Talk:Roman numerals/Archive 10

Archive 5

←

Archive 8

Archive 9

Archive 10

Biquinary?

In the Quinary article, there is a section Biquinary which suggests that Roman numerals follow this pattern. The section doesn't have any citations apart from the usage of Bi-quinary coded decimal but it obviously has the kernel of truth. The problem is that bi-quinary is a positional notation whereas Roman numerals are pre-positional: order is not intrinsically important (and crucially there is no notation for zero) for the obvious reason that positional notation wasn't invented for another 500-1,000 years.

The section also includes abaci of the Chinese and Japanese variants which are much closer to the early decimal computer systems. But it seems to me that Roman numerals are a proper system with alternating multipliers of 5 and 2 rather than a decimal system, even if that's what we always think of it as. The vocabulary is I, V, X, L, C, D, M, which shows this sequence quite adequately. The subtractive notation for IV, IX etc. came later. Without that addition, it is entirely unambiguous regardless of ordering.

But I can't find any good mathematical source which recognizes this. I have Ifrah's "From One to Zero" but that was published (in English) as long ago as 1985 and I'm not near enough a good library to find more recent sources and Google doesn't help.

So is biquinary the correct term or is there some better mathematically accepted term? I ask here as it's probably watched better than the Quinary article. Chris55 (talk) 20:39, 8 April 2021 (UTC)

We've been through this one - although it's certainly an interesting idea! Roman numerals are really more decimal than (bi)quinary - the primary RN notation is actually I, X, C and M: with V, L and D having an essentially intermediate function. Quinary Roman numerals would likely use "VVV" for 15 rather than the essentially decimal pattern inherent in "XV". But the scare quotes around "base ten" in the article text are there for a reason - strictly speaking, to be "base anything" implies positional notation, doesn't it? Anyway, much more relevant to the specific (quinary) article. Heaven deliver us from controversy on the topic on this page! --Soundofmusicals (talk) 23:15, 8 April 2021 (UTC)

Yes, that's the way it's worked out in practice. But it conceals a far simpler and more elegant system. Your point about "VVV" is incorrect. I'm not suggesting it's a quinary system but a biquinary system: no more than 4 I's (or X's, or C's or M's) and no more than 1 V (or L or D). In regular expression notation it would be:

M{0,4}D{0,1}C{0,4}L{0,1}X{0,4}V{0,1}I{0,4}

It's easy to see how this simple system got perverted: with IX there are 3 less grooves to chisel out on the stone than VIIII and it's easier to read in the bargain.

But I was not asking for this to be put on the page at this stage. I was asking for a decent reference and a more specific name, if available. If there isn't one, then I may have to write it myself :) (It won't be the first time...) Chris55 (talk) 08:01, 9 April 2021 (UTC)

If there really IS no recognised "specific name" and no "recent/decent" reference you may even be the very first in this particular field. In "Wikipedia-speak" this is "original research". In this case OR of the most literal and laudable kind, but OR nonetheless, and therefore not what we do here. But of course you already knew this, hence, I take it, the :)! My point was that this discussion belongs on the talk page for quinary rather than here - in fact the categorical statement that RNs are an "early biquinary system" is a bit questionable - it really does need a good source. Still a matter for that article (and its talk page) rather than here though, at least until/unless we decide that RNs ARE really biquinary after all. --Soundofmusicals (talk) 10:44, 9 April 2021 (UTC)

'biquinary' refers to one of two orientations for subradices 2 & 5 under decimal, in which the digits are sorted into 2 groups of 5 each (as opposed to 5 groups of 2 each, which is far less common). RNs do resemble a biquinary decimal system, with the reservation that it's not a place-value system. I considered this point myself, but decided it was a bit too technical, as well as veering on OR (unless there are sources comparing biquinary to RNs). Xcalibur (talk) 14:52, 9 April 2021 (UTC)

Just a couple of points to add here. First, Soundofmusicals' suggestion that Chris had invented the term biqinary is absolutely incorrect. I first came across the term in the 1960s when Dad taught me how to count to 100 on my fingers in the Roman fashion, rather than 10 in the primary school method. Next, assuming that Roman numerals are not biqinary because they don't use a positional system is also palpable rubbish if one then calls them decimal! Colloquially that's called having your cake and eating it. :-o I'd even suggest that this is no more OR than observing that the numbers count in fives and two alternatively – which clearly everyone since Romulus was a lad knows. Martin of Sheffield (talk) 16:32, 9 April 2021 (UTC)

Roman numerals certainly are not "decimal" either, but trying to edit garbage out of this article is almost hopeless. Decimal implies positional notation, and so does bi-qunary. And bi-quinary means a position with 2 symbols alternated with a position that uses 5 symbols, which is not what is happeing here. You can claim decimal is bi-quinary because we use two sets of digits (those less than 5 and those greater or equal to 5) and you would be making just as much sense. Lets just stop this, you do not need to bloat this article with nonsense because the word sounds intellectual and you are proud of finding it.Spitzak (talk) 16:42, 9 April 2021 (UTC)

"because we use two sets of digits (those less than 5 and those greater or equal to 5)" – really? If you want to argue that a system that does not use positional notation cannot be described in terms of bases, then fine. A mildly eccentric but consistent point of view. Perhaps you would like to summarise "a system counting alternatively in fives and twos using non-positional symbols for the powers of ten" in a way that does not encourageforce bloat. Martin of Sheffield (talk) 17:04, 9 April 2021 (UTC)

RNs do resemble decimal and bi-quinary, though they don't use place-value. the roman abacus had a bi-quinary design, which worked well with the numerals. after a brief search, I found a couple references, although they're not RS: [1] [2] Xcalibur (talk) 17:13, 9 April 2021 (UTC)

Interesting suggestions and quotes. But it's easy to confuse the bi-quinary used in abaci and early computers, which undoubtedly are decimal positional systems with the actual Roman numerals which clearly aren't. If they were, the symbols at each power of 10 would be the same, but they're not. It underlines why I'm uneasy at using the term biquinary. One can extend the term, but this goes against scientific ethos. Perhaps we call it pure biquinary. Chris55 (talk) 21:37, 9 April 2021 (UTC)

It's worth having a quick look at Sexagesimal. The proto-cuneiform variant uses separate symbols for 1, 60 and 3600 so would fail according to some of the posts above. However the same system was also represented by place notation by the Babylonians and so would qualify. Surely this is the crux of the issue: the sexagesimal system is represented by both positional and non-positional notations; likewise the Roman system is bi-quinary, but is represented by a non-positional notations. Oops! Sorry about the signature. Martin of Sheffield (talk) 07:51, 10 April 2021 (UTC)

(The last post before this is currently unsigned! A good, thoughtful post, so sign it, whoever!)

Pace Martin of Sheffield but I certainly didn't say that the use of the terms "quinary" or "biquinary" would be OR, but was referring to Chris55's desire for a "more specific term": and his (as I took it) light-hearted threat to make one up! Good to see the old "RN Gang" coming together again, by the way, but we HAVE been over this ground at least once in the past. I do hope any changes this iteration produces will enable us to mention the basic "decimality" of RNs without going too much into technical hairsplitting that might leave our readers more confused than ever, and Spitzak bewailing "bloat". Still, it is always pleasant to be as accurate as we well can be, bearing in mind that this is not a specialised teatrise but an article in a general "popular" encyclopedia. Perhaps a footnote at the place we call (with slight inaccuracy) RNs "basically decimal" pointing out that they can also be described (or have described, with good sources) "quinary" or "biquinary"? Encourage the earnest reader to seek further information without bogging down a succinct summary for everyone else. --Soundofmusicals (talk) 23:17, 9 April 2021 (UTC)

Yes,"quinary" and "bi-quinary" have been discussed here several times over the years.[3] I do fear that using either in the article might teach the reader a new word but might not teach them its correct usage, would pull them up short and would make the article seem more impenetrable. "Biquinary" is not a clear and readily understandable word. The reader that does see how to break it down might recognise the quinary as V but look in vain for a Roman symbol for 2. OTOH, it's true "decimal" fits Roman words for numbers better than their numerals. If I was the only one editing the article, it might briefly describe RN as a system of ones, fives and tens, and move sharply on! NebY (talk) 01:35, 10 April 2021 (UTC)

I might dispute '"Biquinary" is not a clear and readily understandable word', but otherwise I think that is a very sensible suggestion. Avoid decimal and biquinary althogether. Martin of Sheffield (talk) 07:51, 10 April 2021 (UTC)

There has been an article in Wikipedia called "Bi-quinary coded decimal" since 2004, but it received its first reference in 2013 and it's not clear even now where the term biquinary (also used in the article) came from, though the OED cites a Bell Labs report from 1946, and a 1949 quotation from Hartree. The sense used there is very different: 2 bits for 5 and 5 bits for 0-4; designed for error checking and simple logic. I realise that this article has been plagued by over-long dissertations on "orthography" which is why I gave the one-line definition above. The UK is still in lockdown and libraries don't open until Monday so I haven't had a chance to check up. No, Soundofmusicals, I wasn't joking about having to make up a new term, but I'm not yet convinced it is necessary. But the number of responses to this thread suggests the issue touches a nerve. Chris55 (talk) 11:57, 10 April 2021 (UTC)

I wouldn't call my previously proposed section a "dissertation" by any stretch. I agree that the debate dragged on too long; partly that's my own fault, since I was hoping for a breakthrough that never came. Trying to add a single brief reference as a "compromise" was also a mistake, since it only dragged it out further (I dropped the main proposal in September 2020). anyway, as I said, RNs are not quite bi-quinary decimal, since they don't use place-value, but they are organized on similar principles. each power of Roman Numerals is organized into two groups of five digits each, punctuated by the 'fives' numerals. that is biquinary, as opposed to qui-binary, which would be 5 groups of 2 digits each. as I also said, I considered adding this, but explaining it in a way that's not overly technical or tangential would be tricky. this may be better suited to another article. Xcalibur (talk) 15:22, 10 April 2021 (UTC)

@Xcalibur (talk) I do hope we're well and truly over the (totally unrelated) topic your last post seems to want to return to! I agree however that the question raised by THIS threaad is really "better suited to another article" and that until the quinary article itself is better "finished" it may be better not to tamper with this one, at least for moment. --Soundofmusicals (talk) 01:55, 11 April 2021 (UTC)

the new arrival made a not-so-subtle reference to my shenanigans on this talk page, so I felt the need to address that. I won't continue that topic further unless someone requests that I do so, because I've had enough of playing tennis with a brick wall. anyway, perhaps an analysis of the biquinary properties of Roman Numerals would be a better fit for Bi-quinary_coded_decimal, just an idea. Xcalibur (talk) 02:09, 11 April 2021 (UTC)

