User talk:YBG/Archive 5

Archive 1

←

Archive 3

Archive 4

Archive 5

Archive 6

Another small question

@Double sharp and Sandbh: I have a few suggested refactoring of your exchange in WT:ELEM § Opinions

1. I think Double sharp's narrative would be better presented by changing his {{cquote}} to include |width=80%. It would make the flow between point 5 and point 5 (cont.) clearer.

If I have DS's permission on this, I will make the change.

2. I think it would be better if the bulk of the discussion between you two were moved to WT:ELEM § Moved from Opinions to Supplements. I think the best way would be to retain Sandbh's initial points 1-4, and follow them by

Details moved below by YBG <date> (with a wikilink directly to the exact location of the details in Supplements)

Response by Double sharp moved below by YBG <date> (with a wikilink directly to the exact location of the response in Supplements)

If I have DS's and S&bh's permission, I will make this change. Perhaps there is a better place to break; I am open to suggestions ... but IMO this leaves gives the opinion the forcefulness of brevity with the advantage of hyperlinked details. If you will look at what I have already done for EdChem and Double sharp in WT:ELEM § Moved from Opinions to Supplements, you will see the sort of thing I have in mind.

Cheers! YBG (talk) 07:47, 22 October 2020 (UTC)

Fine with me! ^_^ Double sharp (talk) 08:49, 22 October 2020 (UTC)

Good for me too! Sandbh (talk) 09:38, 22 October 2020 (UTC)

OK, here is what I've done

rev numbers & diff	time stamps	Changes made
special:diff/984808102 /984941123	[a] 2020-10-22 06:43:04 to [b] 2020-10-23 00:44:19	Changes made by Sandbh & Double sharp, including eliminating the # numbering
special:diff/984941123 /984946494	[b] 2020-10-23 00:44:19 to [c] 2020-10-23 01:30:57	Changes I made to restore the original # numbering
special:diff/984808102 /984946494	[a] 2020-10-22 06:43:04 to [c] 2020-10-23 01:30:57	Comparison to before numbering was changed
special:diff/984946494 /984947422	[c] 2020-10-23 01:30:57 to [d] 2020-10-23 01:39:59	The actual move
special:diff/984947422 /984964950	[d] 2020-10-23 01:39:59 to [e] 2020-10-23 04:20:53	Changes I made after the move

YBG (talk) 04:45, 23 October 2020 (UTC)

Periodic table categories in plural

Could you take a good look are the proposal text in User:DePiep/sandbox2? Should be convincingly strong, but it's long too. Edit if you want to.-DePiep (talk) 05:53, 19 February 2019 (UTC)

I agree with all of your proposed change-to-plural articles. I would be inclined to change more things, however, including "nonmetals" and "metalloids" because these words are in fact used predominately as classes. The metal article stays put because that article is not about the class of but only because that article is used more frequently in common English as a mass noun (a bridge made of metal, a metal cabinet) and not referring to the specific class of ~90 elements. Metalloid and nonmetal, however, are used predominately as class names, although there are exceptions, e.g., see User talk:YBG/Archive 4 § Re nonmetals.

I would also be inclined to change the name of the other named groups, e.g., halogens, pnictogens, calcogens, noble gases, etc. But I'd be fine with changing the most uncontroversial ones now and leave those that need more thought for a later time. So I suppose this is a plea to write the proposal in such a way that it does not categorically state that other things should not be changed.

By the way, regarding the previous suggestion to change group/period article titles (Talk:Group 3 element § Requested move), I think I now agree 100% with the proposal that I rejected then.

YBG (talk) 06:42, 19 February 2019 (UTC)

I wrote "Should be convincingly strong" for two reasons. First is that the concept of class (class (disambiguation), most of these concepts, but I'd pick class (set theory) first). What went wrong in Talk:Group 3 element#Requested move (2013) is that such a discussion cannot enforce !voters to understand and then apply the concept. Then, the closing admin turns to some ""vote counting" when enough people wrote "but I've seen it used in singular once".

Background: Asking the test question "Is instance of [class]?" (popularly: "Is a ... ?") is called "orthography (of Wikidata)". It is applied thoroughly in Wikidata to determine whether a Wikidata-item (QID) is member of a certain class. The property appears as "Instance of [class]" (instance of (P31)). For example chemical element (Q11344) had :d:this discussion. At the moment, all elements are "Instance of chemical element", not part of (P361) (e.g., carbon (Q623)).

So I want the discussion to only be about this is-instance-of-class, not "is part of item".

Second, keep this focussed on not being a list but classes. Not introducing distractions. I stroke all dual use titles (metal/s, pnictogen/s, period 1 element/s). Groups and periods are class: may follow later. Enwiki categories first.

• How to proceed?
  1. About "metalloids, nonmetals": sure these are border cases, I left them out because of distraction (next to "metal"). We could put them in, if strong enough (I don't want the proposal to fail because of such detail discusissions, but maybe these are not details just main causes too).
  2. The proposed text may be too long. What can we cut? Add Red Book as an argument not nom text?
  3. Introduce groups like pnictogens (already in the Red Book): same question, can we keep this to class plurals? Would shift away from "categories".
  4. Major issue: shouldn't we make this a WP:ELEMENTS guideline first to discuss: "Predominantly class = plural (in titles and in body texts)"? -DePiep (talk) 09:46, 19 February 2019 (UTC)
• See, this is what we are up against. People who do not know or explore the concept of "class", and still do !vote. Same as in 2013. -DePiep (talk) 08:18, 20 February 2019 (UTC)

What would be the chances when we throw both steps in one formal proposal (as a WP:ELEM GUIDELINE): "article names & legend texts are plural, with exceptions: metal, sentencing 'sodium is an alkali metal'"? Too complicated? -DePiep (talk) 11:12, 20 February 2019 (UTC)

@DePiep: A few random comments - probably not exhausting what you've written.

• Where you say Asking the test question ... left me wondering what is the "test question" that one would ask? It sounds like you're talking about something like "Does ____ have hair?" as a test question to determine whether something is a mammal, or "Does it make sense to say 'Here are 33 _____s" to determine whether a thing is a count noun or a mass noun
• I looked at a couple of the WikiData discussions you posted and read a screenful or so and gave up.
• I've completely missed the point where you talk about orthography. I've always thought of that term as having to do with writing systems, not with classification systems.
• A minor nit:

  You write: So I want the discussion to only be about this is-instance-of-class, not "is part of item".

  IMO better: So I want the discussion to only be about this is-instance-of-class, not "is-part-of-collection".

• You write The proposed text may be too long, but I'm not sure what all is included in "the proposed text".
• I'm not sure what the difference is between "class" and "category".
• I like the idea of WP:ELEM/Guidelines. But it might not be the best thing to have as a first guideline.
  • Over the course of years of discussion, there are in all likelihood a number of guidelines that we've informally established, but never recorded. It would be good to document such already-informally-established guidelines on a single page.
• The proposed guideline should read "Classes of elements should should be rendered in the plural in legend text and article titles."
• I'm not sure whether it would be better to go for the big-bang approach of putting both steps into a single proposal or to go the more incremental fashion.

YBG (talk) 04:57, 21 February 2019 (UTC)

replies:

re "Orthography": yes, is used to describe a language. The "mass noun check" you mention is part of orthograpy too indeed.

And is also used in Wikidata to describe a data structure (for example: a "chemical element" is a class of isotopes or a "material appearance"? -- gold book says both BTW). That is what these difficult discussions are about. (I left them too, as you did). I mention this, because it is the Wikidata-philosophers way to check "is it a class or an object?". And: "Sodium is an instance of [class] alkali metal(s)"? (or: "Sodium is an alkali metal?")

I think our categories, when approached like this (asking the right questions), are a class definitely. Therefor plural at enwiki (not in Wikidata!).

Yes should be about [not] is part of collection OK.

"Proposed text too long": may be disracting. For the formal proposal, I will reduce the list-clarification (why are some in/out).

'difference between "class" and "category"'?: category=what enwiki uses for our nine metallishness sets, "class"=the generic word.

WP:ELEM/Guidelines: as you say. Later.

OK I will include metalloid/s and nonmetal/s, but not metal/s.

Will ask Sandbh to reconsider, and refine the proposal. You can edit my sandbox2 if you want to. -DePiep (talk) 10:36, 26 February 2019 (UTC)

All sounds good except I'm still puzzled by the use of the word "orthography". The lede in Orthography says (with emphasis added)

• An orthography is a set of conventions for writing a language. It includes norms of spelling, hyphenation, capitalization, word breaks, emphasis, and punctuation.

wikt:orthography gives four definitions:

1. The study of correct spelling according to established usage.
2. The aspect of language study concerned with letters and their sequences in words.
3. Spelling; the method of representing a language or the sounds of language by written symbols.
4. (architecture) Orthographic projection; especially its use to draw an elevation, vertical projection etc. of a building.

As none of these seem to be aligned with what you are talking about, perhaps WikiData uses the term in other less-standard manner that hasn't been added to wiktionary. YBG (talk) 16:53, 26 February 2019 (UTC)

@DePiep: PS, I really love the phrase our nine metallishness sets YBG (talk) 16:56, 26 February 2019 (UTC)

Yet another point. If we are using the red book as an authority, we should probably move rare-earth element to rare earth metals. This, however, should probably be treated as a completely separate issue. It does, however, make the citation of the red book seem a bit less authoritative. YBG (talk) 17:15, 26 February 2019 (UTC)

Living Prime Ministers of Australia

Why did my account get randomly blocked with no warning? I had to make this account to see what happened. I have no affiliation with that sock puppet account. All of my work has been deleted. I spent hours on that. :( LR.2004 (talk) 10:42, 16 August 2018 (UTC)

1st of all, take a deep breath. All is not lost, your work can be resurrected. In a few days or weeks, when I have more time, I will make the appropriate requests to get your work on this template restored.

2nd, this might be a good time to take a wikibreak and concentrate on your off-wiki life.  I have begun a habit of doing this once a year for a month or so, and it has helped significantly in improving my life, both on-wiki and off-wiki.

3rd, please assume good faith on the part of the person who blocked you and deleted your work. Fighting against unconstructive editing on WP is a tireless and often thankless job. We should be grateful for those who do this work, even with the occasional overzealous actions.

Since I do not have access to any special administrative or investigative powers, I can only guess why your account was blocked and your work deleted. But I can assure you that it was not random and not capricious.

I suspect that you are editing from a shared IP address which has been used for unconstructive editing. Sometimes under these circumstances, not only do specific registered users get blocked, but also an IP address or address range can be blocked. It appears that multiple accounts have been used to make similar edits from the same IP address or address range, and a sock puppet investigator concluded that these multiple accounts are being used by the same individual in violation of wikipedia policies, which are quite strict about about enforcing the very limited circumstances under which it is permissible for a user to use multiple accounts.

