Using of SI prefixes with currencies
Joerg Eisentraeger
Hello all,
the people of Controlling departments of companies etc. here often use
the unit k€ (KiloEuro) for 1000 EUR or k$ for 1000 $.
According to the official SI system: Is it allowed or forbidden to use
the prefix k with currencies? I have not found anything about that on
http://www.bipm.org/en/si/.
In my opinion a currency is not a unit of measurement, therefore the SI
system does not apply to currencies at all, including prefixes. Am I
wrong?
Greetings from Germany
Joerg
--
http://www.joergei.de/
eMail address is valid but not been read.
Jukka K. Korpela
> the people of Controlling departments of companies etc. here often use
> the unit k€ (KiloEuro) for 1000 EUR or k$ for 1000 $.
The disease is probably incurable. Don't make it your problem (unless
they try to force you into it); don't expect them to listen to reason.
> According to the official SI system: Is it allowed or forbidden to use
> the prefix k with currencies?
Neither. The SI system is about presentation of values of physical
quantities, so monetary denotations are outside its scope.
The relevant authorities to be consulted are the authorities for
different human languages and, in some contexts, on notations used in
different areas of life.
> In my opinion a currency is not a unit of measurement,
It is not a unit of measurement as covered by the SI, and this does not
depend on opinions.
> therefore the
> SI system does not apply to currencies at all, including prefixes.
The scope of application of the SI does not cover monetary amounts. But
it would also be outside the scope of the SI to forbid the use of
prefixes for other purposes.
Thus, k€ and k$ are just stupid, and they may violate _other_ standards
and recommendations, but you can't really say that the SI forbids them.
--
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/
Christoph Päper
> the people of Controlling departments of companies etc. here often use
> the unit k¤ (KiloEuro) for 1000 EUR or k$ for 1000 $.
They do, but for some reason this is basically limited to kilo / k,
although M could stand for mega and million, likewise T was tera and
(short scale) trillion. A cent is a c€ or c$, by the way.
International commerce (esp. in English) could profit from using
unambiguous G$, G€ and G£ (milliard / billion) versus T$, T€ and T£
(billion / trillion).
It doesn't look that good with ISO 4217 three-letter currency codes,
though: MUSD, MEUR, MGBP...
> According to the official SI system: Is it allowed or forbidden to use
> the prefix k with currencies?
Neither, it's out of scope. OTOH, if it's not forbidden, it's allowed.
> In my opinion a currency is not a unit of measurement,
Currencies are units of course, they qualify the dimension of the
numbers preceding them (or, in some stupid usage, succeeding them).
Jukka K. Korpela
> Joerg Eisentraeger:
>> the people of Controlling departments of companies etc. here often
>> use the unit k¤ (KiloEuro) for 1000 EUR or k$ for 1000 $.
>
> They do, but for some reason this is basically limited to kilo / k,
Where I live, "M" is often used, too. In fact, it is even officially
allowed in Finnish to use "M€", though only when it is crucial to save
space, and here "M" is interpreted as an abbreviation of "miljoona"
(Finnish for "million"), not as a SI prefix "mega-". It would be quite
inappropriate to _read_ the notation as "megaeuro" (though some people
may actually do so).
> A cent is a c€ or c$, by the way.
It would be hilarious, if I didn't have the nagging feeling that there
are people who are stupid enough to take them seriously and use them.
> International commerce (esp. in English) could profit from using
> unambiguous G$, G€ and G£ (milliard / billion) versus T$, T€ and T£
> (billion / trillion).
When ambiguity is essential, numbers written in digits can and should be
used. Anyone who sells or buys something worth a 1,000,000,000 dollars
should afford to write the sum this way. In a book or essay that needs
to mention such sums often, words can used; "one billion" is quite
unambiguous when you have once defined what you mean by "billion".
It's really newspapers and similar publications that might have a need
for short notations like G$. But even if they were standard, and they
aren't, they would be journalistically poor style
> It doesn't look that good with ISO 4217 three-letter currency codes,
> though: MUSD, MEUR, MGBP...
It doesn't, but such notations are actually used. I've even seen kEUR
and (earlier) kFIM.
Joerg Eisentraeger
Hi Jukka,
thank you for your explanations.
BTW:
>Where I live, "M" is often used, too ...