Xcalibur, or Bigdan201 as you are these days, I doubt that a nearly 7,000 character addition is the best way to sell yourself to this excitable community.:) My starting point in this topic was an entry in the Quinary article which read: "A decimal system with 2 and 5 as a sub-bases is called biquinary, and is found in Wolof and Khmer. Roman numerals are a biquinary system." My only change to that was to add the word "early" before the second mention of biquinary, and to add that Roman numerals are not a positional system. I do agree that this page is lacking a simple explanation of how Roman numerals are formed but I have no intention of trying to impose my own view without some adequate citation. And as you have demonstrated, most of what is available is the efforts of maths teachers around the world to introduce a poorly understood topic. I haven't yet found any language or number theorist who has given it proper attention, which is surprising having spent a proportion of of my professional life in those fields. What we are dealing with of course is the slow evolution of common language into a more mathematical formulation and this inevitably involves numerous inconsistencies. Chris55 (talk) 09:10, 11 April 2021 (UTC)

first of all, my numerous references make up a fair chunk of the character length. my sources vary in reliability, but some are RS, and they all converge on the ruleset I compiled. it's true that historical usage was highly variable, and the convention is a modern phenomenon (ie post 1800). if I could put my many arguments into a nutshell, it would be this:

why are VL, IC, IIX, and XXXX not considered standard? where does the article address this, it doesn't.

however, content is decided by consensus here, and there's nothing more I can do. as for biquinary, that's a good start. I was thinking that you could develop it further by analyzing how RNs resemble a biquinary system. to illustrate, biquinary is like this: I (0, 1, 2, 3, 4) II (5, 6, 7, 8, 9) which are two groups of 5 each. the far less common quibinary is I (0, 1) II (2, 3) III (4, 5) IV (6, 7) V (8, 9); 5 groups of 2 each. leaving out subtractives, RNs are similar to biquinary in that you add up tallies until they 'flip over' to either a 'fives' numeral or the next power of ten. while this relates to both biquinary and RNs, it seems more relevant to biquinary, so I maintain my stance that this should be fleshed out on the biquinary article. Xcalibur (talk) 15:31, 11 April 2021 (UTC)

on a tangent, I found an interesting source which covers RNs and their origins: [4] Xcalibur (talk) 17:38, 11 April 2021 (UTC)

I've had a quick look at that site, which appears to be aimed more at children. It's good that it shows finger counting as the origin of RNs, particularly I and V, it's not as clear why V+V -> X. It also misses out that one hand counts units and the other 10s giving a total of 100 which is more than enough sheep for any shepherd. The other criticism is that it attributes Arabic numerals to Fibonacci. He certainly helped with popularising them, but they had already been in Europe for between 200 and 300 years. The Codex Vigilanus (AD 881-976) appears to be the first mention of them in Spain. They had reached Britain by 1115 as witness page 9 of the Textus Roffensis which has a gloss of them. Martin of Sheffield (talk) 19:34, 11 April 2021 (UTC)

I'm not sure if it qualifies as RS. however, it states that the first Western mention of hindu-arabic numerals was by Boethius (although the zero was omitted), and the illustration of numerals goes back to the 10th century. it has some useful points, and describes RN as biquinary, which is why I linked it. Xcalibur (talk) 20:04, 11 April 2021 (UTC)

Eddy and Khan

This reference does indeed include a "set of rules" (not much like any rules ever proposed for here) - and at this point it is indeed not relevant - assuming we are no long mentioning rules (as per long standing consensus only ever opposed by one editor), however it is a full length book, and on page 3, immediately following the "rules" it has a table very similar to ours. Suggest that in current context it is perfectly relevant! --Soundofmusicals (talk) 03:26, 23 December 2020 (UTC)

you're right, there is a table, somehow I missed it. that means the ref can stay. eta: this also addresses my concern that the section didn't have proper sourcing/support.

adding on- while the table is sourced, you'll still want a ref to support the claim that there's a standard form/conventional notation for RNs. otherwise, I'm afraid that counts as OR. Xcalibur (talk) 08:08, 24 December 2020 (UTC)

I'm not inclined to get into even more fruitless debates, so I'll leave a brief note under here. I don't know why you insist upon blatant overuse of scare quotes throughout the article. "quotes" are used for referencing what someone said, they don't belong on every term referenced. I've used apostrophes 'like this', I'm not sure how proper it is, but it seems like a useful way of emphasizing certain words/phrases while avoiding the scare quote effect. italics can be used for emphasis, but that is discouraged; they're mostly used for highlighting foreign and technical terms. I tried to mop up the article accordingly, but if you insist on a weird, pointless style, I won't press the issue. Xcalibur (talk) 05:15, 29 March 2021 (UTC)

an update to all concerned: my edits on April 1 were a stunt I pulled for April Fools; if the rules hadn't been reverted, I would've reverted the content myself that evening. at least JohnUniq understood, which is impressive, given that he thinks the Romans had the printing press/movable type 1000+ years before they were invented, and he thinks there are criminal penalties for using variant forms of RNs. no one suggested those things except you; those are among the dumbest, most disingenuous things I've read, even worse than Spitzak believing that sources with similar/overlapping content must be cut & pasted, Kahastok's claim that a brief reference is exactly the same as an entire section, and so on. just thought I'd clear things up. Xcalibur (talk) 14:41, 3 April 2021 (UTC)

Just for anyone watching here, April Fools Day is not an excuse for edit warring contentious material into articles. Most people find this to be common sense, but for those that don't, we do have WP:FOOLS.

Nor is it an excuse for the series of personal attacks in the above message. Kahastok talk 14:59, 3 April 2021 (UTC)

I was mostly responding to statements made here and in an edit summary. especially: rv: consensus is against false suggestion that the Romans published rules and Is the suggestion that retrospective sanctions would be applied to anyone who broke the rules carefully documented in the two references? the Romans didn't publish anything -- in classical antiquity, texts were copied down by scribes. also, I stated that the convention is from the modern era (post 1800). if you move a rook diagonally, you won't be arrested, but it's called an illegal move in chess. I'll admit, it was a little out-of-bounds, but the same could be said for the following: Otherwise, please use another website to post opinions on life and rules. as well as Just shut up. Or if you really really think "rules" are better, REPLACE THE EXISTING TEXT!!! Or go away. most of all, making up strawman points and pretending I said them is obnoxious, perhaps moreso than anything I just said. I simply wanted to speak my mind, I beg your pardon if it was too hostile. Xcalibur (talk) 23:20, 3 April 2021 (UTC)

From Wikipedia:Rules for Fools: 'Jokes that affect articles ... will be treated as vandalism ... you risk having your account possibly blocked from editing.' Johnuniq (talk) 01:16, 4 April 2021 (UTC)

I was going to respond with "but I thought you didn't like rules?" but that might be pushing my luck. I get that, I realize that it was mildly disruptive and not allowed. I still think that "Romans publishing rules" and "sanctions for breaking them" are total misrepresentations, but that's no reason for me to lash out. that's all. Xcalibur (talk) 23:32, 4 April 2021 (UTC)

finally: I think my biggest mistake here was trying to add a single brief rules/logic reference after I dropped the section in September 2020. however justified it may have been, it only extended the mess another 3 months. I should've realized that would happen. Xcalibur (talk) 15:10, 7 April 2021 (UTC)

postscript on rules/logic debate

I was re-reading all this, and wanted to share some thoughts. it occurred to me that one reason discussion dragged out too long is because I wasn't focused enough on the central issues, so I'd like to address those now. this isn't an argument for inclusion, after all, I dropped the main proposal in September 2020, and there's now a consensus against any mention of rules/logic in the article. as an aside, why didn't you form that consensus earlier? the 'standard form inflation' section formed a speedy consensus for a line of text under Standard Form, yet when I substantially changed the wording to drop the reference to a 'convention' (which was unsupported), no one objected. this makes it obvious that you rammed through the consensus to shut me up, and if you wanted to shut me up, why didn't you proscribe any mention of rules/logic in the first place? that seems like the obvious solution, and would have cut it short by a few months at least. anyway, for educational purposes: Our hindu-arabic number system is built on the digits 0123456789. These digits are increased by powers of ten according to position, which is known as place-value notation. Roman Numerals use the digits IVXLCDM, where powers of ten are built into the numerals themselves. To form a value, you must add & subtract the digits from each other. Thus, RNs are built on the additive/subtractive principle, which is a very different foundation than place-value. Of course, adding/subtracting allows many possibilities: 45 might be XLV, VL, or XXXXV; 99 might be XCIX, IC, or LXXXXVIIII. However, it's apparent that there's a single form of RNs used in books, Star Wars episodes, Super Bowls, etc. (with the notable exception of IIII on clockfaces). If there's so much leeway in adding/subtracting, yet only one accepted convention, *how do you restrict addition/subtraction to the convention?* THAT'S the question I was attempting to answer. that's why most of the criticisms made against my content were not valid. the claim that "I can just list the subtractives instead" is not true, because just listing combinations doesn't explain why they are correct, it doesn't cover the underlying logic; it's a different approach. the claim that I can insert "use the shortest string" is wrong, because it's a shortcut that doesn't directly address addition/subtraction in an assembly code sort of way, and is not in the RS besides. and the claim that "my content duplicates the existing article" is wrong, because the article doesn't address the inner logic of addition/subtraction, it doesn't explain why VL, IC, IIX, IIXI, or XXXX are wrong. that's not to say there are no valid critiques, because there are. you could say that my work is too technical for a general use article, or that it's better to have one explanation rather than two different ones; I respectfully disagree, but at least those points are reasonable. as to the rest, I still haven't seen a source that contradicts my work, or any example of how I'm not following RS, and I haven't seen a single combination of RNs not addressed by my rules. eta: in my examples, the Roman Numeral font is applied only to correct combinations, not incorrect; this is an intentional stylistic choice. not that this will make any difference of course, the debate is long over. I just wanted to address the main points for any who may be interested, since I didn't do this sufficiently before (and for that reason, the fruitless exchanges were partly my fault). also, my second main paragraph above may be of use for the Description section. and again, even though I thought I was personally attacked first, retaliation was not justified. this is probably the last I'll say on this, unless others discuss it further. pardon the long post, I had alot to say. take care. Xcalibur (talk) 00:30, 6 May 2021 (UTC)

Isaac Asimov theorises

Very interesting article. Conscious I'm being pedantic, but should "Isaac Asimov theorises" be restated as "Isaac Asimov theorised"? The man is no longer alive. I'm aware of Historical present but I'm not sure if it applies here in a more formal setting. — Preceding unsigned comment added by 46.7.0.179 (talk) 15:48, 5 May 2021 (UTC)

I went in and changed it to 'theorized'. Xcalibur (talk) 00:33, 6 May 2021 (UTC)

It turns out Asimov doesn't say its his theory, so I've edited that, opting for the past tense. I understand Soundofmusicals' reason for inclusion though I'm disappointed to hear the theory's gained currency. I do wish we could just leave it out, especially as technically we shouldn't editorialise about it and call it speculation (even though Asimov himself, frankly, seems to realise it's amusing but weak), so I've tried to approach taht with a small fact instead. NebY (talk) 08:36, 8 May 2021 (UTC)

Semi-protected edit request on 16 May 2021

Fix incomplete table: (see below)

Base number	I = 1	CIↃ = 1,000	CCIↃↃ = 10,000	CCCIↃↃↃ = 100,000
1 extra Ↄ	IↃ = 500	CIↃↃ = 1,500	CCIↃↃↃ = 10,500	CCCIↃↃↃↃ = 100,500
2 extra Ↄs	IↃↃ = 5,000	CIↃↃↃ = 6,000	CCIↃↃↃↃ = 15,000	CCCIↃↃↃↃↃ = 105,000
3 extra Ↄs	IↃↃↃ = 50,000	CIↃↃↃↃ = 51,000	CCIↃↃↃↃↃ = 60,000	CCCIↃↃↃↃↃↃ = 150,000

Srs42006 (talk) 19:36, 16 May 2021 (UTC)

 Done RudolfRed (talk) 20:22, 16 May 2021 (UTC)

superscripted use

In 16th c. texts there were use examples rather closely resembling the positional system, with thousands (M) or hundreds (C) being written/printed in superscript after the multiplier. E.g., 400 was IIII^C etc. I see rather a lot of those in digitalisations, but to put that into the article would constitute an OR, right? Yury Tarasievich (talk) 21:13, 22 November 2021 (UTC)

If you can provide a link to an example it would go a long way to avoiding WP:OR. Martin of Sheffield (talk) 21:22, 22 November 2021 (UTC)

There are examples of this already in the article, but it is done rather confusingly and seems to fail to realize that a bunch of examples are actually related.Spitzak (talk) 22:37, 22 November 2021 (UTC)

Under "Rare Variants" this is indeed already covered. Doesn't mean the text couldn't be improved of course, in fact it might deserve a new section? --Soundofmusicals (talk) 07:28, 23 November 2021 (UTC)

Well spotted you two! Martin of Sheffield (talk) 09:24, 23 November 2021 (UTC)

Well spotted, yes, but that shouldn't be a spotting subject ))). The text is not well written because of that, and because those do not look 'rare'. So, need more examples, right? Yury Tarasievich (talk) 14:11, 23 November 2021 (UTC)

Someone here seems to be confused by the phrase "well spotted". This refers to someone who has noticed something (anything) already covered by a request for "more information". There is no such thing as an article that does not need scrutiny as to whether suggested additional material is already covered. --Soundofmusicals (talk) 23:29, 24 November 2021 (UTC)

Semi-protected edit request on 2 February 2022

Please add Template:Pp-semi-indef to this page. It's semiprotected, but the little icon is missing from the top right corner of the page. 122.150.71.249 (talk) 00:15, 2 February 2022 (UTC)

 Done casualdejekyll 01:50, 2 February 2022 (UTC)

"Sources" for this topic!

A Google search on "Roman Numerals" does not result in a list of "reliable sources" for an encyclopedia article. To be fair, such was never the intent of the authors of most of those sources. While a great (even exclusive) stress on definable "rules and principles" is appropriate for a "how-to-do-it" approach, that is not what we are trying to do here. The majority of our own recent improvements to the article have demonstrated just how inconsistent and variable Roman Numerals have always been. The current (well argued, in fact, hard-won) consensus on this point is well worth defending. --Soundofmusicals (talk) 04:28, 11 May 2022 (UTC)

I was just trying to tweak the wording, I wasn't making an assertion. As for sources, they vary quite a bit. Some in fact qualify as RS, others not so much. They do however make it clear that there's a commonly accepted standard, along with alot of inconsistency and variety (they also include rules alongside a general explanation, and I never saw a reason to exclude this, but I respect the consensus). Thing is, if there's no standard or convention, why do we have a table of RNs in a certain style?

There's both a modern, post 1800 standard style, and alot of variation; both are true at the same time, which makes it a little complicated. Some of my sources from awhile back made it clear that in the ancient world, there was wild variance (the craziest one I saw was IXCS for 89 1/2!). It's safe to say that there was almost no standard in antiquity, somewhat more consistency but also experimentation in the medieval era, then the emergence of a modern standard alongside variants. The article should reflect this. Xcalibur (talk) 20:19, 12 May 2022 (UTC)

More or less what it already does "reflect", surely? Just one more very good reason for not including "rules" (although far from the only reason). On the other hand the table does succinctly list the USUAL modern forms (labelled as such). We need something of the kind for the sake of our readers. Please (pretty please) DO respect the consensus. Repeated implications that it (the consensus) is invalid for various irrelevant reasons are not helpful. For the record, from first to last, people have been very patient with you - in spite of the odd instance of irritation. --Soundofmusicals (talk) 18:23, 15 May 2022 (UTC)

Most importantly - we need in an online encyclopedia on a subject like this to avoid matter that is speculative, judgemental or plainly contrafactual - no matter how well "sourced". --Soundofmusicals (talk) 18:39, 15 May 2022 (UTC)

The article does reflect this state of affairs, albeit in a somewhat muddled way. I guess that's my main criticism: the article could be more clear and concise in describing aspects of Roman Numerals, specifically that usage can be roughly divided into ancient, medieval, and modern.

I don't actually mind that my content was rejected. Disagreements happen, and sometimes they can't be resolved. The main reason I responded excessively is because I never saw the reasoning against it, and I was personally attacked first. But anyway, that's water under the bridge. I don't think I was defying the consensus, unless you've become allergic to the word "rule". Xcalibur (talk) 20:09, 23 May 2022 (UTC)

Sad you still feel that the current article is "muddled", although no one (including me) has ever claimed it is perfect. Several of us besides you have given up personal ideas in the interests of consensus. Have you ever actually read it right through? The current article does include treatment on historical forms and uses of Roman numerals in its later sections. It is probably a good thing to get the basic description (including variant forms) over before we start looking these supplementary topics, however. --Soundofmusicals (talk) 00:49, 24 May 2022 (UTC)

I have read through the article; it's decent and fairly comprehensive. I guess it's just an organizational/structural issue. I agree that it should start with a basic description, and then branch out. Xcalibur (talk) 01:19, 24 May 2022 (UTC)

Square boxes error

I've noticed that there are square boxes meant to represent Roman numerals. Looking at the text, it's replaced by a box with a question mark inside (⍰). It feels like an error and should be fixed. It's displayed in the table titled Other Roman fractional notations included the following: and in the subsection Etruscan numerals. Callmemirela 🍁 ^talk 16:23, 21 February 2022 (UTC)

This problem is a result of the display capability of the application (or even the device) you are using to view Wikipedia. If you are using Windows on a PC desktop or laptop you can probably enable the intended display of these characters by changing your settings. There is a standard warning about special characters at the head of the article.--Soundofmusicals (talk) 04:42, 11 May 2022 (UTC)

See also Help:Special characters. --Soundofmusicals (talk) 19:29, 15 May 2022 (UTC)

Big Ben...

This (quite incorrect) nickname occurs not only elsewhere in this article, but also as the title for the article on the subject itself. Might one suggest it might just be a little more helpful to include it, wrong though it is? Soundofmusicals (talk) 18:34, 6 July 2022 (UTC)

I tend to correct it when I happen to spot it, but don't go looking for it. WP has a problem with several articlew where a nickname is used rather than the correct name. Personally I don't like it and think that an article should follow the correct names (without wandering off into blind alleys about what constitutes an "official" name). However, I'm just one lowly (15k edits only) English editor so I'll bow to others' judgement. Martin of Sheffield (talk) 18:44, 6 July 2022 (UTC)

All other things being equal I have a strong preference for correctness myself as you may have noticed. Has the clock (as opposed to the bell) actually got a correct name that most readers would recognize? If so it would certainly make sense to use that. As it is you may notice I didn't revert you but added back a reference to the "wrong term" for clarity. If you have a strong preference for your text I wouldn't argue further. Soundofmusicals (talk) 00:45, 8 July 2022 (UTC)

No, I'm more than happy with your phrasing. "Commonly known as" clearly indicates a nickname and is indisputably correct. Thanks. Martin of Sheffield (talk) 07:26, 8 July 2022 (UTC)

As a rule...

This phrase may be a useful clarification - but surely not in the same sentence as "usually" (!) Soundofmusicals (talk) 18:29, 6 July 2022 (UTC)

That's actually a valid criticism. I was just experimenting with the wording, that's all. I don't think it counts as a proper mention of rules/logic as described by the sources. Btw, the whole reason I tried to add a brief mention before was to back up the claim that there's a standard form, I really thought that was an acceptable compromise. I guess the table is enough, though I'm still not sure why one is allowed but not the other. Also, believe it or not, I didn't lash out over controversy/rejection of content, that's fine and I can accept it. Rather, it was all the attacks made against me first. Xcalibur (talk) 18:59, 6 July 2022 (UTC)

I've said it before - but people have been very patient with you on the whole. Going on about "all the attacks made against you" has at the very least gone on far too long - even if it had anything to do with improving the article. Soundofmusicals (talk) 01:05, 8 July 2022 (UTC)

I suppose that's true. Most of all it was the slander & passive-aggressiveness that set off my response (not referring to you). It's not like I went to all that effort just to troll ppl. But at least you ended up having some useful input, even if I was unable to add encyclopedic content to an encyclopedia, which I think is rather backwards. Xcalibur (talk) 03:33, 23 July 2022 (UTC)

Using Unicode points for numerals?