It looks like previously blocked accounts have been used to edit articles about Australian politicians and railroads. Some of your edits seem to match this profile. This could be for a number of reasons - maybe you are the same individual, maybe you belong to a school or club where lots of people have similar interests, maybe you left your account logged in on a public computer. It may even be that you have made unconstructive edits in the past, but have now decided that you prefer making constructive edits.

Whatever the case may be, I really appreciated your effort to create a living Aussie PM template, and enjoyed collaborating with you.

I hope this setback does not discourage you. I encourage you to continue gathering the information needed for this template. When it is complete, you could send it to me using the "Email this user" feature, described at Help:Wikipedia: The Missing Manual/Collaborating with other editors/Communicating with your fellow editors § Email. Then when I get around to restoring the work you have already done, I can complete your work if you are not able to.

I welcome further interaction, either on-wiki or off-wiki. Thank you again for all your hard work. YBG (talk) 21:37, 16 August 2018 (UTC)

Yeah. I don't Wiki that much, but when I saw your awesome template I really wanted to make my own version for Australia. I haven't vandalised anything and I was just really bummed that all my hard work has been lost. I make constructive edits as evidenced by everything I have done, and I really hope there is a way to recover that page, so I can finish the job off! Thanks for your kindness, not many people on Wiki seem to have any social skills, other than banning it seems. Hahaha. LR.2004 (talk) 04:01, 17 August 2018 (UTC)

Age difference between U.S. presidents and their vice presidents

Table

Partnership, president, and vice president			Older	Age difference		Time span years, days       days      starting – ending			Starting & ending events	days2
			P/V	years, days	±d
1	01 George Washington	01 John Adams	POTUS	3 years, 250 days	1346	7 years, 317 days	2,874	1789-04-21 – 1797-03-04	inauguration – inauguration	3868404
2	02 John Adams	02 Thomas Jefferson	POTUS	7 years, 165 days	2722	4 years, 0 days	1,460	1797-03-04 – 1801-03-04	inauguration – inauguration	3974120
3	03 Thomas Jefferson	03 Aaron Burr	POTUS	12 years, 299 days	4682	4 years, 0 days	1,461	1801-03-04 – 1805-03-04	inauguration – inauguration	6840402
4		04 George Clinton	VEEP	3 years, 261 days	-1357	4 years, 0 days	1,461	1805-03-04 – 1809-03-04	inauguration – inauguration	-1982577
5	04 James Madison		VEEP	11 years, 233 days	-4251	3 years, 47 days	1,143	1809-03-04 – 1812-04-20	inauguration – death of VP	-4858893
6		05 Elbridge Gerry	VEEP	6 years, 242 days	-2433	1 year, 264 days	629	1813-03-04 – 1814-11-23	inauguration – death of VP	-1530357
7	05 James Monroe	06 Daniel D. Tompkins	POTUS	16 years, 54 days	5898	8 years, 0 days	2,922	1817-03-04 – 1825-03-04	inauguration – inauguration	17233956
8	06 John Quincy Adams	07 John C. Calhoun	POTUS	14 years, 250 days	5364	4 years, 0 days	1,461	1825-03-04 – 1829-03-04	inauguration – inauguration	7836804
9	07 Andrew Jackson		POTUS	15 years, 3 days	5482	3 years, 299 days	1,395	1829-03-04 – 1832-12-28	inauguration – resignation of VP	7647390
10		08 Martin Van Buren	POTUS	15 years, 265 days	5744	4 years, 0 days	1,461	1833-03-04 – 1837-03-04	inauguration – inauguration	8391984
11	08 Martin Van Buren	09 Richard M. Johnson	VEEP	2 years, 49 days	-779	4 years, 0 days	1,461	1837-03-04 – 1841-03-04	inauguration – inauguration	-1138119
12	09 William Henry Harrison	10 John Tyler	POTUS	17 years, 48 days	6257	31 days	31	1841-03-04 – 1841-04-04	inauguration – death of POTUS	193967
13	11 James K. Polk	11 George M. Dallas	VEEP	3 years, 115 days	-1210	4 years, 0 days	1,461	1845-03-04 – 1849-03-04	inauguration – inauguration	-1767810
14	12 Zachary Taylor	12 Millard Fillmore	POTUS	15 years, 44 days	5522	1 year, 127 days	492	1849-03-04 – 1850-07-09	inauguration – death of POTUS	2716824
15	14 Franklin Pierce	13 William R. King	VEEP	18 years, 230 days	-6804	45 days	45	1853-03-04 – 1853-04-18	inauguration – death of VP	-306180
16	15 James Buchanan	14 John C. Breckinridge	POTUS	29 years, 268 days	10860	4 years, 0 days	1,461	1857-03-04 – 1861-03-04	inauguration – inauguration	15866460
17	16 Abraham Lincoln	15 Hannibal Hamlin	POTUS	196 days	196	4 years, 0 days	1,461	1861-03-04 – 1865-03-04	inauguration – inauguration	286356
18		16 Andrew Johnson	VEEP	45 days	-45	42 days	42	1865-03-04 – 1865-04-15	inauguration – assassination	-1890
19	18 Ulysses S. Grant	17 Schuyler Colfax	POTUS	330 days	330	4 years, 0 days	1,461	1869-03-04 – 1873-03-04	inauguration – inauguration	482130
20		18 Henry Wilson	VEEP	10 years, 70 days	-3723	2 years, 263 days	993	1873-03-04 – 1875-11-22	inauguration – death of VP	-3696939
21	19 Rutherford B. Hayes	19 William A. Wheeler	VEEP	3 years, 96 days	-1192	4 years, 0 days	1,461	1877-03-04 – 1881-03-04	inauguration – inauguration	-1741512
22	20 James A. Garfield	20 Chester A. Arthur	VEEP	2 years, 45 days	-775	199 days	199	1881-03-04 – 1881-09-19	inauguration – assassination	-154225
23	22 Grover Cleveland	21 Thomas A. Hendricks	VEEP	17 years, 192 days	-6402	266 days	266	1885-03-04 – 1885-11-25	inauguration – death of VP	-1702932
24	23 Benjamin Harrison	22 Levi P. Morton	VEEP	9 years, 96 days	-3383	4 years, 0 days	1,461	1889-03-04 – 1893-03-04	inauguration – inauguration	-4942563
25	24 Grover Cleveland	23 Adlai Stevenson I	VEEP	1 year, 146 days	-512	4 years, 0 days	1,461	1893-03-04 – 1897-03-04	inauguration – inauguration	-748032
26	25 William McKinley	24 Garret Hobart	POTUS	1 year, 126 days	491	2 years, 262 days	992	1897-03-04 – 1899-11-21	inauguration – death of VP	487072
27		25 Theodore Roosevelt	POTUS	15 years, 271 days	5750	194 days	194	1901-03-04 – 1901-09-14	inauguration – assassination	1115500
28	26 Theodore Roosevelt	26 Charles W. Fairbanks	VEEP	6 years, 169 days	-2360	4 years, 0 days	1,461	1905-03-04 – 1909-03-04	inauguration – inauguration	-3447960
29	27 William Howard Taft	27 James S. Sherman	VEEP	1 year, 326 days	-692	3 years, 240 days	1,336	1909-03-04 – 1912-10-30	inauguration – death of VP	-924512
30	28 Woodrow Wilson	28 Thomas R. Marshall	VEEP	2 years, 289 days	-1020	8 years, 0 days	2,922	1913-03-04 – 1921-03-04	inauguration – inauguration	-2980440
31	29 Warren G. Harding	29 Calvin Coolidge	POTUS	6 years, 245 days	2436	2 years, 151 days	881	1921-03-04 – 1923-08-02	inauguration – death of POTUS	2146116
32	30 Calvin Coolidge	30 Charles G. Dawes	VEEP	6 years, 312 days	-2503	4 years, 0 days	1,461	1925-03-04 – 1929-03-04	inauguration – inauguration	-3656883
33	31 Herbert Hoover	31 Charles Curtis	VEEP	14 years, 197 days	-5311	4 years, 0 days	1,461	1929-03-04 – 1933-03-04	inauguration – inauguration	-7759371
34	32 Franklin D. Roosevelt	32 John N. Garner	VEEP	13 years, 69 days	-4817	7 years, 322 days	2,879	1933-03-04 – 1941-01-20	inauguration – inauguration	-13868143
35		33 Henry A. Wallace	POTUS	6 years, 251 days	2442	4 years, 0 days	1,461	1941-01-20 – 1945-01-20	inauguration – inauguration	3567762
36		34 Harry S. Truman	POTUS	2 years, 99 days	829	82 days	82	1945-01-20 – 1945-04-12	inauguration – death of POTUS	67978
37	33 Harry S. Truman	35 Alben W. Barkley	VEEP	6 years, 166 days	-2357	4 years, 0 days	1,461	1949-01-20 – 1953-01-20	inauguration – inauguration	-3443577
38	34 Dwight D. Eisenhower	36 Richard Nixon	POTUS	22 years, 87 days	8122	8 years, 0 days	2,922	1953-01-20 – 1961-01-20	inauguration – inauguration	23732484
39	35 John F. Kennedy	37 Lyndon B. Johnson	VEEP	8 years, 275 days	-3197	2 years, 306 days	1,036	1961-01-20 – 1963-11-22	inauguration – assassination	-3312092
40	36 Lyndon B. Johnson	38 Hubert Humphrey	POTUS	2 years, 273 days	1003	4 years, 0 days	1,461	1965-01-20 – 1969-01-20	inauguration – inauguration	1465383
41	37 Richard Nixon	39 Spiro Agnew	POTUS	5 years, 304 days	2130	4 years, 320 days	1,781	1969-01-20 – 1973-12-06	inauguration – resignation of VP	3793530
42		40 Gerald Ford	POTUS	186 days	186	1 year, 13 days	378	1973-12-06 – 1974-12-19	confirmation of VP – resignation of POTUS	70308
43	38 Gerald Ford	41 Nelson Rockefeller	VEEP	5 years, 6 days	-1832	2 years, 32 days	763	1974-12-19 – 1977-01-20	confirmation of VP – inauguration	-1397816
44	39 Jimmy Carter	42 Walter Mondale	POTUS	3 years, 96 days	1191	4 years, 0 days	1,461	1977-01-20 – 1981-01-20	inauguration – inauguration	1740051
45	40 Ronald Reagan	43 George H. W. Bush	POTUS	13 years, 127 days	4875	8 years, 0 days	2,922	1981-01-20 – 1989-01-20	inauguration – inauguration	14244750
46	41 George H. W. Bush	44 Dan Quayle	POTUS	22 years, 237 days	8272	4 years, 0 days	1,461	1989-01-20 – 1993-01-20	inauguration – inauguration	12085392
47	42 Bill Clinton	45 Al Gore	POTUS	1 year, 225 days	590	8 years, 0 days	2,922	1993-01-20 – 2001-01-20	inauguration – inauguration	1723980
48	43 George W. Bush	46 Dick Cheney	VEEP	5 years, 157 days	-1983	8 years, 0 days	2,922	2001-01-20 – 2009-01-20	inauguration – inauguration	-5794326
49	44 Barack Obama	47 Joe Biden	VEEP	18 years, 257 days	-6832	8 years, 0 days	2,922	2009-01-20 – 2017-01-20	inauguration – inauguration	-19963104
50	45 Donald Trump	48 Mike Pence	POTUS	12 years, 358 days	4741	7 years, 308 days	2,864	2017-01-20 –  (present)	inauguration – (present)	13578224
	25 older presidents (thru Obama)		POTUS	Total age difference	97461	Total time span	36858			141579103
	24 older vice presidents		VEEP	Total age difference	-65770	Total time span	32707			-91120253
	49 total partnerships (thru Obama)		ALL	Total age difference	31691	Total time span	69565			50458850
	26 older presidents		POTUS	Total age difference	97461	Total time span	39722			155157327
	50 total partnerships		ALL	Total age difference	31691	Total time span	72429			64037074

