106

My smallest feature request yet:

I noticed on the Stack Exchange ALL Sites list that SO shows questions at 1.5m.

The standard for million, is a capital 'M', unlike one thousand which is a small 'k'.

'm' in science and engineering stands for one-thousandth.

How about we use the big

M

Improve this question
19
  • 34
    I always read numbers like "1.5mb" as "1.5 millibits".
    – Gabe
    Apr 14, 2011 at 0:26
  • 2
    a lower case m typically refers to minute here...
    – studiohack
    Apr 14, 2011 at 1:10
  • This reeks of the Giant S bug to me ...
    – jcolebrand
    Apr 14, 2011 at 1:45
  • 6
    The standard for million, is a capital 'M', when prefixing SI units no ? MW Megawatts but 1.5 Mil questions/dollars/people or 1.5 Million. Questions aren't SI units. shoot if you are right I need to redo all my thoughts on SI :(
    – phwd
    Apr 14, 2011 at 3:24
  • Related (and/or historical context): meta.stackexchange.com/questions/14581/… and meta.stackexchange.com/questions/43662/… ...
    – Pollyanna
    Apr 14, 2011 at 3:42
  • @phwd, I think we have the ability to take the standard for questions and answers where we want, since Stack Exchange rules that domain :)
    – Lance Roberts
    Apr 14, 2011 at 3:43
  • 5
    So, what you're really saying, is that size does matter...
    – Pollyanna
    Apr 14, 2011 at 4:32
  • 4
    @phwd I'll have to correct you there. The scientific notation is not bound to any unit. You can also say that you want to buy 2k tomatoes and it is still correct.
    – Octavian Helm
    Apr 14, 2011 at 8:26
  • To me that's honestly to make it more pleasing to the eye, has nothing to do with scientific notation (your example would be 2 x 10^3). Cannot fit 2000 in your excel row for your report ? Put 2K. So correct for display in tight spaces... Is it actually readable in a number crunching engineering proposal ? no. I dare someone to put that in a technical reference for engineers people would think it is a typo or some other symbol : conductivity, rate constant, Boltzmann constant, dielectric the list goes on.... oO
    – phwd
    Apr 14, 2011 at 9:01
  • 2
    @phwd I'm not quite sure I got your message but just because someone might not understand it, it doesn't mean that it is alright to have an incorrect representation of it.
    – Octavian Helm
    Apr 14, 2011 at 9:05
  • @Oct The correct scientific notation for 2000 is 2 * 10^3 not 2 K. Engineers (at least Mechanical,Material and Chemical) don't use 2 K. Simple.
    – phwd
    Apr 14, 2011 at 9:13
  • 1
    I read many questions that should only count for 0.001.
    – Rosinante
    Apr 14, 2011 at 12:26
  • 2
    Actually, I think SI is wrong to have both M and m, and there's no reason to accept their poor decisions simply because many people use them in scientific literature and engineering. Viva la Revolución!
    – Pollyanna
    Apr 14, 2011 at 14:12
  • 4
    @Octavian just because someone might not understand it, it doesn't mean that it is alright to have an incorrect representation of it. - actually, it might mean that. If a majority of your audience do not understand it, clearly you need to adjust your idea of "correct".
    – Nicole
    Apr 14, 2011 at 15:59
  • 3
    The SI unit of quantity is the Mole (abbr. mol) which is equal to 6.0221417×10^23 things. If we are going to pedantically use SI units, we should say there are 2.5x10^-18 mol questions or 2.5 atto mol so the correct prefix is 'a'
    – JeremyP
    Apr 23, 2012 at 13:58

7 Answers 7

Reset to default
43

TL;DR

I can't believe that there is no answer yet where people could vote for the capital M and lowercase k. This is one.

+1 for international standards here. -1 if you dislike international standards.

Reasoning

Disclaimer: I'm a non-native English speaker and user of SO and SE sites. And perhaps I can speak for some other users who are more familiar with international standards than English or American standards.

Top voted answer

The top voted answer says k is not for kilo but an abbreviation for thousand - I can't find a reference for that argument. It is not mentioned as a abbreviation on Wikipedia. Ok forget Wikipedia - anyone could add anything. Also Oxford dictionary never mentions k as an abbreviation of thousand, just for kilo.

Also

If someone says "I make a hundred k a year", they are clearly not saying they make 100 kilodollars a year.

Why not? That's 8333 USD per month - not far-fetched and exactly what I understand. What number do you have in mind when you read that sentence?

Even in Germany you can say

Ich verdiene 80 k.

And everyone will understand, but not because k is the abbreviation of thousand but k is the abbreviation of kilo and kilo means thousand.

Second answer

At first, I thought this is my answer, but then

[...] users who are unfamiliar with scientific notation [...]

So, who's that then? In Germany you learn that at school. And as already said, for English speakers, the character is the same, it's just the casing.

Other answers

We cannot have fractions of votes/rep/views/etc

That's true but not a real argument

29.7m is easier to read than 29.7M

According to which design principle or typographic principle please? I'm willing to discuss this on GD.SE

Let's just use E notation...

Well, that's the difference between international and scientific. I don't want to need a master's degree to visit an SE site.

Improve this answer
9
  • 8
    Your link to the Oxford dictionary does mention it. You just didn't look far enough down the page. So you did find a reference for that argument, actually.
    – Dan Getz
    Jun 4, 2015 at 5:04
  • 3
    And here's the OED stating that "m" is an abbreviation for "million": oxforddictionaries.com/definition/english/m#m__9
    – daiscog
    Jun 4, 2015 at 9:35
  • Thanks. I was looking for k being an abbreviation of thousand, not for m being a million. But that doesn't matter, it is also a nice finding and a good reason to downvote this answer.
    – Thomas Weller
    Jun 4, 2015 at 14:34
  • 6
    You completely misinterpreted my comment: If someone says "I make a hundred k a year", they are clearly not saying they make 100 kilodollars a year.. I wasn't saying that 100,000 was an unreasonable amount for someone to make. I'm saying that you read as: "I make one hundred thousand dollars a year" not "I make one hundred kilodollars a year". It was providing evidence that k can mean thousand rather than kilo
    – Kip
    Jun 4, 2015 at 15:59
  • 7
    But 100 kilodollars and 100 thousand dollars are the same thing!
    – Luigi
    Jun 4, 2015 at 18:53
  • 5
    Tl;dr: "Everyone here is an engineer and if they aren't they should be."
    – Nicole
    Jun 8, 2015 at 1:16
  • @NickC: here? This is Meta.SE, the meta site of all sites. I'm not so sure if users of Christianity, Physical fitness, Poker, Pets or Islam are engineers... So certainly the E notation does not fit here.
    – Thomas Weller
    Jun 9, 2015 at 18:11
  • 1
    @ThomasWeller stackexchange.com/sites#traffic
    – Pollyanna
    Oct 21, 2015 at 19:45
  • If anyone is not familiar with scientific notation they should just go learn it. That's not a good excuse.
    – dyslexicanaboko
    Aug 8, 2022 at 13:59
44

Sanity check: If the number got into the billions, would you really expect to see "1.5G"?

You seem to be under the impression that the "k" in "1.5k" represents the SI unit "kilo". It doesn't. It is an abbreviation for "thousand" in English. It was borrowed from the SI unit, in the same way that "ski" was borrowed from Norse. But it has now has its own meaning separate from the SI unit. If someone says "I make a hundred k a year", they are clearly not saying they make 100 kilodollars a year.

Likewise, the "m" in "1.5m" doesn't mean the SI unit "mega". It is an abbreviation for "million". If that number got to billions, we'd see "1.5b", not "1.5g".

Improve this answer
6
  • 16
    Of course not, for billions I would expect T - tera.
    – Danubian Sailor
    Jan 3, 2014 at 17:52
  • @ŁukaszL. I'm not sure if you're serious or trolling... most English-speaking countries use short scale (1,000,000,000 = 1 billion; 1,000,000,000,000 = 1 trillion), whereas most other European countries use long scale (1,000,000,000 = 1,000 million; 1,000,000,000,000 = 1 billion). I was using short scale. en.wikipedia.org/wiki/Long_and_short_scales
    – Kip
    Jan 6, 2014 at 13:57
  • 23
    I'm pointing out that misunderstanding with common names. Therefore, I prefer SI wherever applicable.
    – Danubian Sailor
    Jan 7, 2014 at 10:33
  • 2
    Please add an official link for abbreviations of English numbers - I can't find one. It is not mentioned as a abbreviation on Wikipedia. Ok forget Wikipedia - anyone could add anything now. How about Oxford dictionary never mentions k as an abbreviation of thousand, just for kilo.
    – Thomas Weller
    Jun 3, 2015 at 19:04
  • 2
    @ThomasWeller Your link to the Oxford dictionary does mention it. You just didn't look far enough down the page. It says it's written either capital or lowercase.
    – Dan Getz
    Jun 4, 2015 at 5:01
  • 4
    You seem to be under the impression that the "k" in "1.5k" represents the SI unit "kilo". because that's what it is
    – njzk2
    Jun 4, 2015 at 15:53
24

