Jakob Nielsen reckons it's time to rehabilitate the passive voice in writing. And it's all in the name of search optimisation - as opposed to search engine optimisation (SEO). It's an approach that might have legs but is more likely to result in a lot more gibberish appearing online.
The idea is that people surfing, and especially shopping online, scan web pages in a cursory way that favours words over to the left. By altering a sentence so that key words come first - something that probably involves using the passive rather than the active voice or, in rare cases, flipping the word order round – you can capture their attention for longer. If you look at the results they got from capturing users' eye movements, readers also seem to favour short measures. So, it is at least good to know that conventional newspaper and magazine layout ideas were right all along. It's the reason why this blog has such a narrow template. (OK, it's a sample of three at Nielsen's site, with just one piece of running copy, but it fits my prejudice.)
The passive voice has its uses. It provides a handy way of altering the rhythm of a paragraph amid a lot of active-voice sentences. It is also dangerous.
The passive voice makes copy far less readable and, as commenters at Boing Boing pointed out, lets you get away with conjecture far more easily than the active voice. Converting everything to active voice is a highly effective editor's tool for working out whether a writer has stood up a fact or not.
Readability and vagueness are not the only the problems with altering copy to try to rank better in search engines at the expense of readability. The search engines are moving targets. This week's re-ranking of sites by Google, apparently based on their use of outbound links, is an effective demonstration of that.
If people start trying to force keywords into prominent positions, they may find themselves victims of a future splog cull - because the people putting those together don't care about readability, only about the SEO aspects of a web page. They will happily lob keywords anywhere they think will help them in their quest to appear higher on results pages. For that reason, tt's not hard to imagine the developers at Google or another search engine focusing on keyword position as a way of identifying splogs and then dropping them way down the rankings. The search engine software is getting better at understanding the structure of copy - look at the way that Google now handles word stemming to get different variations of a keyword. Not so long ago, you had to use the OR operator to get the same effect.
You also need to look carefully at what actually appears on the search-engine results pages. Google tries hard to pull relevant sentences that contain the keywords out from the copy. The blurb underneath the page's title may not be the intro paragraph but a completely different sentence. Are you seriously going to render all your copy in passive form just to get keywords upfront? Especially when Google plays nice and puts the keywords in bold.
I worry when people start to talk about search-optimisation tactics for copy. What works today is unlikely to work tomorrow because the wiring in software can change much more quickly than the wiring in people's brains. And it is the people who we should be writing for, not the machines. They should learn to do it our way.