Comments

I made this primarily for my own benefit, to validate my gut reaction that the President is usually older than the VP - Obama/Biden notwithstanding. But the data actually shows the 50 "partnerships" are split almost exactly evenly: 26 presidents are older than their VP and 24 are younger. Omitted from the above chart are instances when the vice-presidency was vacant. YBG (talk) 05:28, 27 June 2018 (UTC)

On average, the president is slightly older than the VP, and the difference is decreasing slightly over time. Haven't checked the weighted average yet. YBG (talk) 01:59, 3 July 2018 (UTC)

The 26 older presidents are older by an average of 3,748.5 days (97461/26); the 24 younger are younger by average of 2740.4 days (65770/24); in the average partnership, the president is older by 633.8 days (31691/50). The weighted averages are 3854.4 days (144177171 days²/37406 days), 2786.0 days (91120253 days²/32707 days), and 756.7 days (53056918 days²/70113 days) respectively. YBG (talk) 14:14, 22 July 2018 (UTC)

The non-weighted averages will stay the same throughout the duration of the Trump/Pence partnership, but to bring the weighted averages up to date, add 4741×2316 to the 1st and 3rd numerator and add 2316 to the 1st and 3rd denominator. YBG (talk) 19:21, 23 July 2018 (UTC)

Here's the data presented tabularly:

Average number of days the U.S. president is older than his vice-president
As of November 23, 2024, 2864 days into the Trump/Pence partnership.
(red values update automatically every day)

	Count	Average difference		Weighted average difference
	n	∑ [POTUS BD]–[VP BD] n		∑ ([POTUS BD]–[VP BD]) × ([End Day]–[Start Day]) ∑ [End Day]–[Start Day]
Presidents older than VP	26	-3748.5	97461 26	3906.08	155157327 39722	141579103 + 2864×4741 36858 + 2864×
VPs older than president	24	-2740.42	-65770 24	-2785.96	-91120253 32707	0-91120253 + 0×4741 000 32707 + 0 000
All partnerships	50	633.82	31691 50	884.14	64037074 72429	050458850 + 2864×4741 69565 + 2864×

YBG (talk) 06:28, 25 July 2018 (UTC)

Ex officio

The following is in response to the comment that I made on JTRH's talkpage. YBG (talk) 05:56, 15 August 2018 (UTC)

The Vice President is not "ex officio" the President of the Senate. The Vice President is the President of the Senate. No modifier is necessary. Best, JTRH (talk) 00:47, 15 August 2018 (UTC)

@JTRH: It sounds like you are using "ex officio" as though if someone holds a position "ex officio" it somehow means they are occupying the position unofficially or temporarily. The President of France is one of the two co-princes of Andorra. To say that he is a prince "ex officio" does not take away from him being a prince, it merely makes a statement about why he is prince. In the same way, the VP is President of the Senate ex officio because he holds that presidency for no other reason than that he is the He is the VP. That is what the wiktionary article means. YBG (talk) 00:56, 15 August 2018 (UTC)

Ex officio means you do Y as a function of doing X. It's not a necessary modifier or qualification for the Vice President. "The Vice President of the United States shall be President of the Senate" is not "ex officio," that's his primary Constitutional responsibility. The term is unnecessary in this case. It's no more necessary than saying the President is "ex officio" Commander-in-Chief of the Armed Forces. JTRH (talk) 01:07, 15 August 2018 (UTC)

@JTRH: I agree that it is unnecessary verbiage. YBG (talk) 03:17, 15 August 2018 (UTC)

xref to

Hard to select namespaces suggestion at MediaWiki. YBG (talk) 07:08, 23 April 2019 (UTC)

History of the periodic table

Hi! A while ago, yourself, ComplexRational, and I agreed we would work on History of the periodic table to bring it to the FA status. I'm trying to write my ideas on that article's talk page from time to time to make sure ComplexRational understands what I'm doing, and I generally have similar expectations from him. You also wanted to take it on, so you could voice your comments on our little topics of discussion, read some interesting literature on the topic (I could give you a few pointers if you're interested), or write with us. ComplexRational wrote a good basis article; I'm now trying to fill it with details on some aspects, and ComplexRational will probably join me in filling in those details soon. So if you want to join, now's a great time!--R8R (talk) 18:48, 24 April 2020 (UTC)

Note to self

This might qualify as being a "non groupic" PT, but certainly that cannot be said of "our" OT. (See special:diff/973773980) YBG (talk) 15:10, 30 August 2020 (UTC)

Since you liked DA's PT...

...you may like User:Double sharp/Teaching periodicity. It also has a periodic table colouring electronegativity as a gradient from red to violet (hopefully pretty) as well as one dividing only by blocks and metallicity (with an explanation as to why I prefer that). Not proposing it for WP, just thought you might be interested. ^_^ Double sharp (talk) 11:14, 6 October 2020 (UTC)

Question re Aufbau

@Sandbh: I am discussing this in user space so that you can correct what is apparently a misunderstanding on my part without bothering the rest of the world.

My understanding of the Aufbau principle or Madelung rule is that it predicts that the shells will be filled in this order:

1s, 2s, 2p, 3s, 3p, 4s, 3d, 4p, 5s, 4d, 5p, 6s, 4f, 5d, 6p, 7s, 5f, 6d, 7p, ...

and that each subshell is filled completely before the next one is started.

Given that s, p, d, and f subshells hold 2, 6, 10, and 14 electrons respectively, this means that the subshells are filled in this way:

1122222222333333334433333333334444445544444444445555556644444444444444555555555566666677555555555555556666666666777777
ssssppppppssppppppssddddddddddppppppssddddddddddppppppssffffffffffffffddddddddddppppppssffffffffffffffddddddddddpppppp

Or when laid out in two dimensions, we get:

ss                              
ss                        pppppp
ss                        pppppp
ss              ddddddddddpppppp
ss              ddddddddddpppppp
ssffffffffffffffddddddddddpppppp
ssffffffffffffffddddddddddpppppp

Or this:

                              ss
                              ss
                        ppppppss
                        ppppppss
              ddddddddddppppppss
              ddddddddddppppppss
ffffffffffffffddddddddddppppppss
ffffffffffffffddddddddddppppppss

This is what I meant by being more consistent with Aufbau.

Have I missed something? I know that Aufbau does not represent what is observed expirementally. But what I said was that Sc/Y/Lu is more consistent with Aufbau, not that it is more consistent with expiremental results. YBG (talk) 09:55, 25 October 2020 (UTC)

Well, the aufbau rule has a regular pattern. It hasn’t been derived from first principles. Claims for such a derivation are made from time to time; none have been accepted. Scerri would otherwise be shouting such a derivation to the rooftops. So the fact that the Lu form corresponds to it whereas the La form requires a split d-block doesn’t mean anything fundamental.

The aufbau rule does work well in the sense that every time it fails to predict the right configuration/s it always, inevitability, resumes making the right prediction at least until it fails the next time.

That said, when used this way, the La version complies more closely to the aufbau rule than the Lu form.

How do you see this? Sandbh (talk) 12:13, 25 October 2020 (UTC)

@Sandbh: I'm trying to understand your point of view here. Can you please explain to me what you mean by "this way"? Thank you. YBG (talk) 06:47, 26 October 2020 (UTC)

I'll take it one step at a time. The aufbau rule is used to predict the electron configurations of the gas phase atoms. For example, Aufbau predicts La should have one f electron in its electron configuration. Thus it should be [Xe] 5f1 6s2. Whereas, in real life it is [Xe] 5d1 6s2.

There are a couple of wrinkles to bear in mind. (1) Gas phase configurations aren't necessarily the same as the configurations of the atoms in their condensed or bonded states. (2) That said, gas phase configurations are relatively easy to measure.

Are you with me so far? Sandbh (talk) 08:31, 26 October 2020 (UTC)

@Sandbh: OK, Aufbau predicts gas phase electron configuration, and (a) prediction is not always accurate, and (b) Gas-phase config not always the same as condensed config. YBG (talk) 08:42, 27 October 2020 (UTC)

Mostly. Electron configurations were largely worked out first via spectroscopy. Aufbau was then “invented” or retrofitted onto the observations, and subsequently used to attempt to predict the configurations of then unknown atoms. Sandbh (talk) 10:00, 27 October 2020 (UTC)

@Sandbh: I'm puzzled by your use of "invented" and "retrofitted". Invented speaks of the first exposition of a concept. Retrofitted speaks of something being adjusted after its initial exposition. YBG (talk) 23:44, 29 October 2020 (UTC)

@YBG: Yes, aufbau was "invented" or proposed as a "rule" that seemed to match the spectroscopists' observations. That is OK science, as a first step of proposing a rule or theory that seems to match the evidence.

Janet went as far as asserting that, in the case of the discrepancies, the spectroscopists' observations must be wrong[!], and retrofitted or adjusted the electron configurations of the elements to match aufbau! Janet turned out to be wrong.

After more configurations were confirmed via spectroscopy, these confirmed that aufbau had a significant number of discrepancies compared to the situation on the ground.

Aufbau therefore does not explain anything, although it is an OK mnemonic, provided you have your wits about you. I recall Double sharp telling me that when he was taught aufbau, he and his fellow students were expected to memorise all the exceptions! Sandbh (talk) 01:41, 30 October 2020 (UTC)

Hat note of convenience 1

I am not YBG, but I watch this talk page, and would be interested in getting a few things clarified from User:Sandbh about this latest reply of his. So I would like to ask him first if he is agreeable to that, and also to ask YBG if he is agreeable to me doing so here. Double sharp (talk) 18:00, 25 October 2020 (UTC) @Double sharp: It am glad to have for you to have a discussion here, provided it is not a discussion that should a wider audience. But I would prefer for the present that this conversation be confined to what is helpful for me to understand Sandbh's point of view, which I do not yet comprehend. Thank you. YBG (talk) 06:47, 26 October 2020 (UTC)

premature off-topic discussion, contrary to my request YBG (talk) 05:04, 27 October 2020 (UTC)

@YBG: Ah, thank you. Since User:Sandbh has not yet agreed to it, I will not ask my question yet, but it would be simply about clarifying something he said in this thread. ^_^ Double sharp (talk) 09:29, 26 October 2020 (UTC) @Double sharp: On my part, feel free to ask your question. Sandbh (talk) 21:58, 26 October 2020 (UTC) @Sandbh: Thank you. My question is: in your 12:13 post yesterday you say in the first paragraph "the Lu form corresponds to it whereas the La form requires a split d-block", although you then say it doesn't mean anything fundamental. But in the third paragraph you write "when used this way, the La version complies more closely to the aufbau rule than the Lu form." Is either one a mistake, or is it that you see the Aufbau as being used differently between the two cases? If it is the latter, what's the difference? Double sharp (talk) 22:01, 26 October 2020 (UTC) Double sharp, I will not attempt to answer for Sandbh, but if you have not read his published article (freely-available and linked from my talk page) then you perhaps should as it certainly explores some of what you are asking. For myself, I would note that I think there is a larger epistemological question (or maybe several) lurking under the surface of your question and other parts of the current discussions – one that I think is interesting but which I doubt is ready for WP (on DUE and OR grounds). Sandbh's article is delving into areas that are very much worth being explored in the literature to see if they influence scientific consensus... but unless there is a lot more out there than I am aware of (which is possible), I think there will need to be more work in the primary literature before we can make much use of (parts of) what Sandbh has written. EdChem (talk) 22:40, 26 October 2020 (UTC) @EdChem: Thank you. I've read through his article, but I confess I don't remember something addressing this question very clearly there, so I'm asking. Maybe it is there and I did not get it. Regarding epistemological questions, I do think there is something of that kind going on here and throughout the whole La vs Lu dispute, but that is from my personal experience arguing it out and so we definitely can't use it. I'm not aware of any RS saying it unfortunately, but it would definitely make an interesting topic to explore in the primary literature. But I'd rather not impose too much on YBG's talkpage given his comment earlier today, so if you want to talk a bit more about that, I'd rather do it on my or your talk page. Double sharp (talk) 23:23, 26 October 2020 (UTC) @Double sharp and EdChem: Wow, it's been a busy morning between WP:ANI (Softlavdner reverted my revert, so now I have that to deal with as well); R8R's talk page; and now here. BTW it is good to see you here, EdChem. And it is nice to be able to talk about hard-core content. ^_^ So, addressing Double sharp's interesting question first. 1. Here is an idealised Lu table: ss ss pppppp ss pppppp ss ddddddddddpppppp ss ddddddddddpppppp ssffffffffffffffddddddddddpppppp ssffffffffffffffddddddddddpppppp The Lu form corresponds to the aufbau rule. 2. Here is an idealised La table: ss ss pppppp ss pppppp ssd dddddddddpppppp ssd dddddddddpppppp ssdffffffffffffffdddddddddpppppp ssdffffffffffffffdddddddddpppppp The La form does not correspond to the aufbau rule, since it features a split d-block. That said, the correspondence of the Lu form to aufbau does not meaning anything, since aufbau is simply a pattern, lacking any accepted ab initio derivation. Several authors claim to have achieved such a derivation; none of these claims have been accepted. Now, let us set aside concerns about the basis of aufabu, and look at how it used in the literature, which is to predict the gas phase configurations of atoms. The first thing about aufbau is that whenever it fails, it always resumes course sooner or later. So it is a curious kind of approximation. The second thing about aufbau is that it's not a very good approximation since it yields about 20 errors up to the first 100 or so elements. The third thing about aufbau is that if you look at its predictions for differnentiaing electrons (de), it is more accurate at predicting these for the La form than the Lu form. How bizarre is that?! But there you go. A caveat with regard to de is that DS has raised an objection as to how these are worked out, in a few cases. I addressed this in my article, at note 1: "Sometimes the differentiating electron (d/e) is not immediately apparent. For example Z = 40 Zr is 4d2s2 and Z = 41 Nb is 4d4</s1. Here the d/e seems to be d2s−1. In such cases the d/e is taken to be the newly added d-electron, rather than the s-electron that was already there (so to speak). So, yes, as per my first paragraph, aufbau does not mean anything. That has not stopped some muggles (by which I mean non WP-editors) saying Lu must be "it" since it corresponds with aufabu. And if we counter-factually assume it does mean something, it works better (as per my third paragraph) for the La form in any event! Muggles respond to the latter via perceptual filtering. Since it does not correspond to their world-view, they adopt the see no-evil, speak no-evil, hear no-evil approach. Very, very few muggles understand or aware of their subconscious perceptual filters nor that these are on auto-pilot. DS, I hope this clarifies the sitation. Sandbh (talk) 01:46, 27 October 2020 (UTC) Placeholder re EdChem's thoughts. Sandbh (talk) 01:46, 27 October 2020 (UTC)

Hat note of convenience 2

@Double sharp and Sandbh: Please take this conversation elsewhere. Apparently I was not clear enough. I wanted this thread to be reserved, only about my trying to understand Sandbh, and was hoping that until that was accomplished, it would not be cluttered with other things that distracted from that purpose. Apparently I was not clear enough for either of you to understand that. YBG (talk) 04:46, 27 October 2020 (UTC) @Double sharp and Sandbh: ... Unless you are willing to wait until I have reached an understanding of Sandbh's point of view. YBG (talk) 05:06, 27 October 2020 (UTC) I'm more than happy to wait. Sandbh (talk) 06:09, 27 October 2020 (UTC) No problem with me either, and my apologies for misunderstanding you. Double sharp (talk) 08:20, 27 October 2020 (UTC) My apologies as well, YBG... I did not understand what you wanted. As this is your user talk page, you are free to direct the discussion. EdChem (talk) 10:37, 27 October 2020 (UTC)

RfC: Periodic table categories

Jump to §§ RfC: Periodic table categories, Proposed text of RfC, Questions, Responses to questions, and Other comments about the RfC

This is just a preliminary proposal I'm hosting here on my talk page so that pagewatches and others can chime in. I am open to all comments, including suggestions that this RfC would be a total waste of time.

I am conflicted on this issue. Although currently leaning toward eliminate, I love the categories and their colors; they are what piqued my interest in WP:ELEM and solidified me as a confirmed wp-a-holic.

I will look at this again in a few hours, but after that I'm taking a short wikibreak. In the event that incivility erupts, I hereby authorize any one of EdChem, Sandbh or Double sharp to declare a moratorium on this thread until I am back on line. YBG (talk) 08:54, 6 November 2020 (UTC)

Proposed text of RfC

Jump to §§ RfC: Periodic table categories, Proposed text of RfC, Questions, Responses to questions, and Other comments about the RfC

Should we eliminate the system of chemical element categories based on the metal-to-nonmetal trend used to color elements in the WP periodic table?

Arguments for eliminating

1. Although many of these categories are found in WP:RS, there is no WP:RS consensus on any particular system of categories.
2. Picking any one system from the literature for prominent display throughout en.WP would give it undue prominence, violating WP:DUE.
3. Trying to harmonize various published systems would be synthesis, violating WP:SYNTH.
4. Taking categories recognized in the literature and combining them into our own system constitutes original research, violating WP:OR

Arguments for retaining

1. While there is no universally agreed-on system of categories, the literature attests to the individual categories and to the idea of a system of categories.
2. Eliminating these categories will hide useful information from our readers.
3. The categories help our readers understand the elements and orient themselves in the periodic table, especially the micro tables in the element sidebars.
4. The usefulness of WP's system of color categories is attested to by how frequently other wikipedias, websites and books have copied our system.

YBG (talk) 08:54, 6 November 2020 (UTC)

Questions

Jump to §§ RfC: Periodic table categories, Proposed text of RfC, Questions, Responses to questions, and Other comments about the RfC

1. Is the question phrased neutrally? Yes/No (if no, please explain)
2. Are the arguments expressed well? Yes/No (if no, please explain)
3. Are there any important arguments that are missing? Yes/No (if yes, please list)
4. Should any of these arguments be removed as unimportant? (if yes, please list)
5. Do the listed arguments need to be reordered? Yes/No (if yes, how?)
6. What would be the best place for this RfC to be hosted?

   a. WT:ELEM

   b. Talk:Periodic table

   c. Some other place (where?)

7. How much should be quoted in the central RfC area? (see Wikipedia:Requests for comment/Article topics § Maths, science, and technology)

   a. Just the question

   b. The question plus a very brief summary (to be written) of the pro/con arguments

   c. The question plus the arguments listed

   d. The question, arguments listed, and additional arguments

8. Should this RfC be coordinated with the nonmetal category RfC?

   a. Yes, nonmetal RfC before this one

   b. Yes, nonmetal RfC at the same time as this one

   c. Yes, nonmetal RfC after this one

   d. No, so long as they are not simultaneous

   e. No, it does not matter - before, after, or simultaneous

9. Should this RfC be coordinated with a Group III RfC?

   a. Yes, Group III RfC before this one

   b. Yes, Group III RfC at the same time as this one

   c. Yes, Group III RfC after this one

   d. No, so long as they are not simultaneous

   e. No, it does not matter - before, after, or simultaneous

10. What should be the next step?

    a. Drop it

    b. Solicit more comments at WT:ELEM

    c. Take the RfC live.

Responses to questions

Jump to §§ RfC: Periodic table categories, Proposed text of RfC, Questions, Responses to questions, and Other comments about the RfC

Other comments about the RfC

Jump to §§ RfC: Periodic table categories, Proposed text of RfC, Questions, Responses to questions, and Other comments about the RfC

Comments (DS)

I think the RFC question should just be the question. Arguments, as I understand it, should go below with !votes, so that the statement is neutral.

That being said, I think your arguments for eliminating are very consistent with my case, so I feel good that I was understood well. I just think that they should be a part of our !votes instead. ^_^ I might take your statement if it's all right with you (and probably expand it with sources), and then add one sentence to 1 saying "The literature does not always organise categories into a system at all". BTW, my replies to points #2 and #4 for retention would be:

2. A category scheme displayed with colours gives readers the impression that there is a generally accepted idea of the categories, and that for the most part each element belongs to exactly one category, which is not at all true. Many elements in fact belong to multiple categories among those common in the literature, and displaying just one is not only WP:UNDUE, but it also hides useful information.
4. Although similar category schemes to the WP one are indeed common in other Wikipedias, websites, and books, they often have differences in precisely the areas where there is no clear dominance of a category in the literature.

I also think it needs to be stressed to avoid misunderstandings that the proposal is only about colouring. No one, AFAIK, is proposing to eliminate all mention of categories. I don't have a problem with discussing categorisation in articles and in infoboxes so long as each category is given its WP:DUE weight. So it would look less like what we have currently where each element is shoehorned into just one category, but something more like the "minor planet category" entry in an article like 90482 Orcus. Notice also that all three categories mentioned for Orcus in that infobox (TNO, plutino, possible dwarf planet) are also mentioned in the lede. That should address concerns about hiding knowledge. (Although it indeed should be questioned if they should be called "categories" or something else.) ^_^

As for an RFC: I am iffy about RFCs for this in general, simply because I suspect the majority of editors are not going to be in a good position to comment. To get it, you need to probably also have some general knowledge of the literature situation. Over at WT:ELEM, we already have a base of editors familiar with those issues, and anyone else if interested can always come if a little note is placed at Talk:Periodic table mentioning that something's going on at our pages. A discussion at WT:ELEM is not and has never been limited to project participants.

I think it makes more sense if the group 3 thing comes first, because it makes a difference on whether La or Lu gets the d block colour by default. And I think it makes more sense if this comes before the proposed nonmetal recategorisation question, because if the "blocks alone" proposal passes, then it makes that question moot. I think that's consistent with what I've said before at ELEM. Double sharp (talk) 22:33, 6 November 2020 (UTC)

Comments (Sandbh)

Ditto YBG: I love the categories and their colors; they are what piqued my interest in WP:ELEM and solidified me as a confirmed wp-a-holic.

RFC text. I suspect the proposed text of the RFC ought to say what you would replace it with. I doubt people would be comfortable with voting for the unknown.

RFC host page. The best place to host the RFC is at Talk:Periodic table. Our project is too small, insufficiently representative, and does not represent an NPOV "venue". Since the RFC is about the periodic table, it ought to take place at that talk page.

What to display. A periodic table is like a map.[1] A map of the world could e.g. display climate zones; crop types; elevations; or political territories, with overlaps at the boundaries according to disputed claims; etc. A periodic table could display e.g. blocks; dates of discovery; elements only; groups i.e. 18 of them, plus the Ln and An; metal-metalloid-nonmetal; or phases i.e. solid, liquid, gas; etc. Perhaps the question is what is the most generally useful information to display, given it's not possible to display everything concurrently.

Colour categories. 1. The Encyclopaedia Britannic PT, which predates the WP PT, features nine colour categories; 2. the most popular Google PT, which postdates the WP PT, features nine colour categories. Atkins, in his book The Periodic Kingdom (1995) surveyed the periodic table as if it was a continent, comprised of regions and territories. Gonick & Criddle, in The Cartoon Guide to Chemistry (2005, p. 38) refer the reader to 3. the LANL PT which they describe as "a wonderfully information-rich" PT. This has ten colour categories. 4. The ACS PT has ten colour categories.[2] 5. The RSC PT has eight main group colour categories; one TM colour category; and one each for the Ln and An.[3] 6. 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.

Category boundaries. Such boundaries are rarely sharp (Atkins 1995, p. 4). We could note this in the lede periodic table graphic, although none of the above six sources felt the need to do this. In any event, we elaborate this in the main body of the periodic table article, and note that "Categorisation as described here can vary among authors".

No RS consensus in literature. That's not quite right. The creators of the WP:PT and its colour categories did so on the basis of what they understood to be the consensus in the literature, noting unanimity is not a requirement for consensus. See: archive 1 of the periodic table talk page, and some related commentary here: Template talk:Periodic table/Archive 1.

Which goes first?. Group 3 was put to an RFC on 20 Jul and closed, without success, on 6 Aug 2020. In this light, I submit the non-metal RFC should now be afforded the opportunity to be put.
--- Sandbh (talk) 07:06, 8 November 2020 (UTC)

Withdrawn

I have now withdrawn this proposed RfC. YBG (talk) 02:40, 14 November 2020 (UTC)

Arbitration

Please comment here Wikipedia:Arbitration/Requests/Case if you wish, as a witness. Jehochman ^Talk 02:50, 13 November 2020 (UTC)

Regarding categories

Maybe I'll try, since I'm the one who was convinced about that view till today when Jehochman suggested on EdChem's talk page about a style guide, maybe it'll help.

You can sort of rationalise it this way. The problem is that (1) sources in English tend to prefer using categories to blocks-only, but (2) sources in English also cannot exactly agree on what categories to use. We also can often find the situation that sources not focusing on an element often take a different view from sources that do for radioactives: At is often called a halogen by those who don't know much about it and called a metal by those who do. In this situation I suspect you're more or less forced to have UNDUE either way. Because if you don't show any categories it's UNDUE weight towards that no-categories approach as most sources show some categories, but if you show any one particular set of categories it's UNDUE weight because of the problem: why that approach and not another one?

So, is our category scheme OR? One can make the case that it's UNDUE and partly synthesis, and I still have doubts about it, but since a more experienced Wikipedian Jehochman has suggested arbitrarily choosing one common way, I guess it may well be the best of all options. If we interpret policies strictly, pretty much every solution has UNDUE lurking behind it. But if we take it as "well, most sources seem to show something like this, even if it's not exactly this" – then, I guess you can make the argument that some form of categories are better than none. Because many sources do colour in a table like that. Even if they are fuzzy about what exactly "halogen" means, what exactly "transition metal" is, it's still common to colour somehow.

We can argue about some things, maybe; I think there is some general support for reinstating halogen nonmetals as a category just because most people do it for F-Cl-Br-I-(At); the problem is just At. I guess, for such cases, one can sort of make case-by-case justifications. Is it SYNTH? Maybe, but the alternative might just be worse UNDUE.

So maybe this is the sort of issue where a strict application of policy is just impossible because of the state of the literature. Double sharp (talk) 19:24, 13 November 2020 (UTC)

Another way to think of this is that any diagram or table will oversimplify whatever it is representing. Some oversimplify one way, some another. Don't let perfect be the enemy of good. Imagine you are explaining chemistry to a first year high school student. What do they need to know? Any extra complexity an be explained in the article text (who say this alternative, who says that alternative), or footnotes. Jehochman ^Talk 19:37, 13 November 2020 (UTC)

So this is your reasoning:

1. We have to show either something or nothing.
2. Showing nothing is rare in the literatore, so going that route would violate WP:UNDUE
3. The only common "showing something" alternatives in the literature are (a) blocks only and (b) schemes similar to our metallicity-based categories
4. Showing blocks only is rare in English literature, so going that route would violate WP:UNDUE
5. Our scheme agrees with most published category-based schemes in the essentials: the identity of most of the categories, and the membership of those categories except at the boundaries
6. Our scheme disagrees with published category-based schemes only in peripheral areas: the categorization of boundary elements and whether to divide or combine certain categories
7. In those perpheral areas, our scheme follows the majority of published category-based schemes.
8. Therefore, our category-based scheme does not violate WP:UNDUE
9. Furthermore, few published PT show multiple categories for each element; so adopting such a scheme would also violate WP:UNDUE and in addition would do our readers a disservice by overwhelming them with unnecessary complexity.

Have I got this line of reasoning correct? YBG (talk) 20:03, 13 November 2020 (UTC)

I think it's almost correct, yes. The question is just #7. I think that that needs to have one addition at the end: "or specialised review articles when the element is extremely rare and difficult to study". That is for superheavies and astatine which are not covered well by general sources for obvious reasons (i.e. being mayflies with half-lives measured in hours at best). Of course that's just a side note, your main idea of the line of reasoning is absolutely correct.

@Sandbh: I recall you mentioned that Po and At are getting called metalloids less and less in the current literature. I suppose Po is going to metals, but what is At going to? Double sharp (talk) 20:38, 13 November 2020 (UTC)

OK, what about the following as a WP:ELEM policy:

Except as required for the context of a particular article, periodic tables will be consistent in form throughout the articles that come under the purvue of our project. The common form shall follow as near as possible the consensus of WP:RS. The RS consensus will be determined from the RS corpus of general chemistry textbooks available in English and used in high schools and colleges. For particular issues, if the specialized professional literature which examinex the issue directly differs from the general textbook consensus, the project will, by consensus, choose which consensus to follow, the RS textbook consensus, or the RS specialist consensus.

I think gives us the flexibility needed to cover the weird edge cases, and also to adjudicate the group 3 issue whichever way we think best. YBG (talk) 00:34, 14 November 2020 (UTC)

Yes, I think that's fine, except that I've written to Sandbh saying that I do not seek to change the group 3 issue at this time. This is because he's given information that says that Scerri is writing an article in Chemistry International that will summarise the IUPAC project's preliminary conclusions. That information will be important for deciding the issue, so I agreed that it is better to adjudicate things later. At least once it's publicly released, possibly even later depending on the chemistry community's reaction to it and the subsequent report. Then we can see how it develops from there. Also, I think we should say "if there is no exact RS consensus, an option close to the centre of RS variation will be chosen". Double sharp (talk) 00:43, 14 November 2020 (UTC)

I agree we should change nothing group-3 wise now. My point is that this wording gives us freedom to go either way on that issue at the appropriate time. YBG (talk) 01:03, 14 November 2020 (UTC)

I agree with you on that, I just clearly didn't express it clearly. ^_^ Double sharp (talk) 01:05, 14 November 2020 (UTC)

@Double sharp: WRT to my perception of what is a happening to Po and to At, there is a risk that my perceptual filters are turned up too high. For example, when I brought a new car, I suddenly started seeing lots of the same kind of cars on the road.

Anyway, At, like Po, seems to be becoming more widely regarded as a PTM.

I say this in the context of the 35 citations for the condensed At is expected to be a metal paper. Now, At stills seems to still be able to behave like halogen sometimes, but we see that kind of behaviour in Au, too.

Oh, I happened upon this 2019 paper arguing for Cn as a relativistic noble liquid with a band gap of ~6.5 eV, which I see we already cite in our article. That's nonmetal (insulator) territory. It already has nine citations. Sandbh (talk) 01:48, 14 November 2020 (UTC)

• I have now withdrawn this proposed RfC. YBG (talk) 02:39, 14 November 2020 (UTC)

@Sandbh: OK, so it's happening in favour of what we're showing, which sounds good. ^_^

On copernicium: seems to me that this will certainly cause a problem with nearly everybody's category scheme if and when the results are experimentally confirmed. I suggest to not worry about it right now since the entire category-showing literature will have to solve it then. Perfect is enemy of good after all. ^_^ Double sharp (talk) 11:06, 14 November 2020 (UTC)

@Double sharp: If copernicium has a band gap of 6.5 eV I guess it will be a colourless liquid, unless relativistic effects come into play as they do with e.g. gold. Diamond is a wide-bandgap semiconductor (5.47 eV). Since it's expected to have an unfilled d shell, I guess it ought then to be regarded as a transition non-metal, and a noble one at that, once its applicable properties are reasonably well "confirmed" in the literature. Sandbh (talk) 23:29, 14 November 2020 (UTC)

Sandbh Well relativistic effects are responsible for this happening in the first place. And you're making my point by getting a new category name for Cn. ^_^ So I think we will wait till it's actually for sure characterised and the literature starts having to deal with it.