So you change it to this:

enter image description here

Which leaves users who are unfamiliar with scientific notation confused about the inconsistency. So then you also uppercase the K:

enter image description here

Which means you have to uppercase the K on the user signatures:

enter image description here

Which brings hundreds of scientific users here to tell you that K means Kelvin and Kilo- is k.

Which leaves you wondering:

Did anyone really think we only had 0.0015 questions in the first place?

Edit, response to comment —

This is not nonsense. It's the reality of dealing with a situation where your "accuracy" is in the eyes of your users. StackExchange has many non-scientific users, and is not itself a scientific system. So whichever you choose, you will be "wrong" to some portion of your users:

  1. As it is now — Some, like Lance, will think its silly that you didn't follow scientific convention, but will know what was meant anyway.
  2. If the M is capitalized — Many will think it is a typo. Some will think its unprofessional, some will be put off by it, some won't notice or care, and some will come here to complain.

Neither is a complete win. So SE has to decide which of the two scenarios is worse, and pick the other. I vote for visual style.

Improve this answer
7
  • 9
    Yes, that should be a lowercase k.
    – Gabe
    Apr 14, 2011 at 3:27
  • 1
    I think a lot of our users will be familiar with scientific notation, but then again it only takes one to make a post on a problem.
    – Lance Roberts
    Apr 14, 2011 at 3:42
  • 29
    I don't agree. Visual consistency is not more important than factual accuracy. It has yet another implication which one might not think about initially. This network is about sharing knowledge. Users which didn't learn the correct scientific notation will learn it here and they will think that it is correct the way it is because, hey Stack Exchange did so it must be correct.
    – Octavian Helm
    Apr 14, 2011 at 8:21
  • 1
    @Octavian - let's be clear - users come with their own preconceived notions of "accuracy" and without anything explaining scientific notation, I don't share your optimism that they will magically learn it simply because SE's capitalization is unexpected.
    – Nicole
    Apr 14, 2011 at 15:53
  • 8
    @Fosco - huh? Did you read the whole thing?
    – Nicole
    Apr 14, 2011 at 16:42
  • 12
    Personally, if it was changed to a capital K, I might try for 0 rep just so I can say I reached absolute 0.
    – ughoavgfhw
    Apr 14, 2011 at 22:03
  • @ughoavgfhw The system has never allowed users to have 0 rep.
    – wizzwizz4
    Jul 19, 2021 at 22:23
11

My 20 millidollars on this say the world could do with more Mega, Giga and Tera.

Guys, Stack Exchange's got two megavisits/day without needing gigadollars of VC funding!

...especially if you know how to pronounce mega, giga, etc. awesomely.

Guys, Stack Exchange's got two me-ga!visits/day without needing gi-ga!dollars of VC funding!

I'm sure it sounds te-ra!weird now, but after a few ki-lo!times I'm sure it'll feel just normal!

Improve this answer
7

Let us stick to SI notations. Let us use capital M for mega and small k for kilo.

People not familiar with SI notations should learn it ASAP.

I would have voted for Thomas Weller but he spoke against the E notation so let me write my own here with the last clause removed.

Improve this answer
2
  • 1
    As has been stated 4 years ago M does not represent mega, but million. Similarly k does not represent kilo but thousand. Rather just agreeing with the question, can you address why the other answers are incorrect? There's no logical argument here why anyone should use your suggestion.
    – ben is uǝq backwards
    Jun 4, 2015 at 11:48
  • "E notation" is only sensible to people who speak Fortran as their native language.
    – The Photon
    Jan 20, 2020 at 22:06
0

The answer is quite straightforward:

The "m" used in this context is not a prefix of the unit; it is a suffix of the number.