Should we be using the Unicode points for all the Roman numerals in this document, for example, “Ⅰ, Ⅱ, Ⅲ, Ⅳ, Ⅴ” instead of “I, II, III, IV, V”? I would be able to take care of it if it is a good idea. Leejordan9 ✉️ 🛠 19:18, 20 January 2022 (UTC)

Per Numerals_in_Unicode#Roman_numerals, Unicode Roman numerals are intended for compatibility only, and they say that in most circumstances it is preferable to use ordinary letters. Using relatively obscure Unicode characters risks some readers not being able to read the text, and risks accessibility issues for e.g. screen readers. I would thus oppose any conversion. Kahastok talk 19:44, 20 January 2022 (UTC)

Thanks for clarifying! Leejordan9 ✉️ 🛠 20:50, 20 January 2022 (UTC)

• There was a "special characters" template on the page, which I have removed, since it does not actually apply. Imaginatorium (talk) 15:32, 16 February 2022 (UTC)

Hope I am forgiven for reinstating this template - which does still apply, at least in the few instances where there is actually no more accessible alternative to the odd Unicode character. On the whole we do avoid Unicode Roman numerals for the reasons stated by Kahastok. -Soundofmusicals (talk) 02:03, 24 July 2022 (UTC)

Update

I had an idea, which I never fleshed out during discussion, so I'll mention it now. One way this article could be expanded is with a section on arithmetic with roman numerals. This is discussed in one of my sources, namely [5]. It's a possibility, at least. However, I won't be the one to do it, because the last time I tried to improve this article, I received nothing but personal attacks, disingenuous nonsense, and gate-keeping in return.

to tie up a few loose ends: That is not to say that the status quo cannot be improved. It can. But these apparently-endless discussions are not improving it. that's funny, since my involvement here ended, this article has been mostly dead with few edits, just as I predicted. Isn't that what you'd call a "logical nonsense"? but congrats on defeating me, I guess! also, This reference does indeed include a "set of rules" (not much like any rules ever proposed for here) it's actually about 70% similar. That's the thing, I never saw a valid point made against the proposed content, and why would you balk at a brief mention that follows the source? It just seems to be a power/gatekeeping thing, where if you let me so much as add a mention, that's invading your territory. I was hoping the process here would be objective, where you'd consider points with impartiality, and if there are no clearly stated objections, then content goes up. instead, it was alot of IDONTLIKEIT and WEDONTLIKEYOU, which is unfortunate. progress towards an ideal encylopedia this will never be a serious encyclopedia, so long as it's run like this. I think WP will continue to be what it is, a giant cliffs notes of varying quality and reliability.

Finally, if anyone here is willing to change their mind and assent to the inclusion of logical analysis drawn from RS, then it's never too late! All offenses would be forgiven if you repent. In the meantime, the suggestion/source at the beginning stands as an alternate/additional means to improve the article. Xcalibur (talk) 09:11, 24 November 2021 (UTC)

Your reference does claim that they converted to additive, then did math with the individual symbols, reducing the result by sorting it and then replacing excessive numbers of the same symbol with the equivalent larger symbol. Conversion to subtractive was done when writing the result. This seems to make sense, though I think anybody doing math for a short time would have made a lot of shortcuts, such as II×V going directly to X rather than to VV. In any case it can be described with a *SHORT* paragraph, not something like that article! I just now tried to describe it here.

The reference also gets into a ridiculous mess again with "rules". If you want "rules" they are trivial:

1. Number is sum of symbols. As few symbols as possible are used (ie use the largest possible symbol first)

2. Write the symbols sorted from largest to smallest

3. There are *ONLY SIX* substitutions: IIII->IV, VIIII->IX, XXXX->XL, ... Do not try to make "rules" that restrict some larger set of substitutions to these six, especially when it seems to take far more text than just listing the six.

This article is still a bloated mess and we need to make it SMALLER, not larger. Please.Spitzak (talk) 19:07, 24 November 2021 (UTC)

As the article pointed to by Xbig points out, the Romans didn't do arithmetic on paper, they used a counting board, a sort of abacus. The Roman abacus was a 4+1 type and as such setting up numbers, performing arithmetic and writing the results down was trivial. Why I'd bet young Septimus could do it age 8. Martin of Sheffield (talk) 23:13, 24 November 2021 (UTC)

I'm not saying add a large section on this, it can be a small, concise one. It's true that the Romans mainly relied on the abacus, and that would have to be taken into account. I think the issue with this article is not so much length, as a somewhat muddled structure and flow. as for the rules, I've said my piece and discussed that at length; if I couldn't resolve philosophical differences then, I doubt I could now. If you have specific questions about rules-based analysis, I'll answer, otherwise I'll avoid diving into that rabbit-hole. Xcalibur (talk) 12:12, 25 November 2021 (UTC)

I forgot to ask, is the assertion that I'll be arrested for editing Wikipedia articles carefully documented in the references? Keep in mind, consensus is against the false suggestion that Wikipedia was invented by the ancient Romans, so please use another site to post opinions on rules and life (because this is a serious encyclopedia, and not an absurdly unprofessional MMORPG). btw, if I really wanted to be disruptive, I'd go into ILLUSTRATE mode by trying to remove the table of numerals in the article and trotting out the same arguments (it largely repeats information, it'll trick ppl into thinking variant forms are somehow illegal, etc), but I'm not petty enough to do something like that. What can I say, I gave this my best effort.

Anyway, arithmetic with roman numerals remains a possibility which could be elaborated in the article. here are more sources: [6][7][8] Xcalibur (talk) 16:27, 4 December 2021 (UTC)

Though directly transcribing to decimal is certainly how any modern person would deal with Roman Numerals, this can be described last. I think the huge objection to the "rules" is that you spend a ridiculous amount of time trying to restrict the "subtractive" patterns down to the six that can actually be used. For obvious reasons this should not require more than six rules. From your arithmetic refs it is clear that subtractive can be considered a transcription shortcut, not used when working with the numbers. This means roman numerals are just a sum of the value of the digits, that the smallest number of digits must be used (ie the arithemetic has to convert excess symbols to larger ones), and they must be sorted from largest to smallest (required so there is a unique version of each number and to make math easier). That is 3 rules (not 10 or whatever). Subtractive can be described by a rule that *only* allows the patterns, start with "if a digit is repeated 4 times then a sequence can be replaced..." and list the 6 replacements, and thus eliminate all the need to make a mess of "rules" to try to restrict subtractive notation. I agree the current article is still a mess and can be a lot shorter. Also just as before, if your "rules" are better, YOU MUST REPLACE THE OLD TEXT!!! Describing it twice is a waste of both yours and the reader's time.Spitzak (talk) 17:27, 4 December 2021 (UTC)

pardon me, I think it's out of my system now. my main complaint wasn't the disagreement, so much as the passive-aggressive personal attacks leveled against me for pursuing this previously.

As I said, I can answer specific stuff on my proposed section (which I dropped in September 2020) if you're interested. transcription between RNs/hindu-arabic is convenient for us, but it's not based on the structure of RNs. Roman Numerals are not a cipher, in which you replace our numbers with the RN equivalent. rather, it's based on addition/subtraction of symbols (primarily addition), which is a different, more rudimentary means of expressing values. Conversely, our hindu-arabic numbers are built on altering the power applied to the base number, which is a good deal more sophisticated than what is essentially an expanded tally system (RNs). with Roman Numeral tallying, various combinations are possible, and in fact, we see many variants, especially in ancient history (one of my sources mentioned IXCS for 89.5!) But in modern times (especially since 1800), certain combinations became preferred, which ended up becoming a convention, as shown in the table. My whole idea for the rules proposal was to logically analyze the modern convention based on the sources. I continually edited, renovated, and polished the content into what I believe is its final form of 7 rules, with some help from my fellow editors here (who were in fact useful, in spite of everything). As for the insistence of replacement, I believe that a basic description and analysis side-by-side could be complementary, and in fact this approach is in my sources (LURNC/Seitz, Shaw, et al). But it's safe to say that this did not go over very well, to say the least, and there's nothing more I can do on that account.

My current proposal is to develop a section on arithmetic with Roman Numerals. this could 1. describe the computation, which includes addition/subtraction and multiplication/division (detailed in the JSTOR source) 2. give a brief historical background on how the Romans pursued math in a pragmatic manner, as opposed to the theoretical study of the Greeks, and 3. describe briefly how they typically used an abacus for operations, with a link to the main article at Roman abacus. Xcalibur (talk) 19:09, 4 December 2021 (UTC)

I've had a quick look at your sources. The Mastin source really doesn't add anything beyond calling Roman numerals a "clumsy and inefficient system" and dropping an addition in from nowhere. Goodmath characterises them as "so damned weird, hard to read, hard to work with" and then gives us his modern way of doing things. Kennedy is altogether more scholarly but again states "Arithmetic was commonly done on the abacus or counting table. What methods they used for written calculations, if any, have apparently been lost, and the methods described here may be a modern invention". A good source for investigating roman arithmetic is Stephenson who, whilst also reinventing things, does at least start from artifacts.

those sources are not comprehensive by any means, just a starting point thrown together from a search. Mastin is brief, but it supports the claim that Romans focused on practical math rather than theoretical study, mentions the Roman preference for using the abacus, and the example illustrates the need to convert between subtractive and entirely additive forms. Goodmath's comments aside, it delves into calculation methods. Kennedy is easily the best source I've found so far, and the admission that the methods may be modern, and ancient methods lost, is not too much of a setback (many things from the ancient world are now lost, btw). We're not strictly limited to the use of Roman Numerals in the ancient world, although that can and should play a prominent role. thanks for the Stephenson source, it looks like useful material. Xcalibur (talk) 22:09, 4 December 2021 (UTC)

Just so long as it is clear that these are modern methods developed by people trained in Arabic numeral based arithmetic who despise the Roman numbers. Mastin and Goodmath don't mention it, but the methods are not based on any extand Roman sources and are at best, synthetic. Martin of Sheffield (talk) 22:59, 4 December 2021 (UTC)

Keep in mind, that was from a brief survey of sources. There may be more out there, which would allow for more development (especially from authors who aren't hostile to the number system). However, I'm disinclined to pursue this further, at least for now. I mainly wanted to indicate that there's another open avenue for development. BTW, about my criticism of Wikipedia -- there's no denying this is a highly successful crowd-sourced project with some good ideas. However, it's also plagued with issues of gatekeeping, gaming the system and bias, and that's not about anyone in particular, I've seen it site-wide. Xcalibur (talk) 01:43, 11 May 2022 (UTC)

