Style guide revisited

The Measurements units style guide, published by UKMA in 2012, is now available on line with an index. In this article, we outline the purpose of the Guide and note recent enhancements.

There has long been a divide in British education after the age of 16 between arts and science. Those who can write, often can’t do sums or measure. And those who can use numbers to solve problems are often unable to write about their achievements. Add this to the ‘make it up as you go along’ nature of old measures and the stage is set for confusion and muddle when writers come to put metric measurement units into their work.

This may have been in the minds of the authors of the UKMA Style Guide. Its purpose, when first published, was to encourage the correct usage of metric measurement units.  It was printed on card and distributed to national and local newspapers throughout the UK, with mixed results. It was also made available for download from the UKMA web site.

Recently, an interactive index has been added for those wishing to search for particular letters, words or topics. It remains possible to download the original Guide or to order a printed copy. Further information may be found on the UKMA web site:

So, even if you prefer arts to science and writing to measuring and calculating, you have no excuse for misunderstandings when it comes to using metric measures. Follow the Style Guide and your meaning should be clear and there could be less risk of confusion.

21 thoughts on “Style guide revisited”

  1. Very nice. I have minor quibbles with two points:

    *The bar is a discouraged unit from Table 8 of the SI Brochure. If used, its SI definition (100 kPa) should be given on first use. I recommend omitting the bar or adding its definition in parentheses with a cautionary note.

    *The space between number and unit is mandatory. Section 5.3.3 of the SI Brochure states, “. . . and a space is always used to separate the unit from the number.” Omitting this space seems to be a UK national practice, and doing so has zero basis in the SI Brochure. Writing 5m to save space is no more correct than 5metres to save space. I recommend striking the parenthetical phrase in its entirety.

    Finally, you might consider a link to directly downloading a free pdf English-language version of the SI Brochure. However, this may change soon to edition 9 if the discussed changes to the SI are approved.

    Click to access si_brochure_8_en.pdf


  2. The letter “x” to represent multiplication could be nicely replaced by the real multiplication × sign.


  3. I looked at the guide and I don’t think it makes it clear enough how degree Celsius should be used. If I wanted to describe a walk of 4100 metres I could quite properly describe it as “a walk of 4.1 kilometres” or “a walk of 4.1 km”. If I want to describe the boiling point of uranium in degrees Celsius, can I write it as “4.131 kilodegrees Celsius” and “4.131 k°C”?

    And on the subject of Celsius, when I copied and pasted the degrees Celsius symbol from your guide (because the degree symbol doesn’t appear on my keyboard) it pasted as “oC” – which is clearly wrong!


  4. 1. John’s comment above: The use of a space is mandatory; I realise it’ll be many
    decades before this is ever likely to be adopted by most people, especially publishers.
    However a start should be made to get all teachers to do it correctly; not just science teachers. It is also something that examination boards should insist on.

    I haven’t checked HM Government publications, Hansard, and many Local Authority Reports, websites etc. – I wonder if they include the space or simply save space like most publishers.
    2. Marauder’s comment: a nice idea. I now know there is a difference. Sadly the proper symbol for multiplication is much more difficult to find when using a computer/tablet/smartphone. The ‘x’ between Z and C on the keyboard is obvious.
    When it comes to handwriting, I’m sure most people couldn’t tell the difference; of course, the sentence/context would make it clear that it is a cross, a letter x , or a multiplication sign. Sadly an issue about using lower case not upper case is ignored etc.


  5. The problem with the mandatory space is that if you don’t have a non-breaking space available (eg plain text email, and indeed this Comment box), it can result in the units being separated from the number and transferred to the next line, making for very poor readability. Banged my head against that one many times, which is why I tend to ignore this rule.


  6. @ Charlie P

    Since the boiling point of uranium is 4404 K, that can be written as 4.404 kK.


  7. In the section on Speed Limits, the symbol for speed should be separated from the symbol for hour by a division sign (that does not extend above or below the height of the symbols) rather than a normal forward slash.