We are not dealing with SI units and prefixes here. As already noted, in written English, "m" is an abbreviation for "million". For example, the population of Wales is roughly 3m people. In this usage, the "m" is not a unit prefix, but together with the number before it, it means "3 million people" not "3 megapeople".

This answer correctly points out that similarly the abbreviation for billion is not "G". Example: "There are over 7bn people in the world" would never be written as "there are over 7G people in the world".

__

OED definition of 'm'

OED definition of 'k'

Improve this answer
6
  • 1
    As usual, English rules are inconsistent: "m" for millions, but then why not "t" for thousands. (Just a non native trolling point of view)
    – Pablo Lozano
    Jun 4, 2015 at 10:52
  • @Pablo They did start off with t and I, and many others, complained. So they changed it to k. The relevant questions have been deleted (why? I don't know.) but here's a comment string from when the change was first made and moved from t --> k meta.stackexchange.com/questions/14581/…
    – Pollyanna
    Jun 4, 2015 at 11:40
  • I know, the inconsistency is in the language itself, not in SE
    – Pablo Lozano
    Jun 4, 2015 at 11:50
  • @Pablo That's because of how English has evolved as a language through common usage. Linguistics is a terrible thing.
    – daiscog
    Jun 4, 2015 at 11:59
  • @daiscog Of course, that's why I wrote that my comment was a little "trolling". No natural language is free of these inconsistencies, but they are always more notable for a non native speaker. I was really shocked when I read that "k is for thousand, not for kilo", because I was convinced it was the opposite
    – Pablo Lozano
    Jun 4, 2015 at 12:56
  • 1
    @Pablo The etymology of the use of k for thousands is a derivation from its use for kilo, so there is a sort of logic behind it, I guess. To be honest, even as a native English speaker the language still occasionally surprises me. It's such a mongrel language! (#digression)
    – daiscog
    Jun 4, 2015 at 13:21
-4

We cannot have fractions of votes/rep/views/etc, therefore it is sufficient to make a design decision based on the visual aspects and lowercase the exponent suffix, as all suffixes will be above 0.

29.7m is easier to read than 29.7M since the suffix is clearly not part of the number, but it's still very easy to understand the magnitude with the lower case letter.

I don't think it's appropriate for Stack Exchange to attempt to follow the SI system so closely that case matters. It is sufficient to loosely follow it, and allow people to hover their mouse over the number for the exact count, or click it and get the exact amount, or learn through trial and error what the suffixes mean.

Improve this answer
6
  • 2
    But if you ask me again in a month or three I'll probably argue in favor of this request. I'm rather fickle about such things.
    – Pollyanna
    Apr 14, 2011 at 3:50
  • 1
    I agree that it's all about the visual aspects, which is why I like the M, since it always throws me off when I see the smaller m; like I'm editing one of my labs/reports.
    – Lance Roberts
    Apr 14, 2011 at 4:02
  • 12
    Since everyone will encounter the SI system every day in his life it would be very appropriate to follow the correct SI prefixes. I'm quite shocked actually that people tend to prefer visual elegance over factual correctness.
    – Octavian Helm
    Apr 14, 2011 at 9:35
  • 3
    @Octavian Standards are great! There are so many to choose from! There is nothing incorrect about the current system - factual or otherwise. Just because the SI system exists doesn't mean that it's the One True Way to represent exponential values. And yes - visual design is important. This isn't a scientific publication - it's a website where it's more important to provide a nice design to attract users than it is to follow some arbitrary scientific rigor. Just be glad we're not using "T" for thousand anymore, and move on.
    – Pollyanna
    Apr 14, 2011 at 12:36
  • 3
    So the content of the website doesn't matter if it is nice to look at. Interesting, very interesting. Apparently Stack Exchange chose the SI standard by using 'k' instead of 'T' for thousands. If one decides to go with a standard then he should also go all the way and implement it correctly.
    – Octavian Helm
    Apr 14, 2011 at 12:45
  • 3
    @Octavian you're not arguing about content, you're arguing about style, and being exceedingly and irredeemably anal in the process. By the way, you must use tabs. Spaces are devilspawn. And crack your eggs big-end first. While you're at it, check an English manual of style or three, then come back and explain how descriptive text must always follow one internationally-defined standard.
    – Nicholas Knight
    Apr 14, 2011 at 13:08

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .