OK - so what I should really have written for the headline, is something like 'Great New York Times piece about how Google is changing the way headlines are written on big new media sites' - which isn't really a headline, but what the hey.
The link probably won't work now, but Collision Detection has a good summary of the key points. There's been a whole heap of comment on this already online - try Simon Waldman or the Technorati aggregation of the blogosphere responses.
(By the way, the image comes from the original NYT story and is by Paul Sahre)
The gist of the story is that the kind of clever, allusive, punning headlines that grab the eye in newspapers can confuse the automated programs that search engines like Google use to gather links and info. These programs are quick and smart, but powerfully dumb when it comes to puns, wordplay and the like - so they find it easier to process simple 'does what it says on the tin' headlines. Result - if you want to achieve high visibility on Google and other search engines, you need to give your online news stories simple informative headlines.
As the article points out, search engine optimisation is a growing business, one that takes in online retailers, spammers and scammers and, increasingly, journalists. The BBC is already experimenting with headlining stories twice - there's a clever alluring hook on the main page for human readers, and a straightforward informational blurb on the actual story page.
In the past, journalists have developed different writing techniques in response to different constraints - production schedules, design formats, new technologies (the NYT story talks about how the inverted pyramid style of news writing was a response to the telegraph). Google represents a new set of constraints - one that can affect more than just headlines.
The natural-language processing algorithms, search experts say, scan the title, headline and at least the first hundred words or so of news articles. Journalists, they say, would be wise to do a little keyword research to determine the two or three most-searched words that relate to their subject — and then include them in the first few sentences. "That's not something they teach in journalism schools," said Danny Sullivan, editor of SearchEngineWatch, an online newsletter. "But in the future, they should."
Two things to think about here: first, Simon Waldman's worries about Google being a force for a certain homogeneity in news writing online...as Jeff Jarvis points out, this really only applies to Google News, not to Google on its own. You can imagine a world in which journalists learn a few new tricks - search engine-friendly headlines, key word laden first paras - without churning out identikit stories (which, of course, they don't do now). It depends if the pressure to be more visible on Google mutates to the pressure to be more visible to Google News.
Second, pace Danny Sullivan, and without wishing to indulge in shameless self promotion, which clearly isn't the done thing on blogs, we do teach this kind of thing in the Online Journalism course at UcCA at Farnham. I do a session on writing online, which revolves around a discussion of usability guru Jakob Nielsen's guidelines for scannable prose, headlines and the rest. Nielsen's been banging on about making text online easy for machines (and humans) to scan and process for quite a while now.
Course, you don't have to agree with everything he says - some of which feels very early web these days (though a big update is imminent). And stories written according to his guidelines might be visible to search engines and easy to scan, but they're not that readable, if you know what I mean.
In the class I teach on this, we usually look at an example of prose that's been Nielsen-ized for the net - last year, I even got the students to Nielsen-ize an alternative set of guidelines for writing online. Overall the aim is to pick up what's useful from Nielsen. Take his approach too far and you end with prose you don't read but process, or parse.
In his piece about how users don't read online, if you, well, don't read, but scroll down (I think that's permitted now...) to the Nielsen-optimised prose at the end - the one that's 'precise, scannable and objective', in his words - and boasts a 124% increase in usability - well, it's basically a bullet point chart - it's not writing any more.
Completely Nielsen-zed prose can be kind of dehumanised - it's all about crude, unambiguous information delivery. It loses the conversational personal flavour that is one of the nice things about much online writing. You could argue that blogs - and the personal rambling they contain, which got started end of the nineties - were perhaps, in part, a response to the kind of homogeneity brought about by Nielsen's influence on the big portal sites of the times and the way they presented information...
I digress... Actually I don't mind Nielsen-ized search engine friendly prose at times. Overall, I guess it's a 'horses for courses' thing. Most machines, and some people - e.g. the RSS reading infomaniacs Clive Thompson mentions in his post - want as much info as fast as possible, all the time, with no buggy ambiguities slowing the raw data transfer.
But the rest of us flip between info-bites and more human writing, depending on what we want and how we feel. And the beauty of the net is that it can do both at once - in theory. The real problem might be if people started to insist that there's only one way to write online...
Recent Comments