  8. @Charlie P 2018-08-28 at 11:59
    “And on the subject of Celsius, when I copied and pasted the degrees Celsius symbol from your guide (because the degree symbol doesn’t appear on my keyboard) it pasted as “oC” – which is clearly wrong!”

    Correct. This is another muddle the world has got itself into. It is more a computer translation problem, but comes up often with trying to get SI to translate across platforms.

    Here are some I have tested out, most come from Libraoffice Greek character map, checked with general internet usage. They C&P into text editors, word processors, spreadsheets and work over the internet into Facebook. I do not use Windows so not tested, Microsoft tend to do their own thing. One that I cannot get to copy is the non-breaking space. These are in symbol, comma, space format
    [°, ±, ², ³, µ, ?, ? ?, ?, ?, ?, ?, ?, ?, ?, ?, ?] In SI usage we get °C, km², m³, µm, which seem to be the troublesome ones.

    The easiest way to get these correct is to have a set C&P into various places. I hope others can add and refine this list.


  9. Unfortunately most (including pi) do not work through this site! Now that is interesting. It proves the point it needs to be sorted.


  10. @BrianAC

    I don’t know until I try whether this will be an answer or a failed attempt. WordPress accepts html codes for special characters according to this article:

    The code for a special entity must be preceded by an ampersand and followed with a semicolon. The special entity may be identified by its name, decimal unicode address preceded by #, or hexadecimal address preceded by #x, no spaces are permitted, but adding space is a way to show the code without it encoding. Support may also depend on browser and operating system.

    Two characters useful in SI discussions are pi and the Ohm symbol. The html name, decimal address, hex address, followed by each as complete html are:
    pi, #0960, #x3c0. π π π
    Omega, #0937, #x3a9; Ω Ω &#x3a9
    After this is moderated, we will both see which, if any, work here.

    For Windows users with numeric keyboards, 4 digit alt-codes work for special characters up to 0255, because I regularly use several. I believe they do NOT work in WordPress for higher Unicode address even when the Alt-x mod has been enabled in Windows registry, but I will try pi, ?, and Omega, ?.


  11. @John Steele:

    Not just UK, it is rife throughout Europe and beyond. Anywhere you find product package designers, they seem to regard it as wasted space which could be filled with lovely branding instead. Especially annoying on food nutrition labels, which have two or three columns of values with units, sometimes with ridiculously large inter-column gaps just sitting there doing nothing!

    @Charlie P:

    Yes, the longer BIPM brochure does not say otherwise—unless someone can point out where? The boiling point of metals is probably a bit too niche for inclusion in UKMA’s handy guide, though. Possibly NIST in the USA does not like this, but it serves them right for clinging onto °F—where I, with as much authority as they, hereby categorically forbid them from doing the same with Fahrenheit. Even though Celsius was a real geezer and units named after people are usually not capitalised unlike symbols, ‘degrees Celsius’ was the exception adopted in 1948. With a PC keyboard in X Windows; Alt+0 gives a ‘°’ character.


    Also, get teachers to distinguish between ordinal indicators and the degree symbol—i.e. not e.g. ‘37 ºC’ 😒… On a PC keyboard in X Windows; Shift+Alt+W produces a ‘×’ character. To access the appropriate virtual keyboard on a smart-‘phone or tablet, you probably have to press and hold one of the buttons while saying the word ‘metric’ three times into the microphone or similar 🗣‌📱‍🖖.


    In WordPress comment boxes, you can type HTML character entities such as &nbsp; for non-breaking space and for [Unicode] non-breaking thin-space. With luck, the latter shall be displayed correctly by most browsers using modern fonts. For the former, again with a PC keyboard and X Windows; Alt+<Space> emits the [8-bit] character itself, which passes through all but the most ancient e-mail software and communication links without re-encoding. Other operating systems may vary, but there is usually a way!


    Wow, that is IMO unnecessarily picky/ overly exacting. The Transport alphabet by Kinneir and Calvert for UK highway signage does now have two solidus (solidi?) albeit neither originally intended for precisely these use-cases. Their Rail alphabet for railway signage did not officially have any, although an all-purpose one was added specifically for data plates including locomotive sub-classes and [pre-ISO 8601] DD/MM/YY dates. Both have no division sign ‘÷’ or fraction slash ‘⁄‍ ‍’ to this day!

    You might expect this to be a dead issue, but BBC have recently instated a new I-can’t-believe-it’s-not-Gill-Sans corporate typeface. The boast that it has Cyrillic (¡) but not Greek character sets immediately rings alarm bells that it won’t have ‘μ’ or ‘µ’ and ‘Ω’ et al 😠. I’ll be mollified by them just writing metric speeds with whatever slashes they have in there and eschewing imperial…


    What were the problematic ones other than ‘π’?


  12. On the point of symbols, I have decoded those that do appear on this site, and they are the same Unicode that I am posting. From this I deduce it is my own browser/OS/settings combination (Firefox on Linux Ubuntu).
    The strange thing is it works on the UKMA Facebook page until I change to Windows for the same test, then some errors come up, the reverse of here it seems!
    The other symbols I posted are not really relevant to SI, just for testing, but Theta, Psi (NOT PSI!), Omega and “Unicode 119330 (1D222 hex)used for frequency”, those I remembered from the 1960’s and never used since. Just for punishment I will post some again as I have just updated everything. [(176)°,(177)±,(178)²,(179)³,(181)µ,(215)×,(981)?, (996)?,(936)?,(937)?,(960)?,(7517)?,(7520)?,(7526)?,(7615)?, (119330)?]. The decimal Unicode I have decoded from this post, so that is what I am sending, no doubt.
    I think that is enough from me on this.


  13. @Charlie P 2018-08-28 at 11:59 says: –
    “If I want to describe the boiling point of uranium in degrees Celsius, can I write it as “4.131 kilodegrees Celsius” and “4.131 k°C”?”

    I agree with Daniel on this one. Once we get above cooking temperatures (around 250° C on my oven) we should be thinking about using Kelvin.
    However, the point is still very valid about kilodegrees, other than using four digits? In this instance I would say 4.131 k°C is the one that is, shall we say least bad?


  14. John Steele says: 2018-09-03 at 12:15 @BrianAC
    “For Windows users with numeric keyboards, 4 digit alt-codes work for special characters up to 0255, because I regularly use several. I believe they do NOT work in WordPress for higher Unicode address even when the Alt-x mod has been enabled in Windows registry, but I will try pi, ?, and Omega, ?.”

    That just about sums it up John, that is where I had got to before I re-read and fully understood the implications on your post. Anything over 255 (FF hex) seems will not work for me. Another case of ‘world standard’ (or at least a world dominant) organization(s) screwing up those that have the audacity to use world standard software (especially when the world standard is European!!). Before anyone comments, FOSS (free open source software) is fully ISO compliant, it has to be. The UK government pay millions to support Libraoffice as of September 2015 and has its very own version [various sites: – It is the big boys that play by their own rules that screw everything up .
    That said, this site is resolving Unicode above 255 (FF) through some ‘backdoor’ trick that the rest of us cannot follow.


  15. @BrianAC

    More a case of “multistandards.” Some sites completely disable html for posts, some allow html characters only, some let you use Unicode directly, some use a mix. It’s a mess. It would be helpful if sites, especially those at all technical and needing special characters, would post a “special character policy” somewhere in their help. Now that I know html special characters work, I know what to do.

    Since the US is so “dual” on its measurement policy, I am pretty good at knowing two or more ways to do everything. 🙂

    Here it seems the “policy” is that Windows alt-codes only work to Alt-0255, above that, or no numeric keypad, use html special characters. I believe I may have omitted a closing semicolon in my earlier test. Retesting Ohm, Ω


  16. I’m convinced that the Daily Telegraph’s style guide (if such a document exists) lists kilometres per hour as “kph” rather than km/h (or if you see yourself as a physicist). So my castigation of this newspaper’s so-called professional journalists may have been a tad unjustified as an editor may have “corrected” their copy.
    Somewhat related, I would welcome a ruling on the use of pound weight; the Daily Mail persists on using “lbs” as the abbreviation, while I maintain that, as example 50 pounds should be abbreviated as 50lb rather than 50lbs. Please advise.