I would absolutely support an arithmetic section. Even if it is likely an anachronism, those caveats are included in the sources and would be mentioned as well. It's also the kind of weird trick that makes strange appearances -- I was first exposed to some version of formalized Roman-numeral arithmetic (multiplication) when I read Shimony's Tibaldo and the Hole in the Calendar (1997) as a kid. I'll have to cross-check what source's system the book might have been using, though. SamuelRiv (talk) 21:19, 23 July 2022 (UTC)

I appreciate your input. I think it's a viable concept for a section, and would do it myself, but I'm holding off due to history here. Xcalibur (talk) 07:38, 30 July 2022 (UTC)

References

Subsections

We ran into some muddles over where a reference to a particular topic should be directed - usually but not always where it is first mentioned - think I have got it right now? --Soundofmusicals (talk) 04:02, 28 August 2022 (UTC)

Internal links

@NebY: I don't understand your reasoning at all. How does making something implied explicit create extra work? It does the exact opposite; making something implied explicit removes the work of having to deduce the implication. Additionally, your revert restores the problematic internal link attached to "subtractive forms in standard use", since it links to a section on "variant forms" not standard use; see WP:EASTEREGG. – Scyrme (talk) 17:04, 28 August 2022 (UTC)

Have you never seen a class bewildered by a teacher explaining the same thing in different ways, when they got it the first time? Far from clarifying anything, it raises questions. Indeed, this has been an ongoing problem at this article, in that different editors have found different ways to describe the same thing, imagining the reader to be confused otherwise; instead, they give the reader the work of getting through another version of the same thing before they can move on. In this specific instance, These are the only forms in standard use is a clean and complete statement that allows us to move on. It quite obviously implies that other forms were or are in use, but this is the section on standard use and we're not going into them here. If we just let the reader carry on reading, they'll find out about them very soon.

Yes, we don't need an anchor for "Irregular subtractive notation" and we don't need a forward link to it in the article. You only placed the anchor for "subtractive notation" because Wikipedia has a redirect Subtractive notation pointing to Roman numerals#Subtractive notation. Wikipedia doesn't have a redirect Irregular subtractive notation - good thing too. I'll clean it up. NebY (talk) 18:31, 28 August 2022 (UTC)

I retained the internal link because Soundofmusicals (talk · contribs) recently added it, and I prefer to avoid reverting another editor when there's an obvious compromise. I don't necessarily object to removing it, but they might. What I do object to is linking the text subtractive forms in standard use to the section on variant forms, but if the link is removed that solves that problem.

You also reverted the addition of "irregular" to a heading because you felt "irregular" implies very infrequent use. I don't agree, but there's no point arguing if we can agree on a different heading that avoids the issue. Would you object to retitling the subheading "Additive notation" under "Variant forms" to "Variant additive forms"? Perhaps also retitling "Irregular subtractive notation" to "Variant subtractive forms" for consistency? Simply having "additive notation" is misleading, as a reader with no prior knowledge may wrongly conclude that "additive notation" is only used in reference to "variant forms" and not Roman numerals in general since the article doesn't mention the term anywhere else. Additionally, consistent headings make the (intended) structure of the article more predictable, and therefore clearer and easier to navigate for both readers and editors. – Scyrme (talk) 19:08, 28 August 2022 (UTC)

Thank you. Soundofmusicals, I hope you'll tolerate my action as in the spirit of collaboration and problem-solving that I hope we've both enjoyed; I certainly have.

As to section headings, this is an old song that SoM and I have often sung: Roman practice varied immensely and consistency was not highly prized, and medieval practice was inventive but fragmented. So I'd use slightly plainer and less technical English: "Other forms", with subheadings "Other additive forms" and "Other subtractive forms". NebY (talk) 20:07, 28 August 2022 (UTC)

Sounds good to me! – Scyrme (talk) 20:34, 28 August 2022 (UTC)

The important things is that we avoid a judgemental approach. There is no such thing as "good" or "bad" Roman numeral, morals don't come into it, at least in the context of an encyclopedia article. At the same time it makes sense to describe a common or usual form that will convey precise and unambiguous information as to which number we are talking about (if only to avoid problems with copyright law!) This needs to be explicit and complete BEFORE we get to the "other" forms, which will mean that things like "subtractive" notation will get split - "usual" subtractive notation has only six possible forms whereas "other" subtractive notation, well... enough said. Some readers, I fear, may well find this confusing? So long as we recognise that we are not quite there yet. --Soundofmusicals (talk) 21:38, 28 August 2022 (UTC)

Yes, if anything I'd like it to be clearer that our "Standard" form is comparatively modern. (Did it only fully emerge when people stopped trying to use Roman numerals for any "serious work" like accounting or even arithmetic?) Even adding "in modern times" to the first sentence of Standard form "The following table displays how Roman numerals are usually written" might be good. Still, the paragraph above that is strong, and I should resist the desire to tinker at low level without properly considering someone reading the whole article for the first time.

You're right, it would be worse to introduce all subtractive forms, common and less common, in one block before describing the rest of the modern standard system. We can be fairly sure we'd confuse more readers that way. For better or worse, "standard" is what we learn in school and even what Excel offers, and it does offer a straightforward starting point.

Yes, not quite there yet and besides, this is Wikipedia. It seems all our articles go through cycles of accretion, even encrustation! NebY (talk) 19:49, 30 August 2022 (UTC)

Regarding the recent edit, it was just an experiment, wording it differently, and in a different part of the article. Admittedly my earlier attempts to insert a single mention were rather clumsy & shoehorned, I thought it fit better there. Xcalibur (talk) 14:42, 30 September 2022 (UTC)

Again, I understand consensus was against me adding "rules" content (or however you describe the instructions given in RS). I just wanted to experiment with a brief mention in a somewhat different way than before, but I guess not much has changed (and wtf is a "motherhood statement", not sure where you were going with that). Xcalibur (talk) 01:06, 4 October 2022 (UTC)

"Clockmaker's IIII"?

There is a limit to the number of times we actually need to repeat the information that RN clocks sometimes use an "additive" IIII rather than the more usual "subtractive" IV! Anyway 10 of the 12 digits of a clock dial contain a "I", not 4. And such a dial is not a quadrant (a quarter of a circle) but a compltee one! The idea that "IIII" is "more aesthetically pleasing" than "IV" in this or any other context is at best very subjective / speculative. All-in-all that edit (even if it had been cited) wouldn't have added anything useful to the article. --Soundofmusicals (talk) 23:34, 1 November 2022 (UTC)

Yes, if it had been phrased correctly, it would have made more sense. It should have said "4 numbers with a V, 4 with an X, and 4 with neither V or X". (the group without V or X is called the 'I' group, hence the poor phrasing of the original editor. Another reason is that the number of characters on each side (right or left) of the watch face is more balanced with a IIII. I found this blog that explains the various theories behind the IIII on some classic watches, here [[9]]. It's in French, but you can translate, if you need to. Dhrm77 (talk) 17:16, 2 November 2022 (UTC)

Good edit. One of the most important things an editor can do is decide what to leave out. Clockfaces are a curious diversion anyway of which all that's known is that all explanations are speculations after the fact, and broadly speaking the English-language Wikipedia does not include technical terms in other languages unless they are commonly used in English-language texts. NebY (talk) 18:04, 2 November 2022 (UTC)

There are many speculative "reasons" given by different sources for the "additive IIII" on (many) clock faces. A good many sources (not just the French blog) go so far as to list ALL of these. Since they are contradictory, and since none are particularly convincing anyway, we have a very long standing consensus that it is much better to ignore them all. If we actually need an explanation (?) the only one that makes any sense is the one (obliquely) mentioned in the lede to the article, good old "tradition".--Soundofmusicals (talk) 04:06, 3 November 2022 (UTC)

Right, since these are just speculations, just giving "tradition" as the reason might be the best approach. Dhrm77 (talk) 14:46, 3 November 2022 (UTC)

As stated above, it's to divide the clockface into even thirds, each representing I, V, and X. As also stated, there's a bit of ambiguity since most of the numerals contain I, so more specifically it's "only I", "contains V" and "contains X", each being 4 hours or 1/3. But unless a source clearly states this, or some other explanation, it should be left out in favor of just mentioning tradition. Xcalibur (talk) 19:32, 12 November 2022 (UTC)

Reason for making this article protected?

Why? John G Hasler (talk) 15:55, 22 November 2022 (UTC)

Persistent vandalism - see the protection log. NebY (talk) 16:14, 22 November 2022 (UTC)

IVPITER

"Isaac Asimov once mentioned an "interesting theory" that Romans avoided using IV because it was the initial letters of IVPITER, the Latin spelling of Jupiter, and might have seemed impious" - that is written in the article. Maybe Isaac Asimov as a Jew translated Talmudic way of writing the number 15 like 9+6 (טו) because Hebrew 15 (יה) is the Lord's name.Evrey9 (talk) 06:01, 18 December 2022 (UTC)

Sadly, no mention of this idea in the Asimov source--Soundofmusicals (talk) 10:04, 1 January 2023 (UTC)

Function "current year"

This is a handy little function that returns the numerical value of the current (Gregorian) year from the built-in date string. Unfortunately, it takes up to 24 hours to increment, depending on your location (rotation of the planet and all that sort of thing). There is no need to anticipate this change - do NOT, as a helpful editor did earlier today try to bring the year "up-to-date" manually. When all is said and done - it's not going to be 2023 for any longer than it was 2022, even in Roman Numerals. Soundofmusicals (talk) 10:00, 1 January 2023 (UTC)

It's not that. All pages at Wikipedia are cached. That is, the wikitext we write is converted to HTML once and the result is saved in a temporary store on a caching server. Someone reading the page gets the HTML from the cache, not from the current wikitext. Many templates use wikitext to do things like calculate the age of a person. The displayed age, like the current year shown in this article, will be wrong on the day of the person's birthday. The incorrect age might be displayed for a week or more. The page should be purged to fix the outdated display. The easiest way to do that is to edit the page and do nothing at all (no edit summary) except click Publish. Johnuniq (talk) 03:59, 2 January 2023 (UTC)

Some editors will also find "Purge cache" on the drop-down menu under Page, depending on their Preferences > Appearnce > Skin setting. NebY (talk) 09:50, 2 January 2023 (UTC)

All very interesting, and no doubt you're right - but the fact remains that it's actually still 2022 somewhere on the planet for 24 hours after it turns 2023 at the international date line - the new year is not instantaneous, but sort of slides round the globe according to degrees of longitude. So regardless of caching and purging the (correct) current year will not be the same at any given moment all over the earth, will it? --Soundofmusicals (talk) 17:28, 2 January 2023 (UTC)

The current version will use UTC, so the year will change the first time the page is purged after midnight UTC (which may be some time later in practice).

It may be theoretically possible to make that more intelligent, but I'm not sure how, and I'm not sure we'd gain much given that (a) it only happens once a year and (b) the actual output is far more dependent on the purge time than on the time zone it's using in a strict sense. Kahastok talk 17:50, 2 January 2023 (UTC)

"No need to represent a number greater than the current year" is not accurate

There is a paragraph in the "large numbers" section about the vinculum which says that the vinculum is part of the standard, but because there is no need to represent such numbers -- "this is hypothetical"

This is total nonsense. These things are part of the standard because very large numbers have often needed to be represented during the Medieval and even early modern period using this system. The standard is based off of classical and medieval usage...

This phrase suggesting that this system is only used for the year is strewn throughout the article, and generally only serves to be misleading. I would suggest that we remove this in all cases... Very large, roman numeralized, numbers have always been a necessity. Arabic numbers are infrequently mixed with Latin text, and so a great deal of medieval literature (which deals with mathematics!) uses these numerals... For example, Boethius' De Institutione Arithmetica. This was written before arabic numerals existed, and, while modern editions substitute arabic for roman, the original manuscripts was produced with things like the vinculum. There are very large (yes, even larger than the year!!) numbers on practically every page....)

I really think this should be removed everywhere it is written, and am tempted to do so immediately -- but wanted to get feedback first. 98.42.80.141 (talk) 19:03, 10 January 2023 (UTC)

To see an example of a MODERN usage which is important and standard, see the 1862 edition of De Institutione Arithmetica (just using this as an example!)

https://openlibrary.org/works/OL1393125W/Anicii_Manlii_Torquati_Severini_Boetii_De_institutione_arithmetica_libri_duo?edition=key%3A/books/OL20554582M

see I, 20 (page 42, or 56 on the pdf)

I'm going to remove some of the more egregious suggestions that this is "hypothetical" 98.42.80.141 (talk) 20:01, 10 January 2023 (UTC)

This article is protected... so someone else will have to remove this stuff. It is really bad, and should be removed urgently. Arxandr (talk) 20:04, 10 January 2023 (UTC)

No, that is not a case of MODERN usage. As already stated in the article large numbers (incidentally mostly rounded) WERE expressed in RN (in various ways hardly constituting any kind of standard) in ancient and medieval times. Modern reproduction of ancient or medieval text that includes representation of "large" Roman Numerals is obviously another matter - since this is done NOT by applying a (hypothetical) "modern" standard but by literally transcribing the VERY non-standard medieval text.

Since you are not at present able to directly edit this protected article - what about regaling us with your suggested version of this section - on this talk page ? Soundofmusicals (talk) 00:43, 11 January 2023 (UTC)

Note: I have already changed "purely hypothetical" to "largely hypothetical" Soundofmusicals (talk) 01:22, 11 January 2023 (UTC)

The link I've provided absolutely is a modern usage. This edition of Boethius is not a recreation of Boethius' formatting, but an edition of the text and graphs... Boethius himself did not use this specific system (it didn't exist yet). As such, the Vinculum is being used as a method of providing information, but is not a direct reproduction... it is the collation of multiple texts, and the vinculum has been used for convenience.

I did some research, in the oldest manuscripts, you will often have the number simply written out if it is too large. But in the later texts, they always use the vinculum, or a similar system.

In this critical edition, the vinculum is a convenient feature, and it is standard for all all critical editions of any text which utilized RN. Whether published by Teubner, Oxford, or Brill -- they use the vinculum...

Knowing this standard, whether one is an editor or a reader of these editions (or anything similar!), is important. Especially since, as the article states, many (or most) scholars agree that this is a modern standard, which has its routes in the late medieval (reminder: Boethius was early medieval).

I'll provide another example, this time of a text which is still in print, and quite a hot seller by comparison. The Loeb edition of Pliny's natural history utilizes the same system... It is obviously standard in the typesetting of any edition of text which used RN. In fact, the Loeb even uses the brackets. https://archive.org/details/L352PlinyNaturalHistoryII37/page/n429/mode/2up (the RN are in the Latin [left] page. -- note that while the number may look like "1332" it is in fact 1332 x 1000, since "miles" just means "thousand" in latin, ie. 1000 paces).

RN are used for all sorts of things other than the year. This article has the assumption throughout it that the only reason to use RN is the year...

I'd be happy to provide some example changes, but I don't think what I'm asking for is too dramatic, and I'm happy to allow anyone else's best judgement here... It just seems presumptuous (and wrong) that there is no modern reason to know the RN brackets or vinculum. These editions were produced either in the 19th or 20th c. There are new editions of Pliny in the works as we speak... The text is not a replication of any particular manuscript, but the collation of all available manuscripts. Arxandr (talk) 02:14, 11 January 2023 (UTC)

"RN are used for all sorts of things other than the year. This article has the assumption throughout it that the only reason to use RN is the year."

Oh dear! - what about all the modern uses listed (with examples) in the article (under Modern Uses", most of which never (or hardly ever) go over 50 or so - and so do not require "vinculum, or a similar system".

Year numbers up to 3999 can be written in "usual" RNs, again without any need for vinculum, "or a similar system". This will be adequate for another 1976 years. I rest my case.

Modern editions of medieval of ancient texts also do not require a "modern" system of "vinculum, or a similar system", but will properly consist of transcription of the numerals used in the original. All the examples I have seen seem to do just that. (Although fanciful "collation" is perfectly possible, if the editor is so moved). On the other hand, the idea that this represents a modern (much less a standard) use of RNs is ludicrous, if you will forgive me being a bit blunt.

Since the stress of this article is descriptive rather than prescriptive (that is, we are describing RNs as they are actually used (with a bow or two to how they have been used at various times and places) rather than setting up prescriptive rules and conditions for "good" RNs, or decrying "bad" ones...

None the less I have made a few changes to the emphasis - have you read these, and what do you think of them? Soundofmusicals (talk) 08:04, 11 January 2023 (UTC)

I don't have a printed copy of Pliny NH VI 26.100 to hand and online versions such as this or this reproduce the vertical lines easily enough but can't show a horizontal bar, necessary as it is to yield1332 thousand paces. So while it would be good to use eg an introduction to a modern scholarly edition as a WP:RS, perhaps we could still say something along these lines:

"Some modern sources describe the vinculum as part of the current "standard" and it is used in some modern editions of classical and medieval Latin texts, as are the parenthetical vertical lines, but is unnecessary for the smaller values for which Roman numerals are commonly used nowadays, such as years." NebY (talk) 13:27, 11 January 2023 (UTC)

Neby, in my comment above, I provided a link which has a scan of one of these pliny books, wherein the vinculum is used. I'll provide the link once more https://archive.org/details/L352PlinyNaturalHistoryII37/page/n429/mode/2up

I think your proposal works well. I'd really like to get the word "hypothetical" out of there... because its not hypothetical. It is a standard used for weird cases such as editing latin texts. Arxandr (talk) 19:06, 11 January 2023 (UTC)

Yes, I should have said "any other printed copy". Your link was claer clear and helpful, but editing Wikipedia has made due-diligence checking rather automatic. Then again, a Loeb edition is a good enough demonstration by itself!

Thanks. My wording may not be quite right yet but yes, lI'd like to replace that editorial judgment by observing where it is used and indicating why it's not seen commonly. NebY (talk) 19:40, 11 January 2023 (UTC)

In what way is it "fanciful" to believe that a stemmatic critical edition of a text is a collation of all available manuscripts? This is an easy thing to prove.. there is no single manuscript which agrees in every way with the CE.

I'll also remind you that the modern period extends from 1500 to present. There are quite literally thousands of editions from the 1500s-1700s which very deliberately prefer RN to arabic in Latin texts.

But luckily, the two editions I have provided links for are late modern. Printed in the 19th or 20th c. Both are still in print.

I checked out the changes. I do not feel that "hypothetical" is the right word. As the article states, the vinculum was used as early as the Roman period, but then was "standardized" in the medieval. This is evident as not all mss. which predate the 12th c. use the vinculum for large numbers, but afterwards, you'd be hard pressed to find any alternative.

These standards exist for such edge cases. The conjecture that it is hypothetical dismisses fields like textual criticism wherein we do find editors making deliberate decisions to use the vinculum to represent large numbers -- not as reproduction but as an act of editorship. Arxandr (talk) 19:05, 11 January 2023 (UTC)

Further modification to article - largely based on the concerns of Arxandr and the positive suggestions of NebY. Please read the full article as it stands rather than interpreting statements out of context. --Soundofmusicals (talk) 21:35, 11 January 2023 (UTC)

Thanks. I'd like to take a little time to improve my hasty phrasing and integrate it better - I had an idea that it might replace part of the following text and there is some duplication now - but I shouldn't spend time on it in the immediate future. NebY (talk) 22:33, 12 January 2023 (UTC)

Thank you -- and I apologize for being so animated about this, but I did feel that that there was a lot that could be improved wrt the original phrasing.

Do we think that it might be useful to add a photograph/screenshot of the vinculum in a modern text? I think it is fair to say that there is no common need for this, but it might be helpful to provide a material example of an uncommon need, such as discussed here. Arxandr (talk) 17:57, 13 January 2023 (UTC)

Thinking to merge two sentences, I wound up reshuffling and copy-editing the whole subsection. We now open with examples, so may not need a screenshot or photo. I do worry a little about whether there should be a small space between the vinculum numerals and the plain ones, eg IV DCXXVII. Kennedy doesn't include vincula, alas.