>"M" is interpreted as an abbreviation of "miljoona"
>(Finnish for "million"), not as a SI prefix "mega-".
The similar we have in Germany. They use not only the k€ (kiloEuro) but
also the T€, where T is the abbreviation for "Tausend" (thousand).
Jukka K. Korpela
> The similar we have in Germany. They use not only the k€ (kiloEuro)
> but also the T€, where T is the abbreviation for "Tausend" (thousand).
That's natural in a sense, and it defeats the idea of using T€ as a
symbol for 1,000,000,000,000 euros, doesn't it? :-) Using SI prefixes
outside their defined scope of use is risky, since letters have many
meanings and you may well clash with some practice that exists
somewhere.
Janis Papanagnou
>
>>Where I live, "M" is often used, too ...
>>"M" is interpreted as an abbreviation of "miljoona"
>>(Finnish for "million"), not as a SI prefix "mega-".
>
> The similar we have in Germany. They use not only the k€ (kiloEuro) but
> also the T€, where T is the abbreviation for "Tausend" (thousand).
I've not yet seen T€ myself in Germany, though I remember former times
(where DM, deutsche Mark, had been the local currency abbreviation, as
opposed to the 3-letter ISO code) where TDM had been occasionally written
for "thousand DM"; I think it was in informal commercial communication.
Joerg, in which contexts did you see that "T€" to appear?
Janis
Joerg Eisentraeger
>Joerg, in which contexts did you see that "T€" to appear?
Janis, I don't remember. Sorry. The German Wikipedia confirms it anyway:
http://de.wikipedia.org/wiki/TEuro
Janis Papanagnou
>>Joerg, in which contexts did you see that "T€" to appear?
>
> Janis, I don't remember. Sorry. The German Wikipedia confirms it anyway:
> http://de.wikipedia.org/wiki/TEuro
Nice quote WRT "TEuro" and "Teuro"! :-)
Janis
Christoph Päper
> The similar we have in Germany. They use not only the k¤ (kiloEuro) but
> also the T¤, where T is the abbreviation for "Tausend" (thousand).
Oh, I thought I had mentioned that. Anyhow, "Tsd." is much more common
in my experience, alongside "Mio." and "Mrd.". I think I've only
regularily seen "bn" in English texts, no language-dependent
abbreviations for thousand and million.
Christoph Päper
> It's really newspapers and similar publications that might have a need
> for short notations like G$.
Exactly, it's fine for headlines and tables.
> But even if they were standard, and they aren't, they would be journalistically poor style
I could live with it. Replacing a phrase like "2 million" with digits
"2,000,000" (or any other local style) really is poor style, because
it suggests a precision that isn't there. In German mass media at
least it's even common to emphasise such values by expressing them as
"a two with six zeros". Another poor style, which in fact seems to be
standard in English for currencies, is <unit><number><multiplier>
instead of <number><multiplier><unit> (in tables and certain
languages, not including English, <multiplier><unit><number> might be
okay, too).
Julian Bradfield
I don't know why you think <unit><number><multiplier> is poor style;
or why Jukka thinks the use of SI prefixes outwith SI is poor style.
In colloquial English speech and writing, the use of, e.g.,
"£57k" is, as you say, pretty standard, and very useful. "k" (pronounced
/kei/, never /kIlO/) is shorter to say than "thousand", shorter to
write than "thousand" or "000", and as you say, it avoids the spurious
precision that is a fundamental problem with simple decimal notation.
In table headings etc. you may see "k£" written, and I myself
sometimes write "k£57" for the sake of variation, but I agree one doesn't
often see it in English. "57k£", which you suggest is good style, is
simply inconsistent with several centuries of established usage in
where to write the currency symbol, and therefore bad style in
English.
More generally, the currency symbol is often omitted when the context
is clear, so one writes "57k" or "4.5M". Although "k" is
always pronounced /kei/ in the pecuniary concept, my impression is
that "M" is always spoken out as "million" rather than /em/ (and
never, ever, "mega"). Presumably because M is seen as a simple
abbreviation of "million", rather than a borrowed SI prefix, even
though its use probably originates by analogy with the SI prefix.
As for the G$ and G£, I find these very useful (in the rare occasions
when I need to talk about such sums), as they avoid the billion
problem (I am one of the few remaining English speakers who refuses to
use the American billion!).
Jukka K. Korpela
> I don't know why you think <unit><number><multiplier> is poor style;
> or why Jukka thinks the use of SI prefixes outwith SI is poor style.
I can answer the latter question. The SI is a well-defined system, with
well-defined special notations. It is generally a sign of ignorance or
over-eagerness to apply the rules of a specific notation outside its
designated scope, including common language and general texts. They
should not be expected to be known to the general audience or to be
intuitively recognized. Being shorthand notations, they are subject to
Korpela's universal law on abbreviations and codes: they are either
misunderstood or not understood. (Think about the T€ case that was
discussed here.)
Moreover, I think there was some discussion in this group some time ago
about the prefixes not being understandable to most people as abstract
prefixes. That is, they don't think in pure SI terms and would be
astonished at hearing a word like "megameter" or notations like "ks" and
"GJ", even though these would often be quite practical. Yet they may
recognize "megawatt" and "s" and "km" and "MJ" and maybe even "GW", if
they've learned them from newspapers.
It would be useful to make people see the idea of SI prefixes more
abstractly and more generally. It would work _against_ the goal to
confuse them into thinking that they can be attached to _anything_,
particularly becase that would ridicule the idea (think about
kilopersons and megadeaths - monstrous expressions that have actually
been used, though fortunately just rarely).
> In colloquial English speech and writing, the use of, e.g.,
> "£57k" is, as you say, pretty standard, and very useful.
There is nothing colloquial, standard, or useful to the rest of the
world, even to English-speaking people.
> it avoids the spurious
> precision that is a fundamental problem with simple decimal notation.
The "spurious precision" problem mainly exists in the minds of people
who think that accuracy can be derived from precision. When I write
£57,000 and you decide that that it indicates an amount exact to ±
£0.50, it's your call, not mine, and you are solely responsible for that
idea.
> In table headings etc. you may see "k£" written,
Rather pointless, since you can write "£1,000". You can also include
",000" into the figures themselves and use just "£" in the header.
Actually, you could also write a descriptive heading like "Expenses" and
put the money symbol into the data items. Saving space is usually much
less relevant than people think. When I see a data item like £47,000, I
can immediately see that it's a sum of money in pounds and I can see the
actual figure. Compare this with "47", which could be just about
anything.
ObMetric-system: Similar considerations apply to indicating physical
quantities in tables. You _may_ use a table heading like "kV" (or "1.000
V") and pure numbers like "47" in the data cells; but it might be wiser
to use a heading like "Max. voltage" with data items like "47 kV".
> and I myself
> sometimes write "k£57" for the sake of variation,
So you think some _additional_ confusion is needed?
> As for the G$ and G£, I find these very useful (in the rare occasions
> when I need to talk about such sums),
Rarity implies that Korpela's law applies particularly strongly. Most
people have no idea of what "G" might mean here, unless you explain it,
and they'll have forgotten it when they see the notation next time.
Julian Bradfield
> Scripsit Julian Bradfield:
>
>> I don't know why you think <unit><number><multiplier> is poor style;
>> or why Jukka thinks the use of SI prefixes outwith SI is poor style.
>
> I can answer the latter question. The SI is a well-defined system,
> with well-defined special notations. It is generally a sign of
> ignorance or over-eagerness to apply the rules of a specific notation
> outside its designated scope, including common language and general
> texts.
This is arrant, and arrogant, nonsense. The expansion of use by
analogy is one of the major ways in which human communicative capacity
is expanded.
Furthermore, the SI is merely a particular narrow codification of a
previously established notation. k and M existed before the SI.
> They should not be expected to be known to the general audience
> or to be intuitively recognized. Being shorthand notations, they are
> subject to Korpela's universal law on abbreviations and codes: they
> are either misunderstood or not understood. (Think about the T€ case
> that was discussed here.)
They may indeed be misunderstood by those people who have never
experienced the social education required to understand them. The same
is true of the SI itself - kN is meaningless to someone who has never
been educated in the SI.
> Moreover, I think there was some discussion in this group some time
> ago about the prefixes not being understandable to most people as
> abstract prefixes. That is, they don't think in pure SI terms and
> would be astonished at hearing a word like "megameter" or notations
> like "ks" and "GJ", even though these would often be quite
> practical. Yet they may recognize "megawatt" and "s" and "km" and "MJ"
> and maybe even "GW", if they've learned them from newspapers.
Indeed. However, these days a very large proportion of the population
in the developed (and increasingly the developing) world is familiar
with the non-SI kB, MB, GB and TB (and those who know that they are
often used with the binary meaning also know that they "should" have
the decimal meaning). From this alone, the generalization to currency
can be made - and indeed, I would bet a large sum that a careful
historical investigation would show that this was exactly the origin
(rather than km, kg etc).
> It would be useful to make people see the idea of SI prefixes more
> abstractly and more generally. It would work _against_ the goal to
> confuse them into thinking that they can be attached to _anything_,
> particularly becase that would ridicule the idea (think about
> kilopersons and megadeaths - monstrous expressions that have actually
> been used, though fortunately just rarely).
kilopersons I admit to not having seen, but megadeaths is not that
rare, or wasn't when I was growing up in the height of the nuclear
stand-off.
But I don't see what's ridiculous about it.
Indeed, it is precisely the abstract and general nature of SI prefixes
that encourages one to apply them to *anything*!
>> In colloquial English speech and writing, the use of, e.g.,
>> "£57k" is, as you say, pretty standard, and very useful.
>
> There is nothing colloquial, standard, or useful to the rest of the
> world, even to English-speaking people.
I'll let the rest of the world decide that, not you.
>> it avoids the spurious
>> precision that is a fundamental problem with simple decimal notation.
>
> The "spurious precision" problem mainly exists in the minds of people
> who think that accuracy can be derived from precision. When I write
> £57,000 and you decide that that it indicates an amount exact to ±
> £0.50, it's your call, not mine, and you are solely responsible for
> that idea.
Arrogant and arrant nonsense again. Even in science (see a discussion
here a few months ago) the conventions about trailing zeros and
precision are poorly applied; in everyday life, they are at best a
strong hint, which needs to be interpreted according to the
context. If you write 57,000 in an entry in an account sheet, then
indeed you had better mean 57,000 +- 0.50, unless you explicitly say
that entries are to the nearest thousand - in which case,
accountants will typically head the columns in thousands, and write 57.
> header. Actually, you could also write a descriptive heading like
> "Expenses" and put the money symbol into the data items. Saving space
> is usually much less relevant than people think. When I see a data
> item like £47,000, I can immediately see that it's a sum of money in
> pounds and I can see the actual figure. Compare this with "47", which
> could be just about anything.
As I've just said, accountants like to head their columns with units
which indicate the precision of the entries.
> ObMetric-system: Similar considerations apply to indicating physical
> quantities in tables. You _may_ use a table heading like "kV" (or
> "1.000 V") and pure numbers like "47" in the data cells; but it might
> be wiser to use a heading like "Max. voltage" with data items like "47
> kV".
What do you think of the practice in medicine of measuring glomerular
filtration rate (and hence heading tables) in units of mL/min/1.73 m^2 ?
>> and I myself
>> sometimes write "k£57" for the sake of variation,
> So you think some _additional_ confusion is needed?
I deal with almost entirely with intelligent, educated, and socialized
people.
>> As for the G$ and G£, I find these very useful (in the rare occasions
>> when I need to talk about such sums),
>
> Rarity implies that Korpela's law applies particularly strongly. Most
> people have no idea of what "G" might mean here, unless you explain
> it, and they'll have forgotten it when they see the notation next time.
I should be astonished if the set of people with whom I might wish to
talk about G£ (which is primarily civil servants and university staff)
included any who does not understand it.
Jukka K. Korpela
>> I can answer the latter question. The SI is a well-defined system,
>> with well-defined special notations. It is generally a sign of
>> ignorance or over-eagerness to apply the rules of a specific notation
>> outside its designated scope, including common language and general
>> texts.
>
> This is arrant, and arrogant, nonsense.
Sounds like I've hit on the nail.
> The expansion of use by
> analogy is one of the major ways in which human communicative capacity
> is expanded.
False analogies and simply wrong use of symbols are a rich source of
confusion and misunderstandings. Besides, this is not a matter of
analogy.
> Furthermore, the SI is merely a particular narrow codification of a
> previously established notation. k and M existed before the SI.
That's simply wrong.
> They may indeed be misunderstood by those people who have never
> experienced the social education required to understand them.
That is, by about 99 % of all people.
> The same
> is true of the SI itself - kN is meaningless to someone who has never
> been educated in the SI.
Quite right. Try asking your next casual acquaintance (outside work or
academia context) whether she or he knows what "kN" means. This gives
you a hint of how many people can be expected to understand, say, T¥ or
G£.
> Indeed, it is precisely the abstract and general nature of SI prefixes
> that encourages one to apply them to *anything*!
They are not abstract, and they are not general except within the SI.
>> The "spurious precision" problem mainly exists in the minds of people
>> who think that accuracy can be derived from precision. When I write
>> £57,000 and you decide that that it indicates an amount exact to ±
>> £0.50, it's your call, not mine, and you are solely responsible for
>> that idea.
>
> Arrogant and arrant nonsense again. Even in science (see a discussion
> here a few months ago) the conventions about trailing zeros and
> precision are poorly applied;
Funnily enough, you accuse me of arrogance and then support my point.
> in everyday life, they are at best a
> strong hint, which needs to be interpreted according to the
> context.
Indeed, except that they aren't even a strong hint. When I write £57,000
or 100 W, I'm not saying anything specific about accuracy. If I agreed
to _pay_ £57,000, then it would be exact, of course. If I just say that
something is worth £57,000, anyone should take it as an estimate of
unspecified accuracy
> As I've just said, accountants like to head their columns with units
> which indicate the precision of the entries.
Accountants do many other things too, out of habit, and often without
ever having thought about them. So do many other people.
> What do you think of the practice in medicine of measuring glomerular
> filtration rate (and hence heading tables) in units of mL/min/1.73
> m^2 ?
Why do you ask? Can't you decide on your own whether that complies with
the SI?
>> So you think some _additional_ confusion is needed?
>
> I deal with almost entirely with intelligent, educated, and socialized
> people.
So you mean they deserve to be confused by your pointless variation of
nonstandard notations?
Julian Bradfield
>> Furthermore, the SI is merely a particular narrow codification of a
>> previously established notation. k and M existed before the SI.
>
> That's simply wrong.
Perhaps a mis-understanding here. I meant that the multiplier prefixes
k and M existed before the SI - not that their use without units did.
(And therefore that the SI has no right of ownership on them.)
In fact, the usage in computers of K without a unit is documented from
1966, so it's not impossible that the usage arose at or around the
same time as the SI (namely 1960).
As a matter of interest, the first OED citation for "k" used in the
form "£3k" is from 1970 - in a job ad for a programmer, which may lend
some support to my contention that the current usage comes from
computing rather than the SI.
>> They may indeed be misunderstood by those people who have never
>> experienced the social education required to understand them.
>
> That is, by about 99 % of all people.
Yes. However, a good rule of communication is to adjust to the
audience - a talk to the general public is entirely different from a
specialist technical talk, and email to, say, academic colleagues can
freely make use of shared knowledge that may not be shared by the
outside world.
> Quite right. Try asking your next casual acquaintance (outside work or
> academia context) whether she or he knows what "kN" means. This gives
> you a hint of how many people can be expected to understand, say, T¥
> or G£.
As I've already said, in a part you chose not to quote, most of the
population is now exposed to k, M, G and even T on a frequent basis.
>> Indeed, it is precisely the abstract and general nature of SI prefixes
>> that encourages one to apply them to *anything*!
>
> They are not abstract, and they are not general except within the SI.
As the OED says:
[From its use as an abbrev. for kilo-.] In connection with Computers K
or k is used to represent 1,000 (or 1,024: see quot. 1970). Also used
transf. to represent 1,000 (pounds, etc.), esp. of salaries offered in
job advertisements.
>>> So you think some _additional_ confusion is needed?
>>
>> I deal with almost entirely with intelligent, educated, and socialized
>> people.
>
> So you mean they deserve to be confused by your pointless variation of
> nonstandard notations?
No, I mean they won't be confused.
Andreas Prilop
> As a matter of interest, the first OED citation for "k" used
> in the form "£3k" is from 1970 - in a job ad for a programmer,
Surely, that means £3072 ? ;-)
--
Top-posting.
What's the most irritating thing on Usenet?
