Computer programming centre-right lovers of Swedish meatballs

English: YouGov logo
English: YouGov logo (Photo credit: Wikipedia)

According to YouGov (the UK’s largest polling company) that is what typical lovers of Linux are – though it’s based on just 272 individual profiles (out of 200,000 or so members of YouGov’s panel). Oh, and they are blokes. More at

(YouGov made at least some of their profiling data available online this morning and it has kept British internet users amused all day.)

Windows lovers are, apparently, somewhat more numerous – there are 744 of them – but also typically younger, less well off and even more right wing. And are also men. Apple pie is their favourite dish and perhaps unsurprisingly they are not as keen on programming. Yes, it’s true: Windows lovers are lusers through and through. See

Admirers of the Microsoft brand, though, tend to be older (still male) and rather more centrist – and numerous. Perhaps this is the Bill Gates effect? People admire his creation in the abstract but there is little concrete love. See

But what of your favourite hipster computer brand – Apple? Turns out they are centrist, female and middle class and like grilled halloumi cheese. It’s harder to make a direct comparison though as (surprise, surprise) Apple users don’t seem to identify their operating system. See

There is lots more to look at – for instance Android users are seemingly very left wing while computer scientists are middle aged men who eat a lot of chicken.

Computer scientists’ lousy citation style

I am reading this book: Soft Real-Time Systems: Predictability vs. Efficiency, and I am struck, once again, by the truly lousy style of publication reference that seem to be preferred by so many

Journal of the American Mathematical Society
Image via Wikipedia
computer scientists,

The style used in the book appears to be that favoured by the American Mathematical Society – the so-called “authorship trigraph” – with references made up of letters from the author’s name followed by the last two figures of the year of original publication eg., [Bak91] which references in the bibliography:

[Bak91]        T.P. Baker. Stack-based scheduling of real-time processes. Journal of Real Time Systems, 3, 1991.

Now it is possible, if I were an expert in the field that I might recognise this reference, but it is far from helpful. When referencing papers written by multiple authors the system is hopeless – using the first letters of the first three authors and ignoring the rest, eg., [DGK^+02] is a real reference in the book to a paper with eight authors. I really doubt many people would get that straight away.

But at least this reference system contains slightly more than the IEEE‘s citation system, which demands papers are merely referenced by a bracketed number in the text, eg., [1].

These reference systems are so widely used that I worried that my own use of the Chicago system – which specifies author and year, eg., (Denning, 1970), would be frowned upon in Birkbeck – but a re read of the regulations showed their demand was for a consistent and well-recognised system.

The ACM, too, promote a sensible citation format eg., [Denning 1970].

Does this matter? Yes. I am sure many readers of these books and papers are students who are conducting literature reviews or similar exercises. Reading the original reference may often be important and having to flick back and forth to a bibliography to check the meaning of an incomprehensible reference is not calaculated to add to the sum of human happiness.

(I don’t have any real complaints about the book though – except that the translation is plainly a bit stilted – for instance, the first sentence of the book refers to real time systems being investigated “in the last years” – a fairly common mistake in syntax from non-English speakers and one that the editors really ought to have corrected. But the occasional infelicity of language does not detract from the book’s overall appeal.)