Talk:Roman numerals

Mathematics B‑class High‑priority

This article is within the scope of WikiProject Mathematics, a collaborative effort to improve the coverage of mathematics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.MathematicsWikipedia:WikiProject MathematicsTemplate:WikiProject Mathematicsmathematics articles B This article has been rated as B-class on Wikipedia's content assessment scale. High This article has been rated as High-priority on the project's priority scale.

Template:FAOL

This article has been mentioned by a media organization: "Library Clock Has 'IIII' Instead of 'IV'". AP Strange News. July 2, 2004.

It's not very common, but you do sometimes see dates written as 4.viii.06 or (as I prefer) 4.VIII.06. A letter I opened this morning had a postmark from Coventry with 1-VIII.

Yes, I believe in entomology (and perhaps other biological sciences) the month is written with Roman numerals. This is in order to avoid confusion--in some areas people put month first, while in others people put day first. Is this worth noting in the "Other Modern Usage" section? Sigil VII 07:02, 15 August 2007 (UTC)[reply]

This whole article is embarassingly uninformed and demonstrates the weakness and unworkability of the whole Wikipedia concept better than any other I've seen. I won't even attempt to fix it. I can now see why so many academics regard Wikipedia as a pathetic joke. Freddy011 (talk) 21:45, 6 December 2008 (UTC)[reply]

Could you give us a clue as to what the problem is? Or is it only you that thinks it ridiculous? --WiseWoman (talk) 19:41, 12 January 2009 (UTC)[reply]

I've read somewhere that Roman numerals were partly responsible for slowing the development of science and math. This was purely because they are harder to deal with, and it takes even a trained user longer to add, subtract, multiply and divide numbers written in Roman numerals that it does someone using Arabic numerals. This greater barrier to entry, as it were, resulted in less research.

Is this true? -- ansible

Maybe yes, maybe not: they certainly could not interact with "our" computing system because of lenght (in chars) of single numbers, which is variable too. For instance: number 77 is expressed in arabic with two chars only, while LXXVII needs six. But the worse is that number 78 needs seven chars (LXXVIII).

Side note. Don't computers use 0s and 1s, anyway? - Rose

But this does not mean that there are operations that you cannot perform with them too. If you try to calculate a square root of some number, you will get the same result with both systems. Maybe obviously this would require a different use of the space.

On the other side, every calculation in roman numbers requires a logical scheme that is different from arabic system. I could not say which is the best: if you are latin-minded (and you are consequently used to decline words, verbs and other object of same frequency making use of a sort of "on-the-fly" developping), you will find it as natural as today we find arabic ones, the longer time only depending on writing.

I think that it is only that arabic system was used by phoenician merchants in the whole Mediterranean area well before that Rome had an influence over a similarly extended territory. A fact is that Rome created the widest empire of ancient world using its numbers, and another fact is that we use arabic system; opinions might evaluate whether it is better for us, but keep in mind that we were born "within" this mentality.

I do think however, that it would be quite complicated to eventually revert our system now :-)

In Latin class I had learned the exact opposite of what Ansible suggested. From what I understand, the Roman numeral system is supposedly really easy to count on your hands with. Essentially, the Roman numerals were quicker to add and subtract with, whilst Arabic numerals are easier to mulitiply with. Just some thoughts... --BlackGriffen

I can kind of see how Roman numerals would be easier to learn how to count. But there's a lot more science needs than simple counting. It's multiplying and division that seem to be overly difficult in Roman numerals. I remember in 3rd grade, learning how to divide using Arabic numbers was hard enough. What are even the rules for doing manual division with Roman numerals? Does anyone even know anymore? -- ansible

As far as I remember, there is no different operational method with roman numbers, since it should be only a matter of graphical rendering, or I didn't meet this point in my studies. The concept of division should be the same in both methods. Roman system has a different approach to rendering, requiring not to consider a linear sequencial scale (as in arabic ones) but a more complex thought about "notable" numeric entities: 99 is in arabic only the number after 98, in roman it is the one before one hundred (closer relevant entity), and is "IC" (really, I am not aware it is wrong, being more purely latin minded, and I have many books with that form too - just checked). A rendering like LXXXXVIIII would be the first result in our current mentality, but a second thought is required to better describe it in latin concepts (so LXXXXVIIII is a wrong form, correct being only IC - or the other one proposed in article).

With arabic numbers you have to learn by heart some concepts like fixed relationships: series of adding or multiplying factors (like 2,4,6.8.... or 3,6,9,12,.... or 4.8.12.16,.... and so on) will be recalled from your memory when you compute the separate parts of a multiplying operation. In roman maths, you will constantly evaluate concrete "weight" of numbers, so you will get your result with less use of memory and deeper instant analysis.

With romans you need to realise "where" in the proportion of values your number is located: is 98 closer to 50 or 100? Am I talking about something that is of (this) kind of proportion or of (this other kind)? Idea is: main identifiable concept = one hundred, my number differs from that of II less (so it's on the left = IIC), while let's say 105 is more than one hundred by 5 (so it's on the right = CV).

Obviously this is easier to perform with sums, and it is true that main progresses in main scientific disciplines were achieved by arabs (just think of astronomy). I wouldn't however agree it simply slowened progress: Roman system might have been better structured to complete a mentality which put humanist sciences before technical sciences, but today's progressed world belongs more to latin civilisation than to arabian one.

I do not believe that the difference between operating with Roman or Arabic numerals is just a question of getting used to one or the other, as some of the above posters seem to imply. A positional number system like the Arabic one(with a symbol for the zero, which the Romans didn't have, and which marks a significant difference) is much more handy for performing all kinds of operations thana system like the Roman. In fact, the Romans did not do any long divisions or stuff like that with their symbols, but they used an abacus for calculations, and an abacus is basically a positioning system. By the way, the Arabic numbers should really be called Indian numbers, that's where they originally came from, although they were introduced into Europe by the Arabs. In reply to the last post, I think the progress in the Western world owes much more to the Arabic digits than to the Roman humanistic legacy....

I would like to quote Georges Ifrah from "The Universal History of Numbers":

Anyone who reflects on the universal history of written number-systems cannot be but struck by the ingeniousness of this system, since the concept of zero, and the positional value attached to each figure in the representation of a number, give it a huge advantage over all other systems thought up by people through the ages. -Calypso

Oh - so the West was able to 'progress' using Arabic zero while the Arabs were not? Just a query from a specialist in the western humanities. Let's not start this silliness. Culture is considerably more complex than ease of computation. --MichaelTinkler

I certainly agree that culture and 'progress' are complex topics. It seems to me, though, that the tools (physical and mental) that are available to people drastically change their outlook on the world. Mathmatics is the basis of science and technology, and arithmetic is the starting point of it all. It seems to me that entire new opportunities became available to us, when we switched numbering systems. However, I'd like to have some references before I write up an encyclopedia article about it. Are there any good studies of history where fundamental practices changed because of better math? Like some example from military history, where someone, because they were able to figure out their logistics better, were able to win some battle. --ansible

Sure it's important, but given that everyone from India to Iceland had the use of the numeral system by some date, we're in Sapir-Whorf fallacy zone to use it as much of an explanation. And what about the Central American zero? There are scholars who insist that the invention of double-entry accounting (Venice, late middle ages) is really what does it for the West. My only goal here is to suppress sweeping, universalistic statements about the zero changing the world. It did so, but very, very, very slowly. --MichaelTinkler

Of course culture is terribly complex, and I certainly did not want to reduce the success of Western culture (or, let's say, the current dominant position of Western culture in economic terms) to the adoption of a certain numerical system. In any case, the main point of my previous post is that the Arabic numerals are intrinsecally better for doing mathematics than the Roman numerals are. --Calypso

Take a look here above: most of the words you use have latin or greek roots. Maybe this is stats, still it's not economy. I agree that the explanation of this concept might be better shown in arabic numbers, but I still prefer a latin "idea", than a fair perfect result.

