, 11 November 2020 (UTC)

Time out (1)

This discussion is unclear about its central question or postulate. It is unuseful because of its unstructured and chaotic flow & layout. It is unhelpful because of extremely long texts that do nothing to make an argument or a point. Unless restructured, we can leave it alone without consequences. I call it basically senseless. -DePiep (talk) 17:24, 12 November 2020 (UTC)

@DePiep: OK. Double sharp (talk) 18:25, 12 November 2020 (UTC)

Eh, nothing personal, nothing wrong with your early contributions Double sharp, even when lengthy. This is about the overall thread so far. Have a nice edit. -DePiep (talk) 00:34, 13 November 2020 (UTC)

@DePiep: Oh, that's good to hear, but I do think I was a little too verbose. I admit I found it difficult to formulate succintly – but I should have tried harder. Have a nice edit too. ^_^ Double sharp (talk) 00:40, 13 November 2020 (UTC)

NO! my statement is: you did nothing wrong, it's the thread itself that exploded! DePiep (talk) 00:44, 13 November 2020 (UTC)

@DePiep: Oh! Okay, I understand what you mean now. Thank you. Double sharp (talk) 00:47, 13 November 2020 (UTC)