Stan
> Hello all,
>
> the people of Controlling departments of companies etc. here often use
> the unit k¤ (KiloEuro) for 1000 EUR or k$ for 1000 $.
>
> According to the official SI system: Is it allowed or forbidden to use
>
> In my opinion a currency is not a unit of measurement, therefore the SI
> system does not apply to currencies at all, including prefixes. Am I
> wrong?
>
> Greetings from Germany
> Joerg
> eMail address is valid but not been read.
Metric prefixes are borrowed in other applications because they are so
convenient. Another example is the "mil" that was (past tense for UK
may be "is" in US) used in precision engineering to mean a thousandth
of an inch.
Dr J R Stockton
2000hsb.googlegroups.com>, Thu, 10 Jul 2008 02:48:52, Stan
<googl...@philhall.net> posted:
>Metric prefixes are borrowed in other applications because they are so
>convenient. Another example is the "mil" that was (past tense for UK
>may be "is" in US) used in precision engineering to mean a thousandth
>of an inch.
I only recall "thou" being used. But if "mil" was used in Britain it
could have come from sources other than SI - the French, or the USAAF,
language, for example.
--
(c) John Stockton, near London. *@merlyn.demon.co.uk/?.?.Stockton@physics.org
Web <URL:http://www.merlyn.demon.co.uk/> - FAQish topics, acronyms, & links.
Correct <= 4-line sig. separator as above, a line precisely "-- " (SoRFC1036)
Do not Mail News to me. Before a reply, quote with ">" or "> " (SoRFC1036)
Markus Kuhn
|> > the people of Controlling departments of companies etc. here often use
|> >
|> > According to the official SI system: Is it allowed or forbidden to use
|> ://www.bipm.org/en/si/.
Currencies are clearly outside the scope of the SI standard;
ISO 31-0 even explicitely says so in its introduction section.
You can do whatever the reader will understand and find convenient.
I personally see nothing wrong with extending the SI notation
to currencies as well, i.e. currency name or symbol follows
the number with a space and may carry an SI prefix, as in 42 k$
or 36 kEUR.
Markus
me
> In misc.metric-system message <625dc427-69ec-46ee-9bd6-971117a62555@x41g
> 2000hsb.googlegroups.com>, Thu, 10 Jul 2008 02:48:52, Stan
>
> >Metric prefixes are borrowed in other applications because they are so
> >convenient. Another example is the "mil" that was (past tense for UK
> >may be "is" in US) used in precision engineering to mean a thousandth
> >of an inch.
>
> I only recall "thou" being used. But if "mil" was used in Britain it
> could have come from sources other than SI - the French, or the USAAF,
> language, for example.
I've never been able to track it down. When I started 20+ years
ago,
folks would call 0.001" a "mil" and a "thou". It caused no small
amount
of confusion betwix the young and old. The only reasonable
explaination
I ever found was the metric "deca, centi, milli" progression. There
was
one short period where I was lead to believe it was a machinists
expression
having something to do with a "millionth" of some "gage" dimension.
But
I've never found anyone who actually knew. I have been suspicious of
some
military connection, much like the "clicks" we deal with today.
Jukka K. Korpela
> Currencies are clearly outside the scope of the SI standard;
> ISO 31-0 even explicitely says so in its introduction section.
It is clear indeed that the SI or standards defining it and its use do
not apply to expression of amounts of money. Strictly speaking, this is
the end of the story as far as our topic is concerned.
However, this does not exclude the possibility of using SI prefixes in
other contexts, and people have various ideas on this and they discuss
them. The issue also indirectly affects the use of the SI. For example,
if some SI notation became widely used outside the SI in a different
meaning or with a different syntax, confusion would arise, e.g. if
letters were used as currency symbol prefixes in a manner that partly
corresponds to SI prefixes, partly differs from it (as in the use of "T"
for thousand or "m" for million).
So I guess there is still some sense in my re-opening the issue.
> You can do whatever the reader will understand and find convenient.
Though superficially self-evident, this is a dangerous line of thought
in a context like this. "The reader" is usually not an individual that
you know, and even if he is, you cannot have accurate information about
the conventions he actually knows (as opposite to what he has been
taught, for example). Usually "the reader" is an unknown set of people,
some of which may not have been born yet (if we expect the text to be
read even after a long time, as I expect some of your texts will). In
any case, it is typically a very heterogenous collection. As writers, we
usually overestimate our readers' awareness of conventions that we use
(even if we explicitly state them in an article or in a book).
Moreover, there's often someone between you and your reader, such as an
editor or an organization that requires you to follow some standards. In
particular, there might be a national standard that explicitly forbids
the use of SI prefixes for currencies or explicitly defines all the
allowed forms of expressing an amount of money, thereby implicitly but
strictly forbidding (or maybe allowing, who knows?) SI prefixes there.
> I personally see nothing wrong with extending the SI notation
> to currencies as well, i.e. currency name or symbol follows
> the number with a space and may carry an SI prefix, as in 42 k$
> or 36 kEUR.
Your examples do not include one with a currency _name_, and I don't
think you really meant to refer to names as well. I don't think anyone
wants to write "42 kdollar", "36 keuro" or "5 Mpound". Maybe you
implicitly referred to the practice of using SI prefixes proper (such as
"kilo") as opposite to using prefix symbols (such as "k"). Some people
actually speak of kilodollars, kiloeuros, perhaps even megapounds. I
always thought this was some stupid joke that started spreading like a
disease and didn't get any better.
The main practical reason for frowning upon k$, kEUR, M£ etc. is that
letter prefixes are not used consistently across cultures and languages,
and there is no activity for standardization of such issues (and no hope
for it and no need for it, if you ask me). This has already been
mentioned in the discussion, but it apparently needs to be emphasized.
We have variation in the use of SI prefixes for SI units, but at least
we can clearly designate some usage as correct and other usage as
incorrect. For money, it's a mess, and it is extremely easy to avoid the
mess by writing e.g. "$42,000" or "42 000 $" (depending on language)
instead of trying to use purportedly international but actually obscure
notations like "42 k$" (which deviates from the established practice of
putting the currency symbol before the number in the economically
dominant world language, English).
Andreas Prilop
>> Currencies are clearly outside the scope of the SI standard;
>> ISO 31-0 even explicitely says so in its introduction section.
>
> It is clear indeed that the SI or standards defining it and its use do
> not apply to expression of amounts of money. Strictly speaking, this is
> the end of the story as far as our topic is concerned.
A similar question is the combination of units and currencies.
I find $/kg and ¢/min very convenient.
Do you think such quotients are "not allowed" by ISO 31?
Jukka K. Korpela
> On Thu, 24 Jul 2008, Jukka K. Korpela wrote:
>
>>> Currencies are clearly outside the scope of the SI standard;
>>> ISO 31-0 even explicitely says so in its introduction section.
>>
>> It is clear indeed that the SI or standards defining it and its use
>> do not apply to expression of amounts of money. Strictly speaking,
>> this is the end of the story as far as our topic is concerned.
>
> A similar question is the combination of units and currencies.
Similar in what sense? Attaching a prefix to a unit name or symbol is
really different from using unit names or symbols in statements or
expressions.
> I find $/kg and в/min very convenient.
I don't, though I might find the former convenient if I lived in the US,
and in that case, even the latter might become tolerable.
But I find some _similar_ constructs familiar, and I have no difficulty
in understanding that a price annotation like "100 kr/kg" as a shorthand
for "1 kg kostar 100 kronor". It's just currency denotations like kr, $,
and в that I'm not familiar with it. Of course I know what they mean, at
least when the context is known, but I am not used to thinking in terms
of such units.
I can even see $/kg and relatives as denotations for units of relative
price. But I don't see your point. When you form a unit for a non-SI
quantity that is defined in terms of an SI quantity, it is certainly in
accordance with the SI to express the SI quantity, such as mass, in
terms of an SI unit, such as the kilogram. The rest is outside the SI -
and not comparable to the use of SI prefixes for non-SI units.
> Do you think such quotients are "not allowed" by ISO 31?
What makes you ask your question as if I had written that some other
expressions are "not allowed" by ISO 31? The statement of mine that you
quoted says rather the opposite: the standard _does not apply_ to
expression of amounts of money, i.e. such expressions are _outside the
scope_ of the standard. The introductory section's statement, however,
makes it absolutely clear that it is inappropriate and wrong to describe
expressions like k$ as "SI notations".