rob waller

Tuesday, January 30, 2007

Octothorpe (you know what I mean)

A colleague’s been debating with a mobile phone company client how to refer to an old-fashioned steam telephone. They want to choose from ‘fixed line’ and ‘land line’, with a preference for the former. She asked for votes from users and ‘land line’ won by a mile. Needless to say, the client wants ‘fixed line’. Actually, a lot of people said ‘none of the above’ and generally refer to ‘home phone’ instead.

This reminds me of a debate we had with BT a long time back, over what to call the # key when it needs to be spoken out loud in voice menus. The international telecommunications standard specifies ‘square’ and BT insist on using that to this day. The debate even went to their usability lab in Martlesham, who backed 'square' ... hmm. We believe 'square' could be there because the standard was translated from another language and the term with it. We went out in the street with a phone and as you would expect, no one called that key ‘square’. A lot didn’t know what to call it, but if they had a name it was generally called ‘hash’. Musicians called it the sharp sign, and someone with computing training called it ‘gate’. In the US it is more generally called the ‘pound sign’ or ‘number sign’. There is an good Wikipedia entry with further names under the headword Number sign.

The most bizarre name for the # sign is Octothorpe. This apparently first appeared in the 60s or 70s, but there is disagreement about its origin. In Elements of Typographic Style (p. 282), Robert Bringhurst says that ‘in cartography, it is also a symbol for village: eight fields around a central square, and this is the source of its name. Octothorp means eight fields.’

Really? Type ‘octothorpe’ and ‘cartography’ into Google and all you get is dictionary definitions quoting Robert Bringhurst – nothing from a cartography source.

Another explanation is given by a retired AT&T engineer, Ralph Carlsen, that he and a colleague made it up, when they needed a word for the # key when developing touchtone phones.

I turn out not be the only person puzzled by this bizarre word. Here are just a couple of the various websites chasing its origin:
http://www.worldwidewords.org/weirdwords/ww-oct1.htm
http://www.robertfulford.com/2005-06-14-octothorpe.html

Octothorpe is a truly bizarre phenomenon – a word that is never ever used for its notional meaning (ie, to refer to a telephone key), that has no real purpose or nuance to add, but that is in the OED (citing the daft ‘fields’ origin as a possibility), and whose origin is much discussed. At least one web dictionary I found claimed that there are variant spellings such as 'Octotherp', as if linguists had toured the country asking gnarled old telephone engineers for the terms that they and their forefathers had used for generations.

So my plan is to invent a new word for something, and get it in the OED. Any ideas?

More seriously, when you research something like this, it reveals the true limitations of the web – with its apparently authoritative websites packed with cut and paste repetitions of unsubstantiated information.

Thursday, January 18, 2007

Ranged left vs ragged right

My heading looks like a mistake ('ranged left' and 'ragged right' are synonyms). But although they refer to the same thing, I think they mean something different.

Whether or not justified or unjustified type works best is a perennial question. The term refers to the practice of padding out lines with extra space to achieve a straight right-hand edge.

It’s less of an issue that it used to be – owing largely to thirty years or so of modernist-influenced design education, and the displacement of compositors (printers who used to set metal type) by graphic designers, unjustified type is the default option these days. Newspapers and printed books are the only genres where justified type is still more or less compulsory – at least for certain types of content. Textbooks are more likely to be unjustified than novels, and newspapers use justification as a marker of formality – The Times uses justified type for news stories and the main editorial, but unjustified type for features and commentary.

Defenders of justified type have not found much support in reading research – psychologists looking for an effect on reading speed or accuracy have found it makes little difference. But typography isn’t just about making reading easy, or providing an efficient channel for words to pass from page to brain – it’s also about articulating meaning. It is part of the language resource that we use to communicate relationships between parts of a text. In particular, it is what makes written language more than just a secondary form, a transcription of speech. Looked at in this way, is there a role for justification?

I think there is a clue in the names we use. Unjustified type is often known under other names, ‘ranged left’ and ‘ragged right’ being the most common. The names didn’t just happen, but were a form of spin by modernist designers who were uncomfortable that the term ‘unjustified’ seemed to imply a lack of quality. In linguistic terms, ‘unjustified’ is the marked form that implies it is the exception not the rule, whereas ‘ranged left’ is a simple neutral description of the way the type is set. The very term ‘justified’ implies something correct, and properly finished, and in an age of symmetry and ornament anything else would have been seen as unfinished – like unplaned timber, or a garment with no hem.

As descriptive terms, ‘ranged left’ and ‘ragged right’ do quite different jobs, and the difference is instructive. ‘Ranged left’ refers to what is going on inside the column of type – the letters are ranged evenly from the left-hand margin – and makes no reference to the space outside the column. ‘Ragged right’, though, draws attention to the resulting untidy column edge.

