Matthew Revell

The Guardian’s media section publishes its Buzz 30 list, today. I hate these things: a few bored hacks do their best to fill a couple of spare column inches with their prediction of what people will consider cool, or hot … erm.

Blinkx, predictably, makes it to the list. The Guardian describes it as:

Dubbed the internet’s “first intelligent search engine”, Blinkx not only searches the web but also news sites, emails, attachments, blogs and your own hard disk. Unlike Google, it also uses artificial intelligence to rate stories, not page rankings. Only three months old, Blinkx has made quite a splash.

This is a perfect example of the crap that makes up these filler “what’s cool” columns. It’s no wonder that one of Blinkx’s founders is a PR expert. Like just about every mainstream hack writing about Blinkx, The Guardian reproduces the Blinkx PR line almost verbatim. I love the reference to “articial intelligence”; ladies and gentleman, I present to you ye magikale arteefishal intelleegence! The only place Blinkx has made a splash is amongst ill-informed, mainstream journalists who are desperate to scoop the Next-Big-Thing.

Why do I dislike Blinkx? Well, here are the first three reasons that occur to me: it relies heavily on Google, throws up poor results and doesn’t fit in with the pro-active way in which people use the web. But if that was all, I’d just ignore it. What I dislike most is the hype surrounding it and the unthinking, lazy press coverage it’s received.

Well done to the Blinkx PR team. If only you’d put as much effort into your product.

The BBC’s Ceefax service turns 30 next week. Readers of BBC News Online were asked to write a story, titled “1974″, in no more than 70 words.

The entries show how 70 well-written words can say more than a thousand that are just slapped on the page. The story by Tok Thompson, Dublin, shows just how much can be packed into a few lines; it’s powerfully evocative.

BBC Ceefax 1974 in 70 words.

I’ve decided to do the Chartered Institute of Marketing’s Professional Diploma in Marketing, at the University of Wolverhampton.

My job is an odd mix of technical know-how, creativity and marketing experience. I reckon that formalising my knowledge of marketing can only help me do my job more effectively.

Many people have asked where LugRadio is.

If you’re a listener to the show, you know we took a summer break. During that time, we swapped servers and because we were all a bit busy, there was some downtime for the LugRadio.org site. Unfortunately, we’ve had some trouble moving the domain to the new server. This should all be sorted out soon, though.

Other news: Ade has set up a blog detailing his experiences with the new, Debian-based Linux distro for the masses. This new distro has the potential to be quite interesting: the stability and package maintenance of Debian with the user friendliness of, say, Linspire. Let’s hope!

Here’s an article I wrote for ContentPeople, in April/May 2003, on Flesch readability.

Clarity is the commercial writer’s goal. With practice, it comes naturally. Finding the right level of readability is usually about gut feeling and consistency.

Increasingly, larger projects are looking to objective methods of measuring readability. The first - and most used - of these is the Flesch readability formula. Devised in the 1940s by American linguist Rudolf Flesch, it measures the average number of syllables per word and words per sentence. Using Flesch’s chart and a ruler, a score is given. An easier way to check a Flesch score is with a software tool, such as Microsoft Word’s grammar checking.

In his book How to write plain English, Flesch admits that using such “a mechanical gadget for this doesn’t seem like an intelligent approach”. His belief, though, is that it reflects the process the brain uses to read. Essentially, Flesch says that it’s easier to read shorter sentences that contain shorter words. Not exactly a major breakthrough.

Perhaps the major flaw of readability formulae is their disregard for context. As John Wild, of the Plain English Campaign notes, “‘The cat sat on the mat’ has exactly the same readability index as ‘The mat sat cat the on’.” It has to be assumed, then, that anyone whose writing is measured using Flesch, or other formulae, already has a mastery of English.

Echoing John Wild’s reservations, Phil Scholfield, of the University of Essex, sees readability formulae as of limited usefulness. “While it is true that usually longer sentences and longer words are harder, that is not always so and several other things can make a text difficult, such as the complexity of its organisation and the difficulty of its thought/content.”

In an effort to achieve a higher Flesch readability score, it can be tempting to forget the actual, human readability of a piece and start slashing away at words. Almost slipping into the spirit of George Orwell’s Newspeak, Flesch suggests that writers should “take first aim at words with prefixes and suffixes, like establishment, available or required” and replace them with “a two-word combination like setting up, in stock or called for“. Rather doubleplusungood for subtlety in English, I’d say.

To be fair to Flesch, he admits that his formula is only useful as a guide. Certainly, for projects with large numbers of writers, the Flesch system can help to produce a uniformity of style. However, as web based translation services prove, language cannot be reduced to computerised rules.