Greetings

I thought that in the Roman era and the Middle Ages, people used the abacus for calculation and the numerals for writing them down. -- Error 00:34, 5 May 2004 (UTC)[reply]

Indeed. The very word "abacus" is a Latin one, though nowadays abaci are chiefly identified with East Asia. --167.206.188.3 07:38, 10 Oct 2004 (UTC)

I would agree with the original post here. Roman numerals do not use a system with an identifiable base. Because there is no standard system for putting particular numerals in place, there is no way to line them up. That's not to say that the arabic numerals themselves were superior, but the organization ceratinly was. You can, for example, line up 153 × 802 vertically very easily, with each digit occupying it's own place. Even with placeholder zeros you could not do this using roman numerals. Simply by memorizing a multiplication table from 0-9, one can multiply and divide any number vertically in this way in arabic numerasl, but not in roman ones. This is why many medieval European mathematicians and astronomers began to use arabic numerals - because they saw the convenience (so claimed and documented by the men themselves, not just conjectured later by historians). Integer exponentiation is not easy with arabic numerals, but requires conversion to multiplication, so one might imagine an even more convenient number system. Exponentiation and its inverse, logarithmetic, with noninteger solutions are still extremely difficult in arabic numerals, but would require infinite iterative processes in any number system (since they result in irrational numbers).

So yes, I think (and have heard in History once or twicethe conversion from roman to arabic numerals was in many ways a result of convenience.75.187.197.2 (talk) 01:41, 15 January 2008 (UTC)[reply]

Ⅱ Ⅲ Ⅳ Ⅴ Ⅵ Ⅶ Ⅷ Ⅸ Ⅹ Ⅺ Ⅻ are all in unicode.

But don't replace the ISO Roman numerals with them. Morwen says they have been deprecated from Day 1 and are there just for compatibility issues. WhisperToMe 17:20, 1 Jul 2004 (UTC)

Are ISO Roman numerals from an ISO standard?--Jusjih 08:38, 16 February 2006 (UTC)[reply]

This article was just cited in an Associated Press story today,with its mention of a supposed Roman ruler ordering a change from "IV" to "IIII",which as I understood it was older(subtractive notation being an innovation even if during classical times).However,the article refers to an "incorrectly" made clock...in Roman times there were sundials,but not clocks!(Could the ruler have been more modern?)--L.E./12.144.5.2/le@put.com

I just corrected (= deleted) this. IIII was very common before the middel ages. It was only in the middel ages that roman numerals where "standarized". When mulitiplication and division became important the rules listed on the page where developed. For a roman it was much more natural to actually write IIII than IV. Generally romans didn't like subtractions. And they even wrote IC for 99. It was a bit more normal to use IX than IV, but VIIII was still heavilly used. Maybe a new section ought to be added about Roman numberals as used in the ancient times. (That is: write so you are understod as the only rule it seems.) Many of these variations have survived because people tend to see them in old books and arcitetctual places. Anyone that understands norwegian might want to incorporate some of the information there about older practices from there.

I noticed that the clock section says it's traditional to use "IIII" rather than "IV" on clocks, and gives a number of reasons why. But I've never seen a clock with "IIII" on it. All the Roman numeral clock faces I've ever seen had "IV" on them. Is this something that was done in Classical times (on sundials, I guess), or is this a European/American thing or something? --209.108.217.226 22:03, 7 October 2005 (UTC)[reply]

The Romans had both sundials and clocks (notwithstanding any article - online or otherwise - which says they had only sundials). Their water clocks (clepsydrae) were based on Graeco-Egyptian patterns, often fairly ingenious in design and manufacture, and could be mechanically sophisticated. There's an excellent Wiki article on them.

Pompey Magnus is known to have off-handedly requested a clepsydra to curtail the long-winded speechifying of senators in debates. Further references to these timepieces may be found in any decent encyclopedia, or from original citations (free of charge) online at the Perseus website at Perseus-Tofts; enter its Lewis and Short Latin Dictionary database using clepsydra as search term. Once the entry is found, pointers lead to various other publications for cross-referencing. dmadams 82.153.120.113 00:57, 27 January 2006 (UTC)[reply]

For those that might be inclined, I have added an article on Roman arithmetic which instructs the reader how to perform addition, subtraction, multiplication and division operations using Roman numerals without converting them to Arabic numerals.

Once you get the hang of it, it is rather easy, but not as easy at using Arabic. denorris 05:24, 31 Aug 2004 (UTC)

That title is spelled wrong, and there was already a Roman arithmetic article. I guess they should be merged...have fun :) Adam Bishop 05:48, 31 Aug 2004 (UTC)

That is what I get for working late. Corrected denorris 06:03, 31 Aug 2004 (UTC)

The following was removed from the article:

But that can't be true because many Romans were educated in Greek whose astronomers used Greek numerals in such a positional notation system, including a special symbol for zero. The Romans could have simply transcribed the Greek version into Latin characters if they had a need for such a system.

The Greeks did not use a positional notation system.

Ptolemy (circa 85 - circa 165 AD) used ο (ouden) to represent nothing (Number Words and Number Symbols, Karl Menninger, 1969, pp 399) but this is only part of the necessary ingredients for a positional notation system.

In Greek numerals, a count of one is represented by α (mia) and a count of ten is represented by ι (de´ka).

A Greek positional notation system would have a count of ten represented by α (mia) in the tens column followed by ο in the units column (αο instead of just ι).

In a further example, the numerals to represent a count of eleven would have been αα. However, the ancient Greeks would have used ια just as the Romans used XI and not II.

I have shown in Roman arithmetic that it is possible to perform the four basic arithmetic functions (addition, substraction, multiplication and division) using Roman numerals without a positional notation system. No doubt the Greeks used similar approaches.

However, the Greeks did have a character to present each count from 1 to 10 (a decimal system) while the Romans only used 2 characters; one to represent 1 and the other to represent 5 (a bi-quinary coded decimal system). Regardless, they were not decimal positional notion systems. That had to wait for the concept of zero as a number in the 6th century.

--Denise Norris 19:00, Sep 22, 2004 (UTC)

We have a misunderstanding here, no doubt due to my poor wording — I will attempt to reword the paragraph to take your confusion into account. You are correct that basic Greek numerals are not positional. Nevertheless, Ptolemy used a sexagesimal (base 60) positional system copied from Babylonia but using non-positional Greek numerals for each digit (0–59), as the following explanations by experts show.

[The sexagesimal system] was taken over by the Greeks (one may guess by the Hellenistic astronomers) from the Babylonians as a convenient way of expressing fractions and (to a lesser extent) large numbers, and of performing calculations with them. It is the first place-value system in history [Babylonian and Greek]. In the translation and notes I use the convenient modern 'comma and semi-colon' notation [devised by Otto Neugebauer in the 1930s], in which 6,13;10,0,58 represents 6 × 60 + 13 + 10 × 60^–1 + 0 × 60^–2 + 58 × 60^–3. Ptolemy uses the system only for fractions, and represents whole numbers, even when combined with sexagesimal fractions, by the standard Greek (alphabetic) notation. The translation follows the mixed notation (thus the above number would be written 373;10,0,58 in the translation, and τογ ι ο νη in Greek [all except the isolated ο have an overbar]. — G. J. Toomer, Ptolemy's Almagest (1998), pp.6-7.