  17. @BrianAC:

    The most reliable way to put non-ASCII characters into WordPress seems to be HTML character entities. I guess it is written by anglophones. Probably a nuisance to submit comments in non-English languages too, even including those using a Latin alphabet with diacritics. Metric far easier to mark up than imperial, naturally. In the comment box, type one of the incantations below starting with an ampersand and ending with a semi-colon, in place of your strange character… What could be simpler?

    ‘ ’←&nbsp; <non-breaking space>
    ‘ ’←&#8239; <non-breaking thin-space> (more correct than &zwj;&thinsp;&zwj;?)

    ‘²’←&#178; (not &sup2;)
    ‘³’←&#179; (not &sup3;)

    ‘Ω’←&Omega; (upper-case ‘O’)

    ‘💩’↔ℓb/ ℔s

    •The non-breaking thin-space is decimal Unicode—for hexadecimal directly precede the number with a [lower-case] ‘x’.
    •The names are case-specific. A lower case &omega; looks like ‘ω’. The same pattern applies to the other Greek letters.

    Could one of the Metric Views editors put a link drawing attention to this comment section on the House rules for comments page? Coding up the decimal codepoint entities was a bit exciting—Wordpress apparently double [de-]canonicalises them, at least…


    You can fairly assume that all British newspapers and media generally have an anti-metric agenda. This is one of their less repellent ideological traits, IMO. Hopefully, the consumers will wise up to this eventually. For imperial units, you can either use the symbols above or make up your own different ‘abbreviations’ every time—there is no authoritative body who can give a definitive ‘ruling’ either way. Better yet; avoid using them altogether to save yourself the angst!


  18. One has to wonder what the BBC is up to with their style guide.

    When I first saw this article yesterday about a kangaroo attack in Australia:

    there was only Imperial units. Surely the Aussies would have reported in metric only, no?

    Now today the BBC has added metric in parentheses, which is at least a half step forward. I’d be curious to know what the process was at the BBC to go from metric (as reported by Australia, I presume) to pure Imperial and then a slight twist to metric in parentheses).

    So much better if they just stuck to metric pure and simple!


  19. @ Ezra Steinberg: 2018-10-14 at 20:33
    Does the BBC have a style guide these days? Maybe that is un-PC and we have a free for all.

    I record and follow the three main TV news channels. The variation from lunchtime to evening on all channels is quite strange.
    From America we expect Imperial, yet on BBC during the recent storm I heard for the first time ever I think, the wind speed as 225 km/h and that from a CBN news reporter. Also of the storm surge of 4 m. Yet that same channel reported similar from Spain all in Imperial. No logic there then.
    I am fairly sure there is a deliberate trait of not revealing the fact that the rest of the world is metric (in particular Australia), America is not metric so that’s OK!
    More interesting was the Russian space mission failure. The part that made most sense was the screenshot from inside the control centre, all in Russian but so easy to follow. Data in SI, just c (for cek) instead of s for sec. Not exactly rocket science as the saying goes.


  20. Ezra:

    Re the kangaroo attack in Queensland, Australia, I have just had a look at some news reports in Australia as well as (in English) in Japan and the Arabian Gulf. I found the height of the kangaroo reported variously as “6-foot”, “1.8 m”, “1.83 m” and “2 m” – and those are figures I found on just a 5-minute search. Mainly metric, but the odd imperial thrown in, even in Australia. (So how big was it really?) The truth is that the height of the animal is of no consequence. Nobody took a tape measure to it. If it had been a smaller animal and the chap had sustained such horrific injuries, the height of the animal would have made no difference. Let’s hope the injured man makes a full recovery.


  21. In response to Ezra’s comment I wonder if the original metric numbers and the metric added in parentheses actually match. I’ve seen a lot over the years of rounded metric vales converted to soft USC/imperial values, then rounded and back converted to metric values that don’t match the original.

    1 m —> 39.37 in –> 36 in –> 0.914 m

    Maybe not to this extreme but this was done to point out this debase practice.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: