Navigation Social Software

On Live Journal mood tracking and zeitgeists…

There are two or three major things I’m thinking about at the moment – and one of them is zeitgeists. Which brings me rapidly to The World according to LiveJournal which is an awesome tracking system of LiveJournal moods over the last seven days. If you go and look at it now and try moods like ‘sympathetic’, ‘distressed’ or ‘nauseated’ you can see that the bombings in London have had a real impact on people’s moods. If you invetigate more thoroughly you can see that other moods have been inversely affected or show more complex relationships. There was a parallel drop on ‘horny’ during the coverage, an enormous drop in ‘irritated’ which then turned into a spike. Fewer people felt ‘guilty’, more felt ‘grateful’.

A couple of obvious things fall out of this for me – you could use this data to articulate relationships in moods really effectively – which things in the world cause reactions, what kinds of reactions do they cause, which moods are more closely correlated or act against one another. I keep looking for clear moods that you’d expect to see appearing twenty four hours after an event like this, but so far I’m only seeing a few (people seemed to become irate in two major spikes – I wonder why).

Another obvious thing would be to use this data to alert people to things that were going on in the world or to track trends over time. I believe that LiveJournal knows which country people are from – combining that data with the stuff from the site would be tremendously useful. Sending alerts to news gathering organisations would be interesting too. Mood expression and collation is such a fascinating area and has some real possibilities for data-mining and zeitgeist taking. Can anyone else think of good ways to get this information from people and to employ it – ideally in an open way? The best I can come up with off the top of my head is AIM status messages using controlled vocabularies and opened up in some spiderable fashion…

Design Navigation Personal Publishing Social Software

On the 'one big site'-ness of weblogs…

Here’s a weird quote about weblogging: “I believe in my heart that people should come up with their own publishing methods. Frankly, it’s boring to surf the blogosphere and see so many sites using the same, tired weblogging tools. The same basic templates, the same ‘post a comment’ form, the same URL schemes! It’s almost as if they’re all small parts of one huge site.” (Adrian Holovaty).

So my immediate reaction is that the fact that there are a limited set of really popular weblogging systems has probably been a good thing, because it means there’s an active and widespread community large enough to be able to self-support, fully explore the boundaries of the software available and push for new functionality. But more importantly, there’s an element in which all weblogs are part of one huge site. And that’s only partly the sense in which all the web is basically one big hypertext entity in which all boundaries between sites are essentially arbitrarily – or culturally – enforced.

More specifically I mean that at that point where a weblog is pretty much balanced between personal publishing (micro-broadcasting or ‘one-to-some’ communication) and social software (something like a distributed discussion board) there are aspects of ‘one huge siteness’ in play – and that that’s precisely why they’re mostly working. We have a roughly common vocabulary about what an entry consists of, a set of structures about how a site works, and systems of trackback, permalinking and commenting that are pretty much interoperable (in one form or another).

I suppose if I wanted push an old comparison (that I never thought really worked) in a slightly different direction, then I’d say that weblogs needed to be ‘like one huge site’ to the same extent that a peer-to-peer network needs to consist of mostly coherent and standardised applications in order to do what it does. Maybe some of the newer responses to writing and interactions between people are demonstrating that ‘siteness’ (heimlich) and ‘unsiteness’ (unheimlich / other) aren’t categories with as much utility as we once thought – or at least that breaching or straddling them provides opportunities for new, powerful kinds of applications.

Hacks Navigation

Against Search Engine Optimisers…

In the middle of the comments for a fairly interesting article about the Googledance that never ends there’s a post from a professional search engine optimiser. He says:

My consulting business website ranks highly in google for a number of search terms that are pertinent to my business. I didn’t get that way using a search engine optimization service. It didn’t cost anything but my time and the sweat of my brow. And it’s really very simple how it works. I tell all my methods in How to Promote Your Business on the Internet.

In summary:

  • Put content on your site that visitors will want to read – and return to. Not just material aimed at potential customers, but stuff anyone will want to read.
  • Post new content regularly
  • Ask for links, and offer reciprocal links

That’s the method I used to make a Google search for software consultant turn up my resume as the #4 search result.

I want to make something clear. This is probably one of the best statements about search engine optimisation I’ve ever read, and it’s still horse-shit. The thing that it says that’s actually useful is that you should have a good site. First and foremost – put content on your site that people want to read and update it regularly. That’s a really really good point and something that people should remember. But it’s not something that a search engine optimiser can help you with, so that leaves you with link-exchange. Which is horse-shit. I’m going to say that again because I enjoyed it so much. It’s horse-shit.

Ladies and Gentlemen, listen very carefully when I say this: There is absolutely positively never any reason whatsoever to go to a search engine optimiser and they may damage your business as much as they help it. The reason they may damage your business is because – for the most part – they are designed to hack the system – to find short-cuts and tricks that fool a search engine into believing your site is something it isn’t. And search engines change their indexing methods all the time to compensate for these tricks. All the time. Google do it monthly! And if they find someone using them – often they’ll penalise the sites concerned.

Here – then – is the big secret of search optimisation. Search optimisation isn’t really about optimising for a search engine at all. It’s about making good quality, cleanly designed, semantically-constructed sites that people want to read, that people can link to and which people can get the gist of in a few seconds. If you make a website well for human beings, then as a side effect – more often than not – search engines will spider it well and rank it highly. And they’ll do this because it’s the best site, not because you’re trying to fool them.

For the most part this is all you need to know:

  1. Highly complex and flashy animation does not help you, it hinders you – your site needs to be easily spiderable and that means that tricky navigational elements probably won’t help. If you have to use them (