In the ‘ranged left’ world view, then, the even texture of type in the column is the most important thing. The column of type is where reading happens, in isolation from other elements on the page.

‘Ragged right’, on the other hand, focuses us on the column edge and, in my view, the term implies a case for justified type.

Edges are critical in graphic design. Pages are made up of different elements, usually aligned in deliberate ways to contribute to the reading experience, whether through meaning, navigation or a simple sense of visual order. The space to the right of a column of type may have a job to do – a job of separating elements, of framing the text, of linking though the continuity of white space. This job can often be done better by a well defined rectangle.

For example, in multi-column layouts, ragged right is fine, so long as the lefthand columns are straight – too much indented type (of the kind you get with legal text) creates a ragged left effect, and the intercolumn space is a less effective visual element.

Here’s a simple demonstration of what I mean.


1: Ranged left is fine so long as you have a straight lefthand column to define the column (most of the time, in fact).


2: Where you have frequent multilevel indention in multicolumn text, it effectively creates a ragged left edge and the intercolumn space no longer forms a clear shape.


3: Justified type restores the vertical alignment.

This note was inspired by a recent discussion on the Infodesign Café. The most thorough and insightful account of the debate is Paul Stiff's 1996 paper: Stiff P (1996) 'The end of the line: a survey of unjustified typography', Information Design Journal, vol 8: 125-152

Tuesday, January 09, 2007

Colour, flags and sensitivity

Working recently for a government department whose corporate colour is green (a kind of greyish, dark green, not exactly emerald or shamrock), we proposed a colour-coding system where orange was among the colours used for navigation.

When someone suggested that this would render the document unacceptable in Northern Ireland, my pedantry alarm went off - surely the juxtaposition of two colours in a coding system does not amount to a flag. But I followed up with a call to the Department of Culture, Arts and Leisure in Northern Ireland. After some consultation we turned the orange into orangey-brown.

The DCAL website is in English, Irish and Ulster-Scots. I had no idea that there was an Ulster-Scots language that is still used, but it seems there is, and I followed up the link to the Ulster-Scots Agency (Tha Boord o Ulstèr-Scotch). Predominant colour: orange, so no surprise there.

The Law of the Penultimate Solution (it's always better than the one you end up with)

Someone reminded me yesterday of a talk David Lewis and I gave at one of the Information Design Conferences in the early 90s. We likened the information design process to a lens. Information designers take requirements and constraints that were previously uncoordinated and unrelated, from different parts of an organisation, and focus them in a single design solution that is coherent and usable by customers.

Then, as we consult and amend our perfectly-focused solution, it shifts just slightly out of focus before it's implemented. The final solution is never as good as the penultimate one.

And finally, there's often a stray photon that arrives from nowhere, completely bypassing our lens. The marketing director's twelve-year old son has drawn a great logo (it's happened); someone has read that italic is always illegible; someone doesn't like green.

Sunday, January 07, 2007

How to operate the shower curtain

Lovers of instructions will appreciate this article in the New Yorker:
How to operate the shower curtain.

And speaking of bathrooms, does anyone know the history of that card you get in hotel rooms worldwide. The one that starts "Dear Guest. Imagine the amount of detergent..." and ends "Towel on the floor means 'change this towel'" .

Even the poshest hotel rooms have this card, which never says "Dear Guest. If you really want to save the planet, take your holiday at home next year".

Friday, January 05, 2007

Money laundering


Eden are an excellent information design consultancy in Amsterdam. They recently came up with new standards for consumer warnings for financial services products (eg, 'investments can go down as well as up'). To simplify the approach, they used washing instructions - a trusted information source - as their inspiration. Have a look at their full case study here.

Wednesday, January 03, 2007

Force 7 confusion forecast

It would be nice to have a version of the Beaufort scale or Richter scale for information. The Beaufort scale is the one for wind, that starts with '0. Calm. Smoke rises vertically' and ends with '12. Hurricane. Considerable and widespread damage to structures'.

So I've had a go.

0. Calm. Ideas rise vertically from page to mind.
1. Light difficulty. Slower reading. Dictionary pages rustle.
2. Moderate difficulty. Reader lightly swaying; visible perplexity.
3. Difficulty. Head shaking, audible groaning.
4. Severe difficulty. Loud muttering, and music heard from helpline queue.
5. Very severe complexity. Foaming. Abrupt movement about room, with swearing.
6. Storm. Whole documents in motion, from table to floor.
7. Brainstorm. Considerable damage to conceptual structures.
8. Typhoo. Reader flattened in darkened room, with cup of hot sweet tea.

Blog Archive

rob waller rob waller rob waller