I still worry about only citing Bongo for the parenthetical vertical lines; as 16th-century he's WP:PRIMARY and he doesn't seem entirely consistent eg |X| for 1,000[10] ie x100, but |M| for 10,000, ie x10. NebY (talk) 20:05, 16 January 2023 (UTC)

Medieval examples

We're probably better off not adding examples of "creative" Medieval usage? Soundofmusicals (talk) 02:02, 18 January 2023 (UTC)

Which examples are you thinking of? NebY (talk) 12:17, 18 January 2023 (UTC)

Perhaps what I am really getting at was examples that good old "average readers" can't copy/use themselves. Soundofmusicals (talk) 19:00, 18 January 2023 (UTC)

Let's hope our average readers never try using RNs for fractions or large numbers! (Though of course some happy nerds will have a go.) But of course we have to try to cover them for the article, and that does lead us to surprising places. I've been thinking of reading some of Pliny's NH, but I hadn't realised I'd want vincula and more for that. NebY (talk) 20:11, 18 January 2023 (UTC)

OK - better explained now - I still don't really like it but... Overall - excellent work. Soundofmusicals (talk) 19:10, 18 January 2023 (UTC)

Oh good - thank you! NebY (talk) 20:11, 18 January 2023 (UTC)

roman numeral

<copyright violation and impersonation removed> NebY (talk) 17:26, 27 January 2023 (UTC)

DO NOT (yes I'm shouting) ADD MY SIGNATURE TO YOUR EDITS. Martin of Sheffield (talk) 16:06, 27 January 2023 (UTC)

The first part of that post was copied and pasted from another web page such as this where identical text appears. The second part copied another editor's post including their signature, impersonating them. Deleted. NebY (talk) 17:26, 27 January 2023 (UTC)

Large numbers (again)

On a slightly different tack the article states that: "The largest number that can be represented in this notation is 3,999 (MMMCMXCIX)" but this is only true if you restrict the thousands to subtractive notation. It would be quite reasonable to go up to 4,999 (MMMMCMXCIX) using additive. Martin of Sheffield (talk) 12:06, 17 January 2023 (UTC)

True, but the statement is in the first section, Standard form, before the Other additive forms section. Still, would "in this manner" be better than "in this notation"? NebY (talk) 12:18, 17 January 2023 (UTC)

On balance, yes. I'm having difficulties typing, I broke my arm last Friday! Martin of Sheffield (talk) 12:26, 17 January 2023 (UTC)

Oh no! I hope it's a break that will heal straightforwardly. I'll make the change. NebY (talk) 12:47, 17 January 2023 (UTC)

Add my sympathies to Martin of Sheffield - my typing is far from fluent and accurate, even with two sound arms! More to the point - this section is now in a bit of a mess. I have attacked some of the things that worried me most but I am putting it the whole thing into my sandbox in an effort to get it clear and accurate, while taking account of various people's well meant concerns. That clear template does more harm than good - adding acres of "white space" in some formats - suggest we might live with the awkwardness of illustration placement in others. --Soundofmusicals (talk) 19:50, 17 January 2023 (UTC)

Can you say in what way(s) you find it a bit of a mess? The distinction between the vinculum and a modern vinculum seemed to be there only for the assertion that we don't need the vinculum for the smaller values for which Roman numerals are commonly used nowadays, and the withholding of examples to the end in a somewhat "well, if we must" manner seemed to be largely to support that observation. Are the mixed examples IVDCXXVII and XXVCDLIX unhelpful, or did you remove them in case they should be presented with spaces as IV DCXXVII and XXV CDLIX? Yes, I had to check how the Pliny example worked when I first saw it and I'm glad we can present it with a citation to a two-page view with translation. Do you find the current "XIII XXXII p. for 1,332,000 paces (1,332 Roman miles)" clearer? Speaking of {{clear}}, if you find you're seeing too much whitespace then yes, let's leave it out. NebY (talk) 20:30, 17 January 2023 (UTC)

To explain the Pliny example further: the parenthetical lines multiply 13 by 100, giving 1,300. The vinculum multiplies that by 1,000, giving 1,300,000. Then another vinculum multiples 32 by 1,000, giving 32,000, so together we have 1,332,000. The p is passus, paces. The Roman mile was of course a thousand paces (mille passus or milia passuum), so 1,332,000 paces is 1,332 miles. A reader familiar with vincula might well read the vincula and the p together as miles, saving a couple of calculation steps. NebY (talk) 20:41, 17 January 2023 (UTC)

Still strikes me as REALLY weird, especially as an example of the statement. Soundofmusicals (talk) 22:08, 17 January 2023 (UTC)

Vincula together with parenthetical lines do look very weird if you're not used to them. But that's a fine and well-sourced example of their use nonetheless. NebY (talk) 22:15, 17 January 2023 (UTC)

But is it a useful example of typical use that a reader will be able to follow? Soundofmusicals (talk) 22:21, 17 January 2023 (UTC)

At this level, Roman numerals are going to be challenging, but we can work together on making it clear. I've offered two approaches, and we could go down the (13 x 100 x 1000) + (32 x 1000) route. Which do you find, if not easiest, least hard? NebY (talk) 22:33, 17 January 2023 (UTC)

@Soundofmusicals: the Antonine Wall examples are of the vinculum (p(er) p(edum) IIII ( milia) CĐXI)[11] and of indicating numerals (leg(ionis) II Aug(ustae) et leg(ionis) XX V(aleriae) V(ictricis) f(ecerunt))[12]. Re,"Pushing them together", in a section about the vinculum, it is appropriate to say that it is not a numeric indicator and to describe both those methods of identifying numerals together - after all, this section is not about either of them. Of course, they might deserve their own section. :) NebY (talk) 22:07, 17 January 2023 (UTC)

This section actually has several digressions that might be separated - I am not sure this would be useful though. I just don't like the close bracketing of the ancient use of overlining with modern "over and under" al the same. Perhaps a return the original text before might be simpler? I am trying very hard to keep anything that might be interpreted as a "reversion" or get anyone's back up while maintaining the integrity of the article.

I intended to avoid multiple edits by using my sandbox but this is not really proved practicable in this case. Soundofmusicals (talk) 22:19, 17 January 2023 (UTC)

We can drop the joined-up serifs. They're not at all visually similar to the vinculum or the overlining indicator. That would save a digression and let us lose an accompanying image too! NebY (talk) 22:28, 17 January 2023 (UTC)

Fine by me - never liked that inage much anyway Soundofmusicals (talk) 22:31, 17 January 2023 (UTC)

Oh good! That could give us simply "The vinculum is distinct from the custom of adding an overline to a numeral simply to make it clear that it is a number; both are seen on inscriptions from the Antonine Wall."

or "simply to indicate that it is", or "to highlight that" or "to emphasise that" or.... NebY (talk) 22:42, 17 January 2023 (UTC)

By all means - any of those - although I'm not sufficiently convinced one of them would be an improvement to make the change myself I wouldn't object to you doing it. Soundofmusicals (talk) 22:50, 17 January 2023 (UTC)

Ah. By joined-up serifs, I meant also over-and-underlining. Do we have a source or an example for distinct lines above and below? NebY (talk) 22:48, 17 January 2023 (UTC)

By all means add them back in if you like. I am too tired to think clearly Soundofmusicals (talk) 22:51, 17 January 2023 (UTC)

Don't think we need a source for something so obvious. TTFN Soundofmusicals (talk) 22:52, 17 January 2023 (UTC)

OK, hope I haven't misunderstood you with this edit. TTFN! NebY (talk) 22:59, 17 January 2023 (UTC)

@Soundofmusicals: Let's try to make sure we're talking about the same things.

1. Vinculum - a line above the numerals, at a small distance from them, that multiplies them by a thousand

2. "overlining" - a line above the numerals, at a small distance from them, that indicates that they are numerals (especially useful in those Roman inscriptions that use many abbreviations)

3. "over- and under-lining" - lines through the top and bottom of the numerals, where serifs would be, which I've called "joined-up serifs".

Is that right? Regarding #3, do you know of any cases where the lines are actually at a small distance from the numerals? I ask because while it might seem to require meticulous phrasing to describe them differently, the appearance of #3 is so different from that of #1 and #2 that they would not be confused, and it's more of a typographer's design choice.

If #3 does not appear with lines at a distance from the numerals, we don't need a gloss in Roman numerals#vinculum to say that #1 and #3 must be distinguished. This simplifies the section significantly. We do want to distinguish #1 and #2 especially as they are both seen in inscriptions from the same period and general location. NebY (talk) 12:43, 18 January 2023 (UTC)

You'll see how following this approach, together with (a) dropping the non-WP:RS website and the argument with it about "standard" and (b) replacing the examples created by following one of Bongo's contradictory examples with an example from a modern edition (supported by referencing that parallel-text edition with a translation on the facing page), gives us a briefer section that stays focused and proceeds straightforwardly. NebY (talk) 14:13, 18 January 2023 (UTC)

Impressed. Bongo is actually the source of the "illustration" at the head of the large numbers section - it seems to have been more a summary of different (contradictory) usages than a "manual" - although "old" it is technically "early modern" and antedates the introduction of "Arabic" numerals by centuries. cf the reluctance of the United States to adopt the metric system. But then of course you knew all that already. Soundofmusicals (talk) 19:20, 18 January 2023 (UTC)

I still don't know what to make of Bongo's |M| = 10000, seen in that illustration. Could it have been a mistake? For three pages before those examples of different usages, he provides a five-part description of the workings of RNs, but the fifth part's defeated me so far.

Digressing, did you know that some in the US speak of "metrication" and some of "metrification"? We had a whole discussion about renaming Metrication in the United States .... NebY (talk) 20:40, 18 January 2023 (UTC)

References

Vertical lines

The indication that vertical lines multiplied by 10 or 100 seems to not be supported by the references. The first reference has no vertical lines in it. I cannot read the second reference. The third reference is for the example given. However it shows a mulitplication by 100 (not 10), and more importantly it does not have an example without an overline as well. I think the actual use was to "box" more digits that are prefixed to whatever number is after it, they are then appended as though you appended two decimal numbers. In this case the entire left+top+right lines indicate a multiplication by 100,000 but that factor is chosen only because the digits afterwards are in the range 10,000..99,999. If the later digits were different then the multiplication would be by a different power of 10.Spitzak (talk) 21:50, 18 January 2023 (UTC)

We do have a very uneven batch of references for these. |X| and |M| are from Bongo, the seventh example on the right-hand page here and the last example of 10000 on the right-hand page here - that's also the image on the right in the article. The 100,000 is much more clearly supported both by our ref, which says "the multiplicative (vinculum) bar for 1000 used in classical antiquity continued to be used under a new name, the titulus, but the three-sided box symbol for multiplying by 100,000 was no longer used (Menninger 1969:281)"(Stephen Chrisomalis, Numerical Notation: A Comparative History), which he expands on page 403 with an argument you may not find persuasive ("When expressions for 100,000 were no longer needed in the early Middle Ages (because of reduced social complexity in Western Europe), they disappeared")[13] and by the Loeb example from Pliny's Natural History. If you flick through the following pages of Pliny, you'll find more including XXV = 2,500,000 (2,500 Roman miles) and XV = 1,500,000 (1,500 miles), neither with any following numerals.[14] NebY (talk) 23:30, 18 January 2023 (UTC)

The first examples without an overline look a lot more like I than "lines", ie IXI and IMI. I feel this may be unrelated to the later use where an upside-down U is drawn.Spitzak (talk) 16:43, 19 January 2023 (UTC)

Fair point. I accepted the long-standing claim about |X| and |M| because I felt I was copy-editing drastically enough already and we do see vertical lines used with the vinculum to make the three-sided box (eg the Loeb Pliny). The Bongo illustrations are horribly unclear, but that's partly because they're so crude - maybe rough wood-cuts or etchings, and maybe worn ones at that. But yes, we should have better sources.