The sexagesimal place value notation, including a symbol for zero, is of course of Babylonian origin. By its adoption in Greek astronomy it also became the standard method in Indian, Islamic, and western European treatises and tables. The method of writing the single digits is insignificant. The alphabetic notation is used in Greek and Arabic texts, Roman numerals in Latin, Hindu numerals in Sanskrit. The essential point, common to all, is the place value notation and the use of a zero symbol. The modification of this notation to decimally written numbers as well, which took place in India, produced the "Hindu numerals" which we use now and which appear in slowly increasing frequency in the later Middle Ages in Arabic as well as in Byzantine and Latin texts. For the computational methods this is of very little importance since it does not matter in what form the individual digits are written. — Otto Neugebauer, A History of Ancient Mathematical Astronomy (1975), p.1113.

Neugebauer implies that writers did use Roman numerals for the individual digits in a sexagesimal system!

— Joe Kress 01:19, Sep 25, 2004 (UTC)

This is an article on Roman numerals, so unless it has a proven impact the Roman numeral system, perhaps this discussion should be moved to Greek numerals.

Also, Ptolemy was a Roman Empire era scholar. Clearly his work did not influence the development of Roman numerals. I would imagine that an article on Roman Era Math would be better suited to discuss the influence of Ptolemy rather than Roman numerals.

My specific objection was the conclusion (badly worded or not) that Roman's had a positional notation system simply because the it is assumed the Greek's had one. Both conclusions are un-supportable. The specialized use of Babylonian sexagesimal counting by Ptolemy in Roman/Greek society is akin to the use of calculus in modern society.

--Denise Norris 07:43, Sep 25, 2004 (UTC)

I don't get how the flipped C works. Can someone explain it better? lysdexia 08:00, 10 Oct 2004 (UTC)

What I was told in Latin class boils down to this: originally, the Romans used the Greek letter Phi (Φ), which was unused in their own alphabet, to denote the number 1,000. To denote half its value (i.e., 500), they halved the letter Phi, resulting in the regular "hardware representation" I+reverse C, which in turn was standardised into D. Reversed-C on its own I've never seen (but that doesn't mean it may not exist). This, by the way, contradicts what the first paragraph says about I+reverse C being an old representation of M, I think someone's got it confused with D. Correct this? — Cwoyte 14:01, Dec 3, 2004 (UTC)

The reverse-C is the "apostrophic C" [U+2183]. The notation (|) for 1000 is likely to be the origin of the M symbol, rather than M-for-mille. Each level of bracketing multiplies the represented value by 10: ((|)) = 10,000; (((|))) = 100,000. Half symbols were used to represent half the value: |) = 500 [hence D]; |)) = 5000, |))) = 50,000. However, there may be no evidence for numbers greater than 100,000 were represented in this way -- the overbar being used instead (later?): single overbar (with small drops, [ on its side) for x 1000; full-three sides for x 100,000. The overbar multiplication was apparently not applied to I,V, L (?).

I have really been looking for fractional notations: The reverse-C appears as '1/4' in some contexts.

It is useful to remember that arithmatic was performed on Abaci, not with the written numbers (as we would with Arabic/Indian numerals). Roman numerals are a direct representation of what is on the Abicus. --Sawatts 11:56, 8 Dec 2004 (UTC)

I highly doubt that this joke could possibly qualify as encyclopedic. Can we remove it and its associated redirect Barney is Satan? -- Antaeus Feldspar 17:50, 10 Nov 2004 (UTC)

I agree. As far as I'm concerned, go ahead and delete it. — Cwoyte 14:01, Dec 3, 2004 (UTC)

Actually, I took the step and edited the Barney bit out. As for the Barney is Satan redirect, I do not know how to remove that. — Cwoyte 22:12, Dec 5, 2004 (UTC)

I've added the Barney is Satan redirect to Wikipedia:Redirects for deletion and {{rfd}} to the redirect (although it doesn't appear due to a bug). — Joe Kress 19:26, Dec 7, 2004 (UTC)

I wonder what the sentence

When describing members of a list, first A, B, C, D tended to be used, then 1, 2, 3 then i, ii, iii, iv.

intends to say.

Romans certainly didn't use 1, 2, or 3. The arabic numbers entered europe long after the fall of the Roman empire. According to Georges Ifrah, Universalgeschichte der Zahlen, the Codex Vigilanus from 976 C.E. is the oldest european work containing arabic numbers.

The sentence seems to require some historical context. Given the fact that this context is not given, I suggest to remove the sentence. --Kune 22:40, 2005 Jan 27 (UTC)

The article said:

Some rules regarding Roman numerals state that a symbol representing 10^x may not precede any symbol larger than 10^x+1. For example, one should represent the number "ninety-nine" as XCIX, not IC. However, these rules are not universally applied.

The last sentence is wrong. MIM for 1999 is not kosher, any way you look at it. So I removed the last sentence. Egil 07:31, 3 Feb 2005 (UTC)

That last sentence reflects that fact that - "kosher" or not - some people do it anyway. Since the role of Wikipedia is not to proclaim what's right, but to describe what's done, I've restored a slightly modified version of that statement. Tverbeek 15:20, 3 Feb 2005 (UTC)

I beg to disagree. There is no reason to report usage that is patent nonsense, unless in very particular cases. The statement However, these rules are not universally followed is just as bad as the previous one. As far as I know, the only usage of 'MIM' and 'IMM' is by people who haven't even bothered finding out what roman numbers really are. The 'pedia should report correct information. Your sentence gives the impression there is doubt about what the correct usage is. This is misleading. And if you feel the need to state that "People do not always bother following rules", it should be done in the context of human behaviour in general. -- Egil 17:23, 3 Feb 2005 (UTC)

One point of that section is that the question of "correct" usage isn't as simple as you state. Not only has usage varied somewhat with time and place, the Romans themselves exhibited some inconsistency in their usage, and a degree of personal preference seems to have been involved. Certainly we can and should spell out the usage that's most prevalent, but since no one can find the original RFC or ISO standard for them, the position that there is an indisputable standard for "correct" usage - and that you have it - seems hard to justify. Tverbeek 20:52, 3 Feb 2005 (UTC)

Tverbeek is right: Even today, there no indisputable standard defining these rules. Moreover, the only documents I know who codify these rigid rules are modern. Are there any Roman documents known who describe the Roman numbering systems? If so, do they codify the stringent rules, or do they allow IM etc., or don't they mention the problem? Are there any medieval documents known who describe the Roman numbering systems? Same questions apply. -- Adhemar, 12 December 2005

The usage XIIX for 18 is attested in actual usage in medieval times, and I think IC for 99 is also. People who actually wrote and read these numerals could communicate unambiguously with a slightly more flexible version of "The Rules", so who are we to be throwing around epithets such as "patent nonsense"? My guess is that "The Rules" were written by printers round about the time that they standardized spelling. Tverbeek is right. Cbdorsett 07:09, 6 Feb 2005 (UTC)

I came across a photo of an ancient inscription with the numeral XIIX (the tomb of Secundinus on the Via Appia). Does anyone know of an ancient example of the use of IC, IM or XM? --Zundark 14:03, 6 Feb 2005 (UTC)

In Rome, archway number 29 of the Colosseum has the inscription XXVIIII. The MathWorld article on Roman numerals cites (Menninger 1992, p. 281; Cajori 1993, p. 32) that "Romans occasionally wrote IM, IIM, etc." -- Adhemar, 12 December 2005

How about VL for 45? I for one don't think this is a 'decimal' system, so the rule about subtracting exactly one-tenth seems suspect. Aleš Wikiak 21:03, 25 August 2006 (UTC)[reply]

I was wondering the exact same thing. I've never seen VL or LD used before. Rather than trying to put it in terms of 10^x and 10^x+1, why not just explicitly spell out which letters may precede others. It's not like there are that many. I can precede V and X, X can precede L and C, and C can precede D and M. That's it. One short sentence. Alexwagner 13:16, 14 June 2007 (UTC)[reply]

Food for thought: Roman numerals were not inspired by a decimal number system... or does anyone care to provide a reference to a document proving otherwise?

• As such, original rules could not have implied that every decimal digit is to be spelled out separately. Note that this has nothing to do with common modern use (or confusion).
• Can anyone provide a reference to a "Roman numerals decimal reform" that would establish the new standard?
• Note that whatever form a number takes (e.g IMM or MIM or MCMIC or MCMXCIX or even MDCDIC, MDCDXCIX, ...) it is always unambigous as to what its value is. The only question is what is the preferred way to write it. Speaking of preference, mine is the one that is easier to look and understand and IM for 999 beats the alternatives any day (one less than a thousand).
• If one wants to specify letters that can precede others (to avoid introducing decimal rules where they do not belong), they would probably have to think about how "distant" the letters are. For example, there are no (currently standard) letters between I and V, hence IV is OK. Same happens with XL for example. Since IX, XC and CM are considered valid, then at least two previous letters must be allowed. Hence VL and LD should also be fine. However, if "in spirit" Roman numberals were to simply show the value in the simplest form, then as 999 is so close to 1000, it should be similar to it... hence IM and not CMXCIX or, worse, say, DCDLVLIV... Then remember that there were other letters in use as well...
• The fact that we today spell out roman numbers while thinking in decimal shows just that ... that we are thinking decimal... only to end up with ugly representations in the end...
• The real fun begins now... is "IXC" valid and, if it is, is it (IX)C = 100-(10-1)=91 or I(XC) = (100-10)-1=89... :)

--Aleksandar Šušnjar (talk) 21:40, 16 June 2008 (UTC)[reply]

A not mathematically inclined friend of mine said that IVC = 100-4 feels better than IVC = 100-5-1. Here's an interpretation of all strings over {I,V,X,L,C,D,M} that treats the unambiguous cases correctly.

-- This software was written in 2008 and is granted to the public domain.

roman :: String -> Integer
roman xs = rom (map val xs)

val 'I' =    1
val 'V' =    5
val 'X' =   10
val 'L' =   50
val 'C' =  100
val 'D' =  500
val 'M' = 1000

rom [] = 0
rom xs = let (ys,zs) = salm xs in last ys - rom (init ys) + rom zs

salm xs = spaf (== foldr1 max xs) xs  -- split after leftmost maximum
spaf p (x:xs) = if p x then ([x],xs) else let (ys,zs) = spaf p xs in (x:ys,zs)

-- 85.179.144.165 (talk) 09:14, 18 September 2008 (UTC)[reply]

The are lengthy learned discussions on this talk page about zero; I haven't read it all, so this may be silly: I think the comment "There was no need for a zero." on the entry "Zero" in the long table of numerals should go; it seems to arise from a confusion of the digit zero and the number zero: The was no need for a digit zero, as e.g. CI clearly means 101, not 11, 110, 1001, or whatever. But the entries in this table are not digits; they are numbers (e.g. 1999). A Roman farmer owning zero cows had as much (or as little) need for the number zero as a modern farmer. -- But perhaps the table should be a table of digits, or rather should be split into two tables?--Niels Ø 10:40, Apr 22, 2005 (UTC)

The clause

Originally, it was common to use IIII to represent "four", because IV represented the god Jove (and later YHWH).

needs context or explanation. I assume this is a taking-the-name-of-the-lord-in-vain kind of thing, but it's too vague as is. Besides, the whole reason we use Rx for recipe instead of just R] is because Rx resembled the symbol for Jove. So why write Rx but avoid IV? kwami 05:07, 2005 May 17 (UTC)

According to a book I have the value for the 5,000 roman numeral is ${\displaystyle {\frac {}{V}}}$?

RyanJ

The overlined V (V) for 5000 is shown under Alternate forms, but it was a relatively recent development during the Middle Ages. The other forms shown were those actually used by the Romans, which were still popular throughout the Middle Ages and Renaissance. — Joe Kress 16:54, 23 July 2005 (UTC)[reply]

I don't understand what is meant by the sentence "so as to not confuse the first two digits of the century with the first two digits of most, if not all, of the years in the century." Is it just me, or does this need to be reworded?

The whole article needs to be reworded! That section should read a little better now. kwami 08:56, 2 November 2005 (UTC)[reply]

It says at Talk:English alphabet that the Romans used Z for 7. Where is the source of this?? Georgia guy 01:33, 12 November 2005 (UTC)[reply]

Z was used for the numeral 7 in Hebrew and Greek, and still is today. The numerical values of the letters of the alphabet came along for the ride with the alphabet itself, and Z likely remained associated with '7' long after it was dropped from writing, just as the Greeks retained digamma (F), qoph (Q), and sampi as numerals long after they were gone from the alphabet. Today, now that miniscule sigma has two forms, the Greeks tend to replace digamma with the alternate form (ς) for '6', and we can imagine that something similar happened with the Romans when they had a new letter G to use instead of the obsolete Z for '7'. See Gematria and Greek numerals. I don't know if the Romans ever used this as a daily system for indexing or anything instead of the Roman numerals, but you can probably find something in Ifrah Universal History of Numbers or Daniels & Bright World's Writing Systems, or many other refs. kwami 02:32, 12 November 2005 (UTC)[reply]

I've heard not only as Z as a alternative form for VII = 7 but also O = XI = 11; F = XL = 40; K = L = 50; S = LXX = 70 although far more frequently S = 1/2; R = LXXX = 80; N = XC = 90 long before Bede used N for nulla; Y = CL = 150; T = CLX = 160; H = CC = 200; E = CCL = 250; P (= sometimes also G) both equal to CD = 400 although CD can be confused with CIƆ = 1000. Were these actually used by the Romans at one time? Does anybody have a source? – Adhemar 18:14, 18 May 2006 (UTC)[reply]

I, too, have seen those "medieval Roman numerals." The ones I have seen, and this is consulting a Webster's dictionary:

I/J	1
V/U 	5
s 	7
X 	10
O 	11
F 	40
L 	50
S 	70
R 	80
N 	90
C 	100
Y 	150
K 	151 (yes, that's a 151-- don't ask)
T 	160
H 	200
E 	250
B 	300
P/G 	400
D/Q 	500
M/φ 	1000
Z 	2000

You're right, I think there should be something on medieval Roman numerals, but I'm not sure if it's worth its own article. (I also have the backward S as 1/2, not a forward S, but that I just pulled off some Web page so I'm not sure of its accuracy.) Again, these are mostly out of the Webster's, so I'm pretty sure they're reliable. (And yes, the CIɔ = ф = 1000 is accurate.) J. Myrle Fuller (talk) 01:37, 30 January 2008 (UTC)[reply]

---

All these other letters were "medeival Roman numerals." Some are listed in dictionaries, others less commonly so, but none were ever used in actual ancient times. Still, they should be listed here, as they indeed are considered "Roman numerals." However, I heard that Z represented 2 000, not 7. No, I do not have a good source; I got this information from an "answers" page (not on google, but somewhere else; I found it during a google search), but if you find a good unabridged dictionary and looked up each letter individually you could find a nearly infallible source, given the research put into these. By the way, where did you get that "CIƆ" from? I've never seen that before as a representation for 1,000. Is Ɔ=D=500, then? 75.187.197.2 (talk) 01:50, 15 January 2008 (UTC)

While this is an interesting analysis of numerals, what about Roman numbers? What were the names? In Arabic, 1=one→first, so forth. What was Roman? Trekphiler 08:51, 14 December 2005 (UTC)[reply]

I'm not sure what you're asking. <1> and <I> both represent 'one' and 'first' in English. If you want the Latin words for numbers, they can be found at List of numbers in various languages, but there's no differece between <1> and <I>. Or you could read the numbers as series of alphabetic numerals: "XIV" as ex i vee, etc. kwami 09:32, 14 December 2005 (UTC)[reply]

It's mentioned that Roman numerals are often used in English for movie titles and other things that come in series. I've noticed, though, that often only the first three in a series are represented this way, with subsequent ones represented with Arabic numbers. Often, also, the first in a series is not given a number at all. For example, Intel makes chips called the Pentium, Pentium II, Pentium III, and Pentium 4. This way they avoid the problem of using IIII or IV. Maybe some note should be made of this.

--Alkali Jack

Not always true. Rocky used Roman numerals for all five instalments, even Bart Simpson notes this: "Rocky V, that was the fifth one!" Superman also used Roman numerals up to Superman IV, and Star Trek did up to Star Trek VI: The Undiscovered Country. The Police Academy series used Arabic numerals for instalments 2 through 6; can't remember if they used a numeral at all for the 7th film. Also, whenever a movie uses Roman numerals for instalments, IV is always used for four. Bricks J. Winzer 22:18, 20 February 2006 (UTC)[reply]

And also compare Mac OS 9 with Mac OS X!. --Error 00:21, 12 September 2006 (UTC)[reply]

Some articles about years now have Roman numerals, but should that be used for very ancient years?--Jusjih 08:40, 16 February 2006 (UTC)[reply]

In the article Roman_Republican_coinage, there are a couple of hints that the Romans may have used the symbol S as a Roman numeral representing 1/2. The sestertius was a coin equal in value to 2+1/2 asses, and its value was marked with IIS or HS. Another coin, the Semis, had a value of 1/2 as, and its value was marked with the symbol S. The as itself was marked with the value I. This suggests that S may have been the Roman numeral for 1/2. --B.d.mills 22:59, 19 February 2006 (UTC)[reply]

Yes, it is presumably an abbreviation for semis (half). The Roman symbols for 1/2, 1/3, 1/4 are S, Ɔ, Z or their graphic variants Σ, ), 2. kwami 02:21, 20 February 2006 (UTC)[reply]

Read a text some years ago, that said the following about the D in the roman numerals (The following how I recall the sentence): D was used as 500, because it is the first letter of Demi-mille, which means "half of thousand", which is 500.

Can anyone substantiate this? -Hecko 21:17, 30 March 2006 (UTC)[reply]

Based on the ones you just took out of your question (L is for Legion, etc.), I suspect the whole pamphlet was based on folk etymology. These symbols have histories, and they were not originally letters of the alphabet. This is true for D as well. However, folk etymology can shape the evolution of words and of symbols. The numerals C and M seem to have been conflated with the letters C and M due to the words centum and mille, which were natural mnemonics for their shapes. It's very possible that demi-mille played a similar role in the evolution of D (the reason why it was conflated with D rather than some other letter), but it isn't the origin of the symbol itself. One of the reasons we know this is that even after it appeared with the modern shape D, there were other variants that were not identical with letters of the Latin alphabet, or even variant shapes of letters. kwami 23:56, 30 March 2006 (UTC)[reply]

In harmonic (musical) analysis, lower-case roman numerals are used to indicate minor triads while upper-case roman numerals are used for major triads. Assuming the chart in the article refers to the degrees in the major scale, the supertonic, mediant and submediant should be indicated with lower-case numerals. I changed the chart to reflect this. The leading tone triad is diminished. It should be lower-case with a superscript 'o' to the right of it. I can't figure out how to add the superscript. Maybe someone else can.

I have seen certain dictionaries that use upper-case roman numerals all around while giving a brief explanation of harmonic analysis. However, as a theory and composition student, I use this system just about every day and I can say that, without exception, the lower-case roman numerals are used for these triads. —Preceding unsigned comment added by Curtis wright 21 (talk • contribs)

I used a small superscript degree symbol (vii^°) to indicate a leading tone because it was closer to the symbol in leading tone triad than was a simple superscript o (vii^o). — Joe Kress 09:09, 16 April 2006 (UTC)[reply]

I was hoping to find mentioned in this topic, the use of MM and mm to indicate million, and the history of that practice.

After introducing this subject I thought one of the experts would incorporate it. After a few months no one did, so I added it to “other modern usage…”

Now I see that it has been removed. Too bad, because when you read a financial report it is common to see both MM and mm to indicate million. However it is difficult to find the definition in writing. Of course it is easy to infer the definition but it would be nice to find it in writing somewhere.

It's beyond me why anyone is still using this archaic system! It tends to be popular among academics with a fetish for ancient things.

Yes, but there's nothing wrong with that, right? Sometimes knowing obscure information can be fun, especially if you get a chance to show it off. It could help one win money on a game show, too.  ;) --Lance E Sloan (talk) 13:52, 3 January 2008 (UTC)[reply]

In a printed book dated of 1668, I found the following: Nec aliter ediderat Fr. Raphelengius anno cIɔIɔxcyI What on earth did he mean with cIɔIɔxcyI? --Ciacchi 22:47, 10 August 2006 (UTC)[reply]

The first two 'signs' are obvious: cIɔ means 1000, and Iɔ means 500. This means that xcyI somehow means 168. However, 168 should be CLXVIII according to the usual rules. It may be the result of a combination of typographic errors and some unknown rules. — Joe Kress 03:59, 11 August 2006 (UTC)[reply]

Assuming y is a badly written v (rather than some shorthand for 150), CIƆIƆXCVI would mean 1596. – Adhemar 18:20, 19 July 2007 (UTC)[reply]

well?

Pece Kocovski 01:45, 7 September 2006 (UTC)[reply]

For the normal ones, just use the corresponding alphabetic keys; for the exotic ones, configure your editor to add a few keybindings, or define your own keyboard layout, or learn the Unicode code points by heart and use the alt-digits combo. There are also some precomposed glyphs in Unicode, but I don't think they're generally recommended. Shinobu 23:46, 11 September 2006 (UTC)[reply]

Several juvenile quips (such as "D-bar is Danny Nickles" and "Barney is Satan") have been placed in the text. In most instances, the sense of the original text is lost. — Unknown 03:41, 13 September 2006 (UTC)[reply]

Fortunately, the page history can be used to repair the damage. Alertness good, worry bad. Shinobu 13:03, 13 September 2006 (UTC)[reply]

Rules regarding Roman numerals often state that a symbol representing 10^x may not precede any symbol larger than 10^x+1. For example, C cannot be preceded by I or V, only by X (or, of course, by a symbol representing a value equal to or larger than C). The "For example ... " part is the thing which is confusing. The first sentence says (for x = 2): A symbol representing 100 ( = C) may not precede any symbol larger than 1000 ( > M). But then it says: For example, C cannot be preceded by I or V, only by X. I absolutely cannot see any correlation between the 10^x / 10^x+1 rule and the I/V/X rule, because V ( =5) will require a logarithm equation to be expressed as 10^x. ;) -andy 80.129.84.239 05:04, 11 October 2006 (UTC)[reply]

For example, C (10^2), cannot be preceded by I (10^0) or V (10^log5 = 10^.7), only by X (10^1).

According to the rule: I (10^0) may not precede any symbol larger than X (10^1), like l (10^1.7). Similar V and L, and X and C. Thus far consistent with the example.

What troubles me is the notion that VL would be allowed instead of XLV. How I thought it worked (only preceding to subtract is shown):

Et cetera. I short, I thought 5*10^i (i integer) was not allowed to precede stuff to subtract. Shinobu 10:36, 11 October 2006 (UTC)[reply]

...are a right fine pain in the ass. When I become Lord of the Internet, this is all that the Wikipedia text shall say. Just a friendly "thanks" to those of you who managed to write this impressive beast. --70.108.140.252 12:55, 22 October 2006 (UTC)[reply]

How were roman numbers originally pronounced? For example, these days the number CMLX (1960) would just be pronounced as 'nineteenhundred sixty', because the reader first converts it to decimal notation, and pronounces the number accordingly. Did the romans actually say 'C M L X' ? 82.94.235.106 10:02, 6 November 2006 (UTC)[reply]

(not that anyone will read this) I would imagine that they used the Latin names for the numbers. Since most of the people were not literate, the speech in general probably wasn't affected much by the writing. But of course I'm conjecturing. Fishal 22:44, 17 May 2007 (UTC)

I think you're right, Fishal. If anyone pronounced CMLX cee em el ex, I imagine people would either think they were innumerate or joking. kwami 23:18, 17 May 2007 (UTC)[reply]

It is said:

It is also worth noting that writing 99 as "IC" or "ic", in Old English texts would appear identical to the word "Ic" meaning "I".

In fact it is most definitely not worth mentioning this, imnsho, because the fact is, the Roman numeral for 1 is I which also means I no less than Ic might have in the past. I'm removing this line from the article text because it simply doesn't make sense to state. D. F. Schmidt 02:32, 11 November 2006 (UTC)[reply]

The following sentence seems to be speculative "original research" and tangential to the page topic:

"The possibility of using IC for 99 provides one explanation why Cadbury's smaller, unwrapped Flakes (for inserting into ice creams) were designated 99 Flakes."

I have removed it. Alki 22:55, 29 November 2006 (UTC)[reply]

"As it relates to the nomenclature of inorganic compounds, only IV should be used. For example MnO2 should be named manganese (IV) oxide; manganese (IIII) oxide is unacceptable."

This explanation makes no sense at all, both IIII and IV represent 4. I have never seen IIII used to represent 4 in science literature, however without a source that states that only IV can be used I wouldn't say IIII is unacceptable. 211.28.194.8 09:56, 20 January 2007 (UTC)[reply]

There is a roman numeral calculator at:

http://www.guernsey.net/~sgibbs/roman.html?#

I wanted to add this to external links but there is no edit link. Also, there is some vandalism on the page in the link section but I cannot get at the section to remove it so I hope that someone else can. —The preceding unsigned comment was added by JeanKorte (talk • contribs) 21:07, 4 February 2007 (UTC).[reply]

"They are also sometimes used in the credits of movies and television programs to denote the year of production, particularly programs made by the BBC." It has been my own experience that they are not just sometimes, but very, very often used in such instances. I can't say anything for particularly the BBC, though. Is there any way this can be backed up? 69.95.237.253 23:35, 7 February 2007 (UTC)[reply]

I thought that roman numerals were often used in copyright notices (including movie credits) in order to obfuscate the actual year the movie was created so that people don't dismiss the movie as being outdated. Can anyone verify this? Jesushouston 00:56, 26 April 2007 (UTC)[reply]

When I was a kid in the 1970's, I used to wonder why films credited the year via roman numerals, it made me mad because I wanted to know the years certain cartoons from WB and MGM were made. My learning of the entire roman numeral system came from this experience. I thank them for it. In the US, most all did that . Andy Griffith show shows the actual date, The Beverly Hillbillies shows the roman numeral, for example. Some programs that do show the number date tends to made it look illegible to read, another circumvent to disguise the actuasl time. —Preceding unsigned comment added by 74.34.136.181 (talk) 20:18, 21 February 2008 (UTC)[reply]

This code is in Scheme. I'm putting it here for verification before introducing into the main article. I claim this code with correctly romanize integers from 1 (I) to 3888 (MMMDCCCLXXXVIII), if not further. I submit that this code would improve the article by clearly and concisely exhibiting how romanization is performed. I wrote this code; any ownership I have is forfeit. Please respond with criticism/improvements and comments. When you feel the code has been sufficiently verified, please move it to the article. 74.130.9.41 02:51, 8 June 2007 (UTC)[reply]

(define romanize
  (lambda (n)
    (cond
      [(>= n 1000) (string-append "M" (romanize (- n 1000)))]
      [(>= n 900) (string-append "CM" (romanize (- n 900)))]
      [(>= n 500) (string-append "D" (romanize (- n 500)))]
      [(>= n 400) (string-append "CD" (romanize (- n 400)))]
      [(>= n 100) (string-append "C" (romanize (- n 100)))]
      [(>= n 90) (string-append "XC" (romanize (- n 90)))]
      [(>= n 50) (string-append "L" (romanize (- n 50)))]
      [(>= n 40) (string-append "XL" (romanize (- n 40)))]
      [(>= n 10) (string-append "X" (romanize (- n 10)))]
      [(>= n 9) (string-append "IX" (romanize (- n 9)))]
      [(>= n 5) (string-append "V" (romanize (- n 5)))]
      [(>= n 4) (string-append "IV" (romanize (- n 4)))]
      [(>= n 1) (string-append "I" (romanize (- n 1)))]
      [else ""])))

This code is in Scheme. I'm putting it here for verification before introducing into the main article. I claim this code with correctly arabicize Roman numerals from I (1) to MMMDCCCLXXXVIII (3888), if not further. I submit that this code would improve the article by clearly and concisely exhibiting how arabicization is performed. I wrote this code; any ownership I have is forfeit. Please respond with criticism/improvements and comments. When you feel the code has been sufficiently verified, please move it to the article.

Please notice this is considerably more difficult than romanization! Notice that incorrectly formatted Roman numerals will fail, at the cost of some extra error-checking code. This could easily be removed if thought too confusing to be worthwhile. In particular, the error-checking has that I may only subtract V and X, X may only subtract L and C, and C may only subtract D and M. V, L, D, and M may not subtract. For example, given the incorrectly formatted numeral IC, this code will throw an error, not 99 (99 is properly written XCIX). Also note that only I, V, X, L, C, D, and M are supported.

Notice arabicizing the empty string will return 0.

74.130.9.41 21:38, 8 June 2007 (UTC)[reply]

(define arabicize
  (lambda (s)
    (let ([len (string-length s)])
      (cond
        [(eq? 0 len) 0]
        [(eq? 1 len)
         (let ([fc (substring s 0 1)])
           (cond
             [(string=? "I" fc) 1]
             [(string=? "V" fc) 5]
             [(string=? "X" fc) 10]
             [(string=? "L" fc) 50]
             [(string=? "C" fc) 100]
             [(string=? "D" fc) 500]
             [(string=? "M" fc) 1000]
             [else (error "Cannot parse incorrectly formatted Roman numeral")]))]
        [else
         (let ([fc (substring s 0 1)]
               [sc (substring s 1 2)]
               [restone (substring s 1 len)]
               [resttwo (substring s 2 len)])
           (cond
             [(string=? "I" fc)
              (cond
                [(string=? "V" sc) (+ 4 (arabicize resttwo))]
                [(string=? "X" sc) (+ 9 (arabicize resttwo))]
                [(or
                  (string=? "L" sc)
                  (string=? "C" sc)
                  (string=? "D" sc)
                  (string=? "M" sc))
                 (error "Cannot parse incorrectly formatted Roman numeral")]
                [else (+ 1 (arabicize restone))])]
             [(string=? "V" fc) (+ 5 (arabicize restone))]
             [(string=? "X" fc)
              (cond
                [(string=? "L" sc) (+ 40 (arabicize resttwo))]
                [(string=? "C" sc) (+ 90 (arabicize resttwo))]
                [(or
                  (string=? "D" sc)
                  (string=? "M" sc))
                 (error "Cannot parse incorrectly formatted Roman numeral")]
                [else (+ 10 (arabicize restone))])]
             [(string=? "L" fc) (+ 50 (arabicize restone))]
             [(string=? "C" fc)
              (cond
                [(string=? "D" sc) (+ 400 (arabicize resttwo))]
                [(string=? "M" sc) (+ 900 (arabicize resttwo))]
                [else (+ 100 (arabicize restone))])]
             [(string=? "D" fc) (+ 500 (arabicize restone))]
             [(string=? "M" fc) (+ 1000 (arabicize restone))]
             [else (error "Cannot parse incorrectly formatted Roman numeral")]))]))))

Roman numerals is a numbers in a different language eg I is 1 II is 2 III is 3 IX is 4 and X is 5 ect.

Not sure how constructive this comment is. No, IX is not 4 and X is not 5, nor are Roman Numerals a language. I'm not sure I understand this. —Preceding unsigned comment added by 75.187.197.2 (talk) 01:01, 23 January 2008 (UTC)[reply]

Should the chart be replaced with {{Unicode chart Number Forms}}:

I don't like how it looks, but that entire group of templates could use some cleanup. —Random832 14:27, 20 September 2007 (UTC)[reply]

I see Japanese characters in my web browser, Firefox, when I look at that chart. However, when I view the PDF, I can see that the chart should contain Roman numerals in place of the Japanese characters. This may vary from one browser to another. It would be nice to have an explanation (or a link to one) of how to get the Roman numerals to appear in a browser correctly. --Lance E Sloan (talk) 14:45, 3 January 2008 (UTC)[reply]

I also use Firefox but do see the PDF Roman numerals, not Japanese characters. Nevertheless, I still see Chinese/Korean/Japanese characters when they appear in Wikipedia articles. The difference may involve how I initially downloaded Firefox because I knew I wanted to see all characters used throughout the world. — Joe Kress (talk) 05:25, 4 January 2008 (UTC)[reply]

i have read that R should mean 250 and N 900... [this unsigned entry was moved from the middle of an entry above]

You might be thinking of an alphabetical numbering system where A = 1, B = 2, etc., but in that case N would be 40, and R, 80. kwami 01:36, 3 October 2007 (UTC)[reply]

ok now i am taking latin 1 on virtual virginia and one of the web sites that my perfessor gave to us was a web page that said that for the roman numeral 4, 5 and 6 they would put llll, lllll, and llllll some times is this true? —Preceding unsigned comment added by 74.127.73.88 (talk) 00:26, 5 January 2008 (UTC)[reply]

I have been told in school that the origin of using I's as units came from counting on one's fingers, where fingers resemble I's. V's resemble the space between the forefinger and the thumb when all five fingers are extended. X's resemble two V's put together. C simply represents "centum," and M "mille." I think the origins of L and D are listed. (Aren't they?) The origins of bars are obvious shorthands using previous symbols.

Is this true, and can anybody find a source for it?75.187.197.2 (talk) 02:09, 15 January 2008 (UTC)[reply]

This folk etymology is covered in the article. kwami (talk) 07:33, 30 January 2008 (UTC)[reply]

How about a bullet pointed list instead of asterisks in "Modern Usage"? Asterisks look a bit shoddy and are not a suitable list marker. —Preceding unsigned comment added by 79.67.83.161 (talk) 15:09, 3 February 2008 (UTC)[reply]

That list already uses bullets. An asterisk on an edit page is automatically displayed as a bullet by the Wikipedia software. — Joe Kress (talk) 09:30, 4 February 2008 (UTC)[reply]

This article describes the use of "subtractive" numerals as something going back to classical Roman times, with a taboo against "IV" as an abbreviation of Jupiter. In one paragraph the Jupiter concept is stated as fact, in another as a possibility.

The development of subtractive Roman numerals is usually described as a Medieval innovation -- any claim of being a classical Roman practice needs citations and hopefully a coin or something showing it in use. The "Jupiter" concept for IV may be completely unfounded and also requires a citation -- and should be presented as a theory not fact unless there are actual classical Roman sources describing this taboo rather than modern speculation.

The origin of Etruscan numerals is described as being "tally stick" notches, and not alphabet letters.

But all of the Etruscan "notch" symbols shown are actually letters of the Greek-derived Etruscan alphabet which you will find here: http://www.unicode.org/charts/PDF/U10300.pdf. And the Greeks had a system of using Greek letters as numbers (a simple, Roman-like system using a few letters in archaic times, a more complicated system using the full alphabet in classical times). So the entire "notch" concept needs LOTS of references and should be presented as a possible alternative if it can be supported at all.

This entire article looks like some inaccurate impressions and speculations by a math major and not a sound scholarly resource. —Preceding unsigned comment added by 75.36.148.166 (talk) 04:39, 17 February 2008 (UTC)[reply]

First of all, no need to shout. I'm deleting your bolding.

About the taboo thing, I have no idea.

The tally stick theory comes from Georges Ifrah, 2000, The Universal History of Numbers. And your objections are original research - because the Unicode range for Old Italic includes a few extra symbols (I, Λ, X, and T) for numerals (1, 5, 10, 50), you decide that that somehow invalidates the article. It does nothing of the kind, and in fact supports the claim you are disputing. Do you have any objection founded on anything other than your imagination? kwami (talk) 07:41, 17 February 2008 (UTC)[reply]

Firstly, it's too bad that 75.36.148.166 doesn't identify himself, because his comments are spot on, especially the last paragraph. Secondly, Kwami, you should get a life that doesn't involve computers or the internet. Freddy011 (talk) 21:59, 6 December 2008 (UTC)[reply]

Proper Images/Unicode Encoding for Etruscan Characters

See previous comment on likely errors in this article about the nature of the Etruscan numeral system.

The Etruscan numerals are currently shown in imitated form using a combination of Latin alphabet letters and numbers, Greek Lambda, a mathematic "circled plus" symbol, and some garbled character put in markup ["span class="Unicode"]⋔[/span].

They should all be actual Unicode characters which you will find here:

http://www.unicode.org/charts/PDF/U10300.pdf

Wikipedia pages are served in Unicode-encoding (UTF-8) so these characters can be typed directly into the page code. But as you likely don't have an Old Italic keyboard layout handy, you can use numbered Unicode entities (ampersand-x-number-semicolon) instead.

But as most readers wont have an Old Italic font installed, each letter should actually be presented as an in-line graphical image (which you'll have to draw, the Unicode chart images are copyrighted), with the Unicode ampersand-x-number-semicolon code as its alt tag.

Sorry that I don't have time to do this myself -- and I don't know what the missing "?" character is supposed to be anyway. —Preceding unsigned comment added by 75.36.157.152 (talk) 04:52, 17 February 2008 (UTC)[reply]

• I think the anonymous editor is right about one thing for sure: the article is pretty light on citations. Kwami provided one - anyone else is welcome to find more. No problem with flagging and blatantly unsupported assertions with {{Cite}}; please go right ahead. I myself have suspected that any taboos against using the name or initials of Jupiter were Christian innovations, but don't have any cites. It doesn't sound logical to me that the pre-Christian Romans would have had a taboo like that, but I'll leave that to those who have studied Roman history and religion. I would also like some more actual support for the origin of the subtractive principle. I have read that sometimes the ancient Romans would pile all the digits together without regard to our modern concepts of ones-tens-hundreds, etc. Subtractive numerals would not work then. When was the one system dropped and the other adopted? Cbdorsett (talk) 22:11, 17 February 2008 (UTC)[reply]

The Origins section comes straight out of Ifrah. As for replacing Etruscan numerals with their Unicode equivalents, that might be useful, but there was a lot of graphic variation that Unicode conceals. kwami (talk) 23:12, 17 February 2008 (UTC)[reply]

"The four-character form IIII creates a visual symmetry with the VIII on the other side, which IV would not (with the exception of square faced watches and clocks, where the opposite number is X)."

This doesn't seem to make any sense. Square-faced timepieces I've seen have the numbers in the exact same order as for round-faced ones. 10 is diametrically opposite 4, and 8 is opposite 4 with respect to a vertical axis. -- Smjg (talk) 19:28, 20 March 2008 (UTC)[reply]

"The four-character form IIII creates a visual symmetry with the VIII on the other side, which IV would not ... "

I read this idea about visual symmetry some years ago, and it didn't make sense either. But I think the mistake is considering the visual symmetry just between the numbers 8 and 4. Your eyes don't behave as a pendulum bouncing back and forth between those two positions of the clock.

A visual appeal (I wouldn't call it symmetry) makes a lot more sense if you divide the whole face of the clock in three parts: The 1st part (the four hours 1 to 4) will contain only Is; the 2nd part (the four hours 5 to 8) will be the only part with Vs in it; and the 3rd part (the four hours 9 to 12) will be the only part with Xs in it. -- Vikfra (talk) 19:28, 13 May 2008 (UTC)[reply]

An excerpt from "The Modern Clock", pp. 428-429...

We often hear stories concerning the IIII in place of IV. The story usually told is that Louis XIV of France was inspecting a clock made for him by a celebrated watchmaker of that day and remarked that the IV was an error. It should be IIII. There was no disputing the King and so the watchmaker took away the dial and had the IIII engraved in place of IV, and that it has thus remained in deference of all tradition.

Mr. A. L. Gordon, of the Seth Thomas Clock Co., has the following to say concerning this story and thus furnishes the only plausible explanation we have ever seen for the continuance of this manifest error in the Roman numeral of the dial:

"That the attempt has been made to use the IV for the fourth hour on clock dials, any one making a study of them may observe. The dials on the Big Ben clock in the tower of the Parliament buildings, London, which may be said to be the most celebrated clock int he world, have the IV mark, and the dial on the Herald building in New York City also has it.

That the IIII mark has come to stay all must admidt and if so there must be a good and sufficient reason. Art writers tell us that pictures must have a balance in the placing and prominence of the several subjects. Most conventional forms are equally balanced about a center line or a central point. Of the latter class the well known trefoil is a common example.

A clock or watch dial with Roman numerals has three points where the numbers are heavier, at the IIII, VIII and XII. Fortunately these heavier numerals come at points equally spaced about the center of the dial and about a center line perpendicular to the dial. Of these heavy numberals the lighter of them comes at the top and it is especially necessary that the other two, which are placed at the opposite points in relation to the center line, should be balanced as nearly as possible. As the VIII is the heavier and cannot be changed, the balancing figure must be made to correspond as nearly as possible, and if marked as IV, it will not do so nearly as effectively as if the usual IIII is used."

Tales Of Symphonia series does not appear to use roman numerals, only an unreleased (or retitled?) game in the series was once planned to use "II".

A table of modern (post-Victorian) Roman numerals has been in the article forever (since 2001). 4000 was originally added as MMMM on 3 October 2005, then changed to MV on 23 December 2006 by an editor who stated that subtractive notation should be used. On 19 June 2007 the original editor changed it back to MMMM. Another editor then added "Not MV" on 6 October 2007. "Clarify" was added to "Not MV" on 29 November 2007 by an editor who noted a conflict with the top of the article. On 12 April 2008 a "pattern" table was added that stopped at MMM. On 15 April 2008 symbols up to IX including MMMM were added to it. On 22 April 2008 an editor changed MMMM in the pattern table to IV, claiming that was the pattern. On 14 May 2008 "Not MV" was "clarified" by stating it was 5000-1000, which actually confirmed that MV was correct. On 15 May 2008 I made MV preferred and MMMM optional.

Now both conflict with the pattern table. Editors who prefer MMMM may be ignoring that the table is for "Modern Roman numerals". But I don't know of citations for any of these forms, just editor's opinions. Should they be removed, as well as all entries above MMM in the pattern table? — Joe Kress (talk) 07:06, 15 May 2008 (UTC)[reply]

"This problem manifested in such questions as why 1990 was not written as MXM instead of the universal usage MCMXC, or why 1999 was not written simply IMM or MIM as opposed to the universal MCMXCIX.

However, these rules are not universally followed."

If the rules "are not universally followed", how can one believe that MCMXCIX was nevertheless "universal"? 10:40, 13 July 2008 (UTC)

Are they? 82.69.77.254 (talk) 14:39, 29 September 2008 (UTC)[reply]

Well, not specifically guitar chord diagrams, but roman numerals are in fact used for chords in general most of the time, especially in Jazz literature. Dgtljunglist (talk) 08:40, 30 November 2008 (UTC)[reply]

On the page, it says uses for VV as 10 were discovered. I know IIX can be used for 8, but have never heard of VV being used for 10.

I'd like to know what source this is from. ZtObOr 23:31, 29 October 2008 (UTC)[reply]

Why is this article called Roman numerals and not Roman numeral? — Reinyday, 16:00, 3 December 2008 (UTC)

I have no idea. te top numeral i can name is ___

                                            MCXMCXI  —Preceding unsigned comment added by 96.240.177.102 (talk) 01:09, 24 December 2008 (UTC)[reply] 

I don't understand why XIX (1 before 10 [which is nine] after 10) is used to represent 19 rather than IXX (1 before twenty). Explanation? ~BRENT NOT MEMBER. :-)