June 17th, 2008


On my way back I stopped at the Borders in Fosse Park. It is always interesting to look at the computer books section in bookshops outside of Oxford which is, I think, rather atypical, to see what is being foisted on people. I was rather surprised to see a copy of the Pragmatic Programmers' Erlang book. This, surely, must be the most bizarre hyping of a language in the history of computer books. I am very suspicious of the ecosystem surrounding the more 'fannish' of the computer book publishers - O'Reilly, APress and the Pragmatic Programmers. It seems to me that 'the gang' publish books about technologies which people then assume are legit because there wouldn't be books about them if they weren't legit. Thus people use those technologies and more books get written about them and so on...
Looking at the blogs on O'Reilly it all seems to resemble fandom more than anything approaching an engineering discipline. If you read the O'Reilly blogs and didn't know better you would think 'I can see why this Linux thing has 90% market share, with the premium Apple thing having its 8% of people willing to pay extra for it, though honestly I can't see why these Microsoft people with their crappy unpopular Windows thing struggling along the bottom even bother'.
I am also rather unconvinced by their analysis of book sales by language, if you read the small print the cut-off is such that if there were 10 approximately equally popular books about a language it could actually be selling as many as one of the 'mid-major' ones without getting onto the list at all http://radar.oreilly.com/archives/2007/05/state-of-the-computer-book-mar-15.html (and the lower tables are essentially just the noise at the bottom).
Then you get things like the bizarre Tiobe league table http://www.tiobe.com/index.php/content/paperinfo/tpci/index.html If you think 'D' really is one of the A list then I have a bridge I would like to sell you.
Of course, none of this probably matters very much, except that I am sure that people do make bad decisions based on it all.