The first mention of vertical lines in this article is this 2004 edit, with no reference but we weren't so strict about referencing then. That does have an EL which is still live, Mathworld on Roman Numerals. I'm not sure whether or not that's an RS itself (I think it's basically Eric W. Weisstein without any editorial oversight, and his representation of the apostrophus with brackets is arguably misleading) but maybe that doesn't matter so much as that he shows us a couple more sources we could look at:

Cajori, F. A History of Mathematical Notations, 2 vols. Bound as One, Vol. 1: Notations in Elementary Mathematics. New York: Dover, pp. 30-37, 1993.

Menninger, K. Number Words and Number Symbols: A Cultural History of Numbers. New York: Dover, pp. 44-45 and 281, 1992.

I haven't yet tried to find online views of those pages - I must deal with some RL first!

It is tempting to simply abandon mention of |X| and |M| until and unless we can find better support for them. I'd like to retain the three-sided box as it's seen with the vinculum and mentioned together with it by Chrisomalis, and not otherwise covered here, but we don't need |X| and |M| in this section. NebY (talk) 21:39, 19 January 2023 (UTC)

I've removed |X| and |M|, at least for now, keeping the three-sided box which Chrisomalis refers to together with the vinculum (see esp 402–403) and which does appear in combination with the vinculum in texts. The subsection seems a bit more readable now. NebY (talk) 10:38, 20 January 2023 (UTC)

To my understanding, vertical lines were not used separately, but were always part of a three-line "box" used in medieval notation to multiply by a large amount, eg 10k or 100k. The vinculum was always a 1k multiplier. In my travails, I saw some sources repeat the vinculum to add additional powers, although I'd have to dig them up. Since Roman Numerals have a pattern of repeating up to 3x, I'd assume up to 3 vinculums would be allowed, bringing the possible range to 1 less than 4 trillion, although this would certainly be an extended/atypical usage in any case. Also, stating what patterns are allowed, aka rules, is not exactly popular around here, I've noticed. srsly, all you had to do was agree that it was too technical for the article, and I would've actually agreed to that. I kept it going not only out of stubbornness, but because I didn't see a solid reason for not documenting the sources in that way, especially since the sources themselves "repeat information", which was a common objection. Pointing out that they were overly technical would've defused the whole thing.

Anyway, back on topic. The over & underline was used to signify that the letters are a roman numeral, and I don't know of any variation on that.

The apostrophus is an alternative to the vinculum (and convenient for typing). The article states that only up to 3 apostrophii were used, which agrees with my intuition on a 3x limit for vinculums. The article also mentions that (I) and I) were the origins of M & D respectively (1 x 1000, 1 x 500), to which I agree. Xcalibur (talk) 06:59, 3 February 2023 (UTC)

Year 4000 Problem

"The largest number that can be represented in this manner is 3,999 (MMMCMXCIX), but this is sufficient for the values for which Roman numerals are commonly used today, such as year numbers:" - Is there year 4000 problem. Because largest Roman numeral is MMMCMXCIX or 3999, ignoring vinculum. But vinculum can't be properly displayed on screen. Epochalypse2038 (talk) 12:01, 18 February 2023 (UTC)

It's actually a Y5k problem! I'd suggest waiting until MMMMCMXC before bringing this up again, then you can start the international panic going. ;-) Martin of Sheffield (talk) 12:08, 18 February 2023 (UTC)

MMMM breaks the rule "No more than three repeating letters". Epochalypse2038 (talk) 20:24, 19 February 2023 (UTC)

See Roman_numerals#Other_additive_forms. There are no "rules", only conventions. Martin of Sheffield (talk) 20:43, 19 February 2023 (UTC)

Hear hear!! This is hardly a pressing problem, even assuming that we're still likely to be numbering years in the same sequence over the next two or three millennia, or for that matter still using Roman Numerals for doing so! Vinculum is irrelevant in the context of current usage of RNs anyway, but if it were otherwise it would be simple enough to standardise a method of ensuring its standard representation "on the screen". To paraphrase Martin of Sheffield's edit there are no prescriptive rules "governing" the use of RNs in any case. But PLEASE do not reopen that can of worms! Soundofmusicals (talk) 21:27, 19 February 2023 (UTC)

Lol, this time I'm not the one who brought it up. Anyway OP, you're correct in that vinculums are difficult (although not impossible) to use for computer inputs. Fortunately, there's an alternative, which is to apply parentheses in place of vinculums, so that 4,000 or MMMM would become (IV). As another example, 6,986 would be (VI)CMLXXXVI. I've seen sources that repeat vinculums, presumably up to three times (since it becomes difficult for the eye to distinguish after three repetitions); and in fact, the related apostrophus is stated by the article to have a 3x limit. Thus, the upper limit for RNs would be 4 trillion minus 1, or (((MMMCMXCIX)))((CMXCIX))(CMXCIX)CMXCIX, or 3,999,999,999,999. Alternatively, you could put (((MMMM))) or 4 trillion as the highest value, since that's more aesthetically pleasing. Either way, that gives you plenty of room to work with. Xcalibur (talk) 02:17, 23 February 2023 (UTC)

Quite right! after all the "system" has already been through many changes. But the real point is that even the standard "maximum Roman numeral" doesn't overflow for any use to which we currently put this archaic and long obsolete system to for another couple of millennia. Hopefully that's even enough time for the Americans to "go metric" lol Soundofmusicals (talk) 02:42, 23 February 2023 (UTC)

But why use the irrational decimal system? It would be far better to standardise on base sixteen in the long run. It makes for a far simpler interface with cyber. :-) Martin of Sheffield (talk) 08:25, 23 February 2023 (UTC)

Now THAT'S an idea! 1,976 years is a long time for us to assume that ANY human "civilisation" will survive, much less anything we would recognise. -Soundofmusicals (talk) 10:52, 23 February 2023 (UTC)

(OT) Always remember that there are 10 sorts of individuals – those that understand binary and those that don't. Martin of Sheffield (talk) 15:03, 23 February 2023 (UTC)

Twelve and sixteen are excellent choices for base numbers. As for measurement, I don't think we'll switch over, metric will remain in a secondary role, with imperial/standard as primary, there are many reasons for this which I won't get into. Finally, two millennia is plenty of time to figure out a keyboard shortcut for adding vinculums if one is eventually needed, in the meantime, there's parentheses as I explained above. Xcalibur (talk) 17:57, 23 February 2023 (UTC)

o 103.209.196.225 (talk) 08:47, 25 May 2023 (UTC)

@Soundofmusicals 103.209.196.225 (talk) 08:48, 25 May 2023 (UTC)

Subject heading change

The original heading (Standard form) goes much better with the wording of the previous section - and also with our philosophy of "describing" rather than "mandating" a "usual" form that is accepted and understood (more or less) everywhere. Under the following heading we group all "variant" forms that "differ from the standard described above" - with a minimum of comment - just mentioning that they have "occurred" (been used at some time or period in a document or inscription). This kind of clarity is incompatible with any kind of attempt to tie down the inconsistencies of historical usage to a set of "rules" - which is why we ended up as we did. Very relieved that the attempt to go back to these was so promptly reverted by its proposer! Well done by the way. Soundofmusicals (talk) 02:31, 23 February 2023 (UTC)

In case you were wondering, I was creating a permalink for the rules content, since I've made further adjustments & improvements since the controversy. I had no intention of stirring up any more trouble than I have already. As for the section title (Standard form vs Roman numeric system), I just wanted to try out an alternative, but I can see where your use of "Standard form" fits with the existing content, such as Other forms. Another reason is that I wanted to use that title for my own content, but that's not a conflict, given the consensus. As I said above, I never would've been so stubborn if you had just pointed out that the rules are too technical for this article. Anyway, hopefully my recent edits were not a big deal. Xcalibur (talk) 17:49, 23 February 2023 (UTC)

Spelling error

I'm new so I can't edit but there is a spelling error where the page says 'the Spanish siglo XVIII (not XVIII siglo) mean "18 century"' instead of the Spanish siglo XVIII (not XVIII siglo) meaning "18 century".

If someone could edit this that would be great Taftaloka (talk) 08:33, 17 June 2023 (UTC)

"Mean" is correct (to take a simpler construction "for example, A means B") but yes, a construction with "meaning" would be correct too ("for example, A, meaning B" or "for example, A (meaning B)"). I've switched to "for". NebY (talk) 10:15, 17 June 2023 (UTC)

Thanks. Although it doesn't really matter, I was also wondering about why the article does not use the Unicode Roman Numerals and instead opts for keyboard characters. (for exampleⅩⅧ instead of Latin characters XVIII) Taftaloka (talk) 01:33, 19 June 2023 (UTC)

See Roman numerals#Unicode and Talk:Roman numerals/Archive 10#Using Unicode points for numerals?. NebY (talk) 12:08, 19 June 2023 (UTC)

Supposed crossword puzzle use

• The notation ID for 499 or IM for 999 is used in crossword puzzles.

I removed this bit, because (a) it is not clear what it means. Clues give words, not numbers, so on the face of it, this is backwards; (b) crossword setters play games with language, so unless extremely clear and significant these games do not belong here. Imaginatorium (talk) 05:42, 6 November 2023 (UTC)

I'm guessing some crossword puzzle had the clue "499" and the answer was "ID", and same for IM. However I think there has to be a link to the crossword puzzle as a reference for this information to stay. It is mildly interesting in that, along with the Excel thing, shows how common the "anything subtracts" idea is in modern times. Spitzak (talk) 15:24, 6 November 2023 (UTC)