Not sure what you mean by the unfilled d shell, though. Cn has 6d¹⁰, it's just that 6d should ionise preferentially to 7s. Or do you think now that the Zn group should go back to transition metals? I don't mind either way. Anyway, I think it can all wait till the literature finally has to deal with it. Double sharp (talk) 23:58, 14 November 2020 (UTC)

Double sharp Oh, with Cn I was thinking about, "Cn2+ is likely to have a [Rn]5f¹⁴6d⁸7s² electronic configuration, using the 6d orbitals before the 7s one, unlike its homologues.

On items such as these I intend to post them to WP:ELEM as, "Speculation: …" in the same way that I did with "Philosophy: …" so folks with no interest in either can skip them. As you say, it may take quite a while for "Speculation: …" to turn into "Proposal …". I recall e.g. we waited several years e.g. before flipping At from a metalloid to a PTM. Sandbh (talk) 01:44, 15 November 2020 (UTC)

@Sandbh: If you wish. Myself, I am personally not inclined to speculate too much as long as they are calculations only. I would like to wait for more info (both experimental and predicted) about Cn and Og especially first. ^_^ Double sharp (talk) 11:02, 15 November 2020 (UTC)

Regarding the nonmetal RFC

Why are you so much better at explaining my logic than I am? (Tongue-in-cheek question, obviously.) ^_^ Double sharp (talk) 10:24, 23 November 2020 (UTC)

Status of the LANL table

The LANL table is troubling. They say, "The halogen elements are a subset of the nonmetals", so At and Ts are nonmetals according to them. Presuming high school students are taught that metallic character increases going down a main group, the LANL table shows it increasing going down groups 13 to 16, but not, for some reason, when it comes group 17.

The text accompanying the LANL table is a shocker. Whoever wrote it didn't know what they were talking about:

LANL says	What's wrong
…aluminum and tin are included characterized [sic] as Metalloids or poor metals	Say no more (note erroneous spelling of Al)
Alkaline earth metals. The alkaline earth metals have very high melting points and oxides that have basic alkaline solutions. Their characteristics are well described and consistent down the group.	In fact, the AEM do not have very high melting points; beryllium oxide does not have a basic alkaline solution; and Be and Mg are quite distinct from the rest of the AEM.
Transition metals. The transition elements are metals that have a partially filled d subshell (CRC Handbook of Chemistry and Physics) and comprise groups 3 through 12 and the lanthanides and actinides (see below).	In fact, group 12 do not have a partially filled d subshell.
Post-transition metals. …are Al, Ga, In, Tl, Sn, Pb and Bi. As their name implies, they have some of the characteristics of the transition elements.	Whoever wrote that had no clue.
Metalloid (or "semi-metal" or "poor metal"). The metalloids are B, Si, Ge, As, Sb, Te, and Po. They sometimes behave as semiconductors (B, Si, Ge) rather than as conductors.	Er, no, metalloids are not poor metals. And Te is a semiconductor, too.
Lanthanides. …comprise elements 57 (La, hence the name of the set) through 71. They are grouped together because they have similar chemical properties. They, along with the actinides, are often called "the f-elements" because they have valence electrons in the f shell.	In fact, La and Lu do not have "valence" electrons in the f shell.
Actinides. The actinides comprise elements 89 through 103. They, along with the lanthanides, are often called "the f-elements" because they have valence electrons in the f shell.	In fact, Ac, Th and Lr do not have "valence" electrons in the f shell.
Halogens. [a]…They generally [sic] very chemically reactive…[b] are present in the environment as compounds rather than as pure elements	[a] eh? [b] A 2012 study (19 citations) reported the presence of 0.04% F2 by weight in antozonite, attributing these inclusions to radiation from the presence of tiny amounts of uranium

Rather than relying on LANL, here are some teaching perspectives:

1999	In CHEM 13 News, a magazine published for teachers of introductory chemistry and mostly written by high school chemistry teachers, Stephen Hawkes argued Po was a metal.
2007	Holt, Rinehart and Winston, who publish chemistry textbooks, published a note for teachers on Why polonium and astatine arenot metalloids in HRW texts.
2010	In the Journal of Chemical Education, Hawkes made the same argument, that Po was not a metalloid. doi:10.1021/ed100308w
2011	Pamela Fujinaka, a high school teacher in Honolulu, uses the mnemonic, "Up, up-down, up-down, up…are the metalloids" i.e. B, Si, Ge, As, Sb, Te in 'Faculty profile: Elements of Great Teaching', The Iolani School Bulletin, Winter.
2013	In JChemEd, I cited Hawkes' argument, confirming the status of Po as a metal. doi:10.1021/ed3008457

The global audience for JChemEd includes instructors of chemistry from middle school through graduate school, professional staff who support these teaching activities, as well as some scientists in commerce, industry, and government.

Presuming high school students are taught about semiconductors (and that all metalloids are such, or exist in semiconducting forms) no semiconducting allotrope of Po is known.

Pinging Double sharp. Sandbh (talk) 01:13, 24 November 2020 (UTC)

Sandbh, JChemEd surely isn't just containing things that are standard in the literature, but also things that are proposed to be standards. You've listed various sources saying that Po should not be a metalloid, for sure. I even agree with them. Jensen has also published an article there saying Lu should not be in group 3. And as you know, I strongly agree with him. The question in both cases I am asking is: how many people are listening to them? Not "how right do we think they are". My case is not based on that; if it were that, I'd still be arguing furiously for Lu in group 3, and Sb and Po as metals (following Rayner-Canham and Overton in both instances, I add!). But I don't because those are not the majority views, no matter how right I personally think they both are, and no matter how strongly I think science is on my side. I don't think WP would work very well for scientifically controversial subjects like group 3 and which-elements-are-metalloids if it didn't have a general idea of following the majority viewpoint rather than what editors believe to be correct, anyway; otherwise it seems to me that what would happen is that every so often partisans on each side change it to their way, believing science is on their side, and then an edit war happens if they arrive at the same time. Not great if you ask me now.

My point is not about whether they are wrong. (Also, the PTM surely do have some of the chemical weakness and propensity for oxoanion formation that you sometimes seem from TMs anyway, so whoever wrote that was not totally clueless.) The point is: this is what they show, this is what they're teaching students, and this is close to what most people are saying. That's why I'm supporting the LANL scheme. Your listing of antozonite (basically a very exceptional situation and not at all the usual occurrence of fluorine in nature) is a good illustration of why I'm not convinced by your point: I feel that we should be focusing on the general, not the specific, and following the general sources for a general colour scheme rather than a specific one. So you're not really addressing the grounds of my support for LANL at all and sorry, you're not convincing me away from the position that WP should follow something like that. Which, I add to reiterate, is distinct from the position that that is really the best way of doing things educationally when off WP. I take the position that WP should show something like LANL, but I don't take the position that LANL is the best scheme outside WP. Double sharp (talk) 01:32, 24 November 2020 (UTC)

Double sharp:

JChemEd as a proposer

Has become pretty conservative since they were folded into the ACS stable. They focus a lot more on technique than content.

Jensen arguments cf. Po arguments

He's not a good example due to the perceived selectivity of his arguments, and the fact that his proposal did not gain traction although, of late, there's been more interest in Lu in group 3. For polonium, OTOH, I'm not aware of any comparable sources in the literature arguing for its treatment as a metalloid. That's understandable given polonium's fully metallic band structure; metallic electrical conductivity; absence of a semiconducting form; and relatively straightforward cation formation.

Which elements are metalloids

I like to think the question of which elements are metalloids is no longer controversial following the publication of my 2013 article in JChemEd addressing this question (31 citations).

Even before then there was hardly any controversy as to the six elements commonly recognised as metalloids, per Goldsmith 1982, p. 526; Kotz, Treichel & Weaver 2009, p. 62; Bettelheim et al. 2010, p. 46; and Mann et al. 2000, p. 2783.

• Goldsmith RH 1982, 'Metalloids', Journal of Chemical Education, vol. 59, no. 6, pp. 526–7
• Kotz JC, Treichel P & Weaver GC 2009, Chemistry and Chemical Reactivity, 7th ed., Brooks/Cole, Belmont, California
• Bettelheim F, Brown WH, Campbell MK & Farrell SO 2010, Introduction to General, Organic, and Biochemistry, 9th ed., Brooks/Cole, Belmont CA, ISBN 0-495-39112-3
• Mann JB, Meek TL & Allen LC 2000, 'Configuration Energies of the Main Group Elements', Journal of the American Chemical Society, vol. 122, no. 12, pp. 2780–3, doi:10.1021ja992866e

Reliability of the LANL PT

I don't regard the LANL PT as a necessarily reliable source.

I distinguish here between LANL the organisation, in contrast to the LANL PT which was devised by someone in LANL who thought it would be a good idea to have such a PT, which it was. Unfortunately, the execution of that good idea was poor. For example, a simple search of the literature would've revealed quite a bit had been written about the status of polonium.

Teaching cf. RS

I follow what you are saying about what is taught rather than what is truth. But WP is not based on what is taught. It is based on (may we recall?) RS. Not just generalist sources but all sources, having regard to varying degrees of reliability e.g. less so; reliable: and top tier. Indeed. a theme of distinguishing RS based on e.g. publisher reputation, author reputation, age of publication, etc pervades WP:RS. When it comes to textbooks, some judgement is required in recognition of the textbook errors phenomenon extensively written about in JChemEd, for over ten years.

The antozonite example is notable since it was reported along the lines of textbooks needing to be rewritten. This kind of thing is what makes chemistry interesting. Rayner-Canham wrote along the same lines:

"We lie to students. Such lies are often justified by saying that we are “simplifying” the topic. But in the process of simplifying, are we hiding the fascinating "messiness" of the periodic table? In this chapter, I focus on two facets of the periodic table. First, that richness, complexities, and controversies, should be highlighted in our teaching."

Here again, Rayner-Canham was speaking in a teaching context. And he treats Po as a metal (a chemically weak one).

No doubt students and teachers, if they felt so inclined, could Google polonium and at the top of the page they would see the WP extract referring to it as a metal.

I forgot to mention Hawke's 2001 article in JChemEd in which he wrote, "…polonium is unambiguously a silvery-white metal" citing sources from 1990 and 1957.

The LANL PT represents an example of needless sloppiness.
--- Sandbh (talk) 10:09, 24 November 2020 (UTC)

@Sandbh: Well, Jensen did gain some traction. That's why Lavelle complained about people citing only him when they showed Lu in group 3. He just didn't gain enough to overturn the establishment totally, but without his article it is doubtful that the situation would have appeared on chemists' radar, it is doubtful that the popularity of La would have declined from its former overwhelming status to just keeping a plurality afloat, and it is doubtful that IUPAC would have instituted the project to make a decision. If you ask me, that's quite some traction. Not to mention that a majority of the specialised sources support Lu under Y. I do think that it is not enough to change things yet because we do not currently have a significant majority of basic sources showing La under Y, nor do we have an expectation that there will be. That may or may not change depending on what the ongoing IUPAC project says, but that is for the future. And BTW, even in 2017 JChemEd still published Kurushkin's periodic table with He in group 2 and Lu in group 3. That's not exactly super conservative for chemistry textbooks especially because of the first one. So the growing conservatism, while it may be a trend, is clearly not absolute. And BTW, it doesn't seem particularly different from the astatine situation: quite a few of the specialised sources support that At is not a nonmetal, but there's not 100% agreement, since Hawkes said it was a nonmetal. Meanwhile I strongly suspect that the usual picture of At in the basic literature is as a nonmetallic fifth halogen that looks like a black solid.

As for the question of which elements are metalloids, I think your own list says it all regarding the pre-2013 stats. That the usual six are common inclusions was generally accepted indeed; that Se, Po, and At are not also metalloids is not all that clear in the literature from that time. It would be interesting to see post-2013 stats. I also think that it is not exactly consistent to refer to Rayner-Canham in support for Po as a metal and not also take into account that he also supports Sb as a metal. Rather, I think that one should not use him as the only source when we know there is significant disagreement on both sides. And that's even though I think that scientifically his stand is the more consistent one. And it seems to me that if Rayner-Canham is describing what should be highlighted in teaching, then by definition of the word should that must also be what is not highlighted in teaching. Lastly: given that the issue is already controversial in the literature, I suppose you probably agree that other editors can have legitimate scientific views for these controversial topics that are not yours. So don't you think that this implies that if we all went for a scientific discussion, and attempted to assess scientific correctness of all the sources, we'd get nowhere?

The policy you cite, WP:RS, links when discussing primary, secondary, and tertiary sources the policy WP:PSTS. That points out that WP articles usually rely on secondary sources rather than primary sources. And that points out that tertiary sources can be helpful in providing broad summaries of topics that involve many primary and secondary sources, and may be helpful in evaluating due weight, especially when primary or secondary sources contradict each other. That seems to be rather exactly the situation we are in in drawing the PT and colouring it in. The policy WP:NC on article titles also reads The choice of article titles should put the interests of readers before those of editors, and those of a general audience before those of specialists. I think that principle is probably relevant here too. Comments at the Wikipedia:Arbitration/Requests/Case page from Beyond My Ken and SMcCandlish also suggests it.

In any case, none of this is really impacting my opinion for about the same reason. If your RFC doesn't gain a consensus, I still plan to open one for the 2010 (~ ACS/LANL) colour scheme. You are completely free to oppose it then, of course.

Also, why are we debating this at YBG's talk page in the first place? I came here only to tell him I was impressed but how much better he was at explaining my stand than I was. I'd therefore like to ask him if he would perhaps prefer this be moved to either my talk page or yours. Double sharp (talk) 10:57, 24 November 2020 (UTC)

The "put the interests of readers before those of editors" principle is a consistent theme (in one wording or another) throughout MoS as well (it's where AT got the idea, actually, back in the very early days before they split into multiple pages), and content guidelines. The point being: "for the readers" is something to keep in mind when making a table, building a template, and deciding on wording. It's also behind the wisdom of relying on reputable general-audience secondary sources (like university-level textbooks, and recent articles in the best science magazines), and even tertiary ones (e.g., encyclopedias of science), rather than highly specialized primary ones (esp. journal papers reporting new, unconfirmed research). In sciences, secondary academic sources (like systematic and literature reviews) are also of unusually high value, of course, but their geekiness often has to be moderated with general-audience material to produce something that many of our readers can grok. (A complex model that scientists agree this year is best will probably be experts-to-experts communication, and much harder material compared to whatever is most common in univ. textbooks, even if different subfields want to squabble about its accuracy from various perspectives).

All that generality said, I have no idea specifically how to arrive at what PT to use (and where – everywhere? by subfield?) on Wikipedia, nor whether it should attempt (on the basis of previously published attempts, not OR) to marry any different PT approaches. Given sufficiently complex template code, one could probably switch between alternative PT models (at least if the user has JavaScript turned on), but that would be a lot of Lua work, and we still need to know what the default view should be (whether that's universal or categorically varying).
 — SMcCandlish ☏ ¢ 😼  13:39, 24 November 2020 (UTC)

That's an interesting suggestion, SMcCandlish. I could definitely see a toggle-switch for the group 3 debate, since that's a relatively straightforward 2- or 3-option one: the space below Y is either filled by La, Lu, or * (the last one only common in some versions of the PT), and you'd be able to click to decide which one you actually see. I am interested in sounding out our resident ELEM template guru DePiep on just whether it could be done in principle should the need arise (i.e., not now). I have no plans to suggest it now, since the IUPAC project (they're the official body for this kind of thing, although they are not always listened to all the time by both specialists and generalists) is expected to say something soon, and whatever they say will have a significant impact on and probably clarify the current situation; so it doesn't make sense to me to make decisions that are likely to have to be relooked at (within some months, I would guess, although the road from first public knowledge of the results to the report and the recommendations may take a while). So, I will stick to my line of not reopening the subject till then. But it is something I plan to keep in mind as a possibility, and I thank you for it.

For the categorisation thing, it is not so straightforward. It's rather a little bit fractally complex. The majority of PT illustrations show categories – but there's disagreement on what exactly some of the categories are and exactly where their boundaries are. So it is a funny problem in which showing any categorisation is arguably undue weight to that one, but not showing any of them is also arguably undue weight because that's rather more of a specialist practice in the English-speaking world. (YBG already knows this, but I'm just recapping it briefly if it helps.) Basically, it's as explained by EdChem at his statement at the ArbCom case request. IUPAC is in this case not of very much help, since it has no plans to sort this matter out. Having a toggle would probably require a massive variety of options in this case, rather than just three, so it may not be very practical. I think this is best handled by what's been suggested by others to us: saying that we show a common category scheme that's near the centre of the range of variation, and just noting the boundaries and categories vary between authors. The precise differences can of course be described in our more specialised articles about each problematic boundary element and each category. I do not think we should worry about possibly compromising on accuracy when sources have widely differing views on just what is "accurate", anyway. As for just exactly what that common category scheme is to be – I have plans to put a candidate up for RFC once Sandbh's current RFC on the nonmetal categories is closed. As always, the result depends on consensus. Double sharp (talk) 19:48, 24 November 2020 (UTC)

Talkflow question: To discuss graphical support like toggle option, where to continue? This thread already does PT-content and Aim-At-Reader topics, so better not here ;-). (Started making notes). -DePiep (talk) 21:08, 24 November 2020 (UTC)

DePiep, does your talk page sound good? Double sharp (talk) 21:10, 24 November 2020 (UTC)

Categorizing: If that's in response to my mention of categories, I didn't mean the template should auto-place articles in categories, or auto-detect categorization. Rather, I meant that depending on how discussions go, it might be determined that the default PT for a particular category/field should be X while for another it should be Y, rather than there being one site-wide default. So we would need a parameter to set that.  — SMcCandlish ☏ ¢ 😼  21:46, 24 November 2020 (UTC)

@SMcCandlish: Oh, it's actually the bins the elements are placed into by the colour codes like {{compact periodic table}} does. Although I am not really sure there is an actual name for them in the literature: we just started calling them "categories" for talkpage-discussions to make it clear what we were talking about. Double sharp (talk) 22:37, 24 November 2020 (UTC)

Ah! Got it. Maybe we should use a fancy-pants word like tranches. LOL.  — SMcCandlish ☏ ¢ 😼  23:24, 24 November 2020 (UTC)

I just logged in a short time ago and was astounded at the alert count. I've never had 36 alerts before. YBG (talk) 04:45, 25 November 2020 (UTC)

@YBG and Double sharp: I continued the conversation here since I knew you were interested. Sandbh (talk) 23:54, 25 November 2020 (UTC)

Double sharp, I think our thread has gone somewhat off topic, which was about the LANL table and polonium. On a scientific evidence basis, as presented, polonium is a metal.

Astatine and its categorisation

A simple literature search produces some easy to read findings:

Unusually for a halogen, no solid evidence for the existence of diatomic astatine has been reported.

People look at the colour trend going down F-Cl-Br-I and think At would therefore be black. This is is simple textbook error given iodine is not the violet colour of its vapour but instead looks like a metal under white light.

Iodine has a band gap of 1.3 eV (hence it has a sub-metallic appearance) so it’d be reasonable to expect astatine to have a smaller band gap, and a sub-metallic or metallic appearance.

The nature of astatine: Timeline

Date	Event
1898	In announcing the discovery of Po, the Curies write: "We believe that the substance we recovered from pitchblende contains a heretofore unknown metal, similar to bismuth in its analytical properties. If the existence of this new metal is confirmed, we propose that it be named polonium in honour of the native land of one of us."
1917, 1923	Photoconductivity was observed in iodine (cf. selenium, another metallic-looking semiconductor).
1927	Herzfeld recognised a simple way of linking atomic properties, density, and the metallic state. Goldhammer probably recognised the idea in 1913. The result was the Goldhammer-Herzfeld criterion for establishing metallic character, based on classical physics. All this was noted by Edwards & Sienko in their 1981 JChemEd article on the occurrence of metallic character in the elements.
1940	The investigators who synthesised astatine said it was a metal.
1971	Based on periodic trends, Batsanov gave a calculated band gap for diatomic astatine of 0.7 eV.
1981	Edwards & Sienko, in their aforementioned JChemEd article speculated, on the basis of the G-H criterion for metallicity, that astatine is probably a metalloid.
1982	In JChemEd, Goldsmith wrote, "Describing the intermediate elements as metalloids is clearly in fashion today."
2002	In their concise book on the chemistry of the elements, Siekierski and Burgess presumed astatine would be a metal if it could form a condensed phase.
2005	In JChemEd, Thayer reported that the +1 state appeared to be the most common for At, occurring in At+, AtI, polyhalides, and a large number of monovalent organoastatine compounds. Of course, the astatide ion, At−, is found in salts. OTOH, +1 as the most common oxidation state looks unusual, given −1 is the most common for F to I. OTOH, it isn't unusual—it's inline with periodic trends of increasing metallic character going down each main group.
2006	Restrepo et al., reported that astatine appeared to share more in common with polonium (a metal) than it did with the established halogens. They did so on the basis of detailed comparative studies of the known and interpolated properties of 72 elements.
2007	Legut, Friák and Sob reported that the extremely rare cubic structure of Po is due to relativistic effects (54 citations).
2013	On the basis of relativistic modelling, Hermann, Hoffmann, and Ashcroft, predicted condensed astatine would be a monatomic metal, with a face centred cubic structure. Their work has been cited 35 times. They start with scalar-relativistic calculations which predict At will be a molecular solid with a band gap of 0.68 eV (cf. Batsanov, 1971). Then they add spin-order effects (approximated), since such effects scale roughly with nuclear charge. That changes the prediction to a monatomic metal with a body-centred tetragonal structure. They then finish by correcting for dispersion interactions. That changes the crystalline structure to face-centred cubic. They mention, but don't model, screening effects of the eventual metallic environment, and higher-order terms such as 'the three-body interactions of the Axilrod-Teller type'. Later, they refer to studying the viability of some alternative structures based on a 4-atom unit cell, with no dispersion corrections, but do not explain why they chose this cell size or the impact of excluding dispersion corrections. I'm a little wary, however, that their prediction of astatine as a metal is based on spin-order effects that scale 'roughly' with nuclear charge, and which are evaluated only in an 'approximation'. The authors do speculate however that, with active cooling to offset radiocative heating, it should be possible to synthesize enough astatine to test some of their predictions experimentally.
2013	The above was summarised, sans Legut et al., in my JChemEd article, Which elements are metalloids, cited 31 times.

Here we are 80 years after astatine was predicted to be a metal, now confirmed via relativtisic modelling.

And that is what the WP PT shows At as.

Re, "So don't you think that this implies that if we all went for a scientific discussion, and attempted to assess scientific correctness of all the sources, we'd get nowhere?" I say good luck to anyone attempting to argue that At is a nonmetal, period.

I suspect it would be more doable to reach agreement that At ought to at least be regarded as a metalloid. That would be a reasonable position to take.

i asked a chemistry professor today about At and he said, "I checked the periodic table on the inside cover of a number of general chemistry textbooks. All the ones that use color coding for metals, metalloids and nonmetals show At as a metalloid."

I guess, I don’t know, that he is talking about modern textbooks. That is consistent with what we know about At i.e. that, "In reactions, it sometimes acts like a halogen, sometimes like a metal."

The foregoing description is however clouded by the extremely low concentrations at which astatine experiments have been conducted, and the possibility of reactions with impurities, walls and filters, or radioactivity by-products, and other unwanted nano-scale interactions. Equally, as Kirby (1985) noted, “since the trace chemistry of I sometimes differs significantly from its own macroscopic chemistry, analogies drawn between At and I are likely to be questionable, at best.”

What to do then?

So what to about astatine? There are no reliable RS supporting its categorisation as a nonmetal

There are lots of RS pointing to increased metallic character compared to iodine. There are about three dozen secondary sources citing At as metal, based on relativistic modelling.

Its tough. If’s safe to call it as "not a metal". But is it a metalloid or a metal? The lack of any substantive evidence for diatomic astatine is almost a smoking gun.

Judging from my sample size of one chemistry professor, it may be safe to call it a metalloid. Whether or not it's a fully fledged monatomic FCC metal in condensed form is something we can address in, say, the different periodic tables section of our periodic table article.

Of course, this leaves the problem of what to call F to I. We could call them halogens (which they are) and leave them under the nonmetal super-category, without explicitly labelling them as "halogen nonmetals". That would be joint compromise solution incorporating your suggestion and my suggestion.

Then we’d be left with HCNOPSSe. The worst acceptable solution would be to call them other nonmetals, for the time being, in the hope that we can eventually settle on a better name. Judging from the RFC I think nobody cares, as long as whatever it is, appears in RS (in the same way that we picked PTM):

Metal						Metalloid	Nonmetal
Alkali metal	Alkaline earth metal	Lan­thanide	Actinide	Transition metal	Post-​transition metal		Other nonmetal	Halogen	Noble gas

At first glance, I suppose the specialist reader would say, "Hey! Isn't At a halogen?" Well, yes it is as a group 17 halogen. Upon seeing the legend, the specialist reader might twig that our PT is not a groupic table (hence we do not show the pnictogens or chalcogens). Rather, it is a metallicity-based table. Equally for the IUPAC-approved names, pnictogen, chalcogen, halogen, and noble gas, IUPAC has nothing to say about the nonmetallic, metalloidal, or metallic status of the elements of these sets of elements.

PS: I suspect the calibre and reputation of WP derives from expectations that WP editors doing their research, and weighing up RS, as we strive to do, and continuous improvement. The outcome of this process may not necessarily be what is taught since it seems to me we do have a responsibility to discount textbook errors we know to be such, based on more reliable RS. --- Sandbh (talk) 06:00, 25 November 2020 (UTC)

Sandbh OK, you're not convinced about At, but there has been a disagreement between us about Sb, no? Last I checked, you were of the opinion that it should be considered a metalloid, and I was of the opinion that it should be considered a metal (which Rayner-Canham does too). I don't think either of us has actually changed our minds. And since you've at some point considered the idea of calling the metalloids weak nonmetals, I suppose you can see Sb as a nonmetal too. So what then? Should we go out, metaphorically with guns blazing, against each other on WP just because each of us think that the other's philosophy of science is misguided? No, that will probably just get us both banned for disruption. But should we attempt to convince each other to the other side? No, that's pointless because we're coming from different philosophies of science. What I consider a categorical argument may quite possibly be seen by you as missing the point and vice versa. As was in fact demonstrated in the group 3 issue. Surely both of us can find a ton of sources to support ourselves anyway, but as DA said, "Citation is not an ultimate truth but just another opinion". Eventually, until we realise what's happening, the end result is going to be anyway that each party thinks the other simply doesn't understand how science works and it becomes like the first option anyway. Well – if we continue to make it about the science, and what the scientific truth of the matter is, there is no way out, because what I see as the truth of the matter is assuredly in many cases rather exactly not what you see as the truth of the matter, as was also demonstrated in the group 3 issue. As could probably have been predicted for any issue on which serious sources are actually divided, no matter how many obvious flaws we think the side we don't like has. The moment we attempt on WP to decide on a scientific basis how reliable a source is, we go beyond the duty of documenting the scientific consensus into contributing to it, which is rather explicitly not allowed; and, more importantly, how is it even possible to do that when we don't agree on the scientific bases? When the scientists don't even agree on the scientific bases? But it seems to me that the more WP-like, and probably only workable, option is: forget about trying to scientifically prove whether things are right or not, and note that WP:PSTS has shown us exactly which sources should be taken in a fraught situation like this one. On group 3, I agreed to pause the discussion because a IUPAC ruling is coming and it will change the calculus (and probably have to address some of those philosophy-of-science issues anyway), so it doesn't make sense to decide based on a situation that may change soon. No such thing is forthcoming for the categorisation issue, which is why I plan to RFC on simply following a common-ish category scheme. Sure, it will contain some things in it that can be seen as clueless. For example, we can rail against Ts as a halogen all we like. I even agree that it's stupid for an element that probably acts like a less melty version of gallium (at least if you could, you know, have it around for a significant amount of time). It simply remains that the general term among both specialists and generalists is "halogen" and that it usually, in English, includes Ts. That seems like something that can be admitted to me. If we say "oh, it's because the science says X" – we're applying it to something where scientists don't agree, so one side or the other is going to feel like it's the wrong version that's up and will be angry. Either they lose and feel even angrier, or they win and suddenly it's the other side that's angry. If we say "look, this is something like what a lot of people do; we follow the literature as a whole, so if you want to change it, start a campaign in the literature that gets noticed and accepted, or join one if it already exists, and be prepared for it to take a while" – then that will be clear and hard to argue against.

But, you clearly have a very different view of what WP should be doing here than I do, and neither of us is going to be able to convince the other, just like the philosophy of science issue. Thankfully, I think there's a simple answer: put it up for an RFC and see what the community thinks. If yours doesn't pass, then I'll do that, and the community will have a say. That should resolve things far more easily and with a lot less words than continuing this discussion. Double sharp (talk) 11:46, 25 November 2020 (UTC)

Antimony

Double sharp, yes, literature support for Sb as a metalloid is strong.

Yes, Sb chemistry is mainly that of a weak nonmetal, as is the case for the metalloids, generally, which is why I considered referring to them as (chemically) weak nonmetals. The chemically weak nature of the metalloids has been known for over 120 years.

Yes, we do have to exercise some judgment as to how reliable a RS is. Like PTST says:

"Appropriate sourcing can be a complicated issue, and these are general rules. Deciding whether primary, secondary, or tertiary sources are appropriate in any given instance is a matter of good editorial judgment and common sense, and should be discussed on article talk pages… Secondary sources are not necessarily independent sources."

The categorisation issue can easily be resolved by taking the best parts of our two ideas.

Could I check that you do not agree to categorising F to I as halogens, and At as a metalloid? Sandbh (talk) 00:22, 26 November 2020 (UTC)

@Sandbh: And that's the problem. I don't agree that Sb chemistry is mainly that of a weak nonmetal. Rather it seems to me to be mainly that of a very weak metal, almost on the borderline but not quite (closer to a metal than a nonmetal). Antimony doesn't readily form anions, but Po has a polonide chemistry. Antimony has a chemistry that is basically first (at strong acid conditions) the aqua cation, then (at slightly weaker acid conditions) some deprotonated oxycations, and then some oxyanions at even higher pH: why, that's more or less just like a standard transition metal or like Sn next door. Also, this is in the more common oxidation state +3 for Sb, which puts it ahead of Po whose cationic chemistry is mostly confined to the +2 state which has a bad habit of self-oxidising to +4 which is definitely less cationic (though maybe because of radioactivity). Just like Po, Sb forms salt-like compounds. Alloys of Sb with metals usually have metallic conductivity. Both Sb and Po have structures with few (six) nearest neighbours that hence probably cause them to be brittle (but in the case of Sb, not so brittle that you can't make a coin out of it). Sb is attacked by acid but more or less ignores alkali; Po is also attacked by acid, but is also attacked by hot concentrated alkali (p. 31). Sb doesn't have a stable form lacking metallic conductivity, and in the stable form it has metallic lustre that the other metalloids don't have (except As, which loses it in black arsenic anyway): the lustre of Si and Ge is more glassy than metallic. So I see physical properties that are more or less like Bi and Po, and chemical properties that are more or less like a normal transition metal (indeed, chemistry of Sb seems actually more metallic than that of Re). In general, Sb seems comparable as a metal to Po. And almost all of that is based on your literature review in the metalloid article. So what's stopping this from going round in circles between us just like the group 3 thing? This is why I think scientific discussion is not the answer to our problems.

Literature support for Sb as a metalloid is strong, indeed. Unfortunately, so is literature support for using "halogens" to cover everything to At, and now increasingly mainly Ts. So no, I do not agree to taking At out of the halogens. If "halogens" is a category, which I believe it should be, I would not agree to taking At out of it, and would be uncomfortable with my encyclopaedist's hat (though happy with my educator's hat) with taking Ts out of it.

I am pretty sure "good editorial judgement" does not mean "analysing the science ourselves". As a general rule, I would guess that when the amount of analysis we do is good enough to produce a journal article out of, we have almost certainly crossed the WP:NOR line. So I find your idea more or less fundamentally incompatible with mine because of those scientific disagreements and different readings of policies. But, there is an easy way out. I'll just put up my proposal for an RFC, provided yours fails, and see if it gains consensus. If it does, then it appears my reading was right. If it doesn't, then I can adapt to that and discuss again. Double sharp (talk) 02:02, 26 November 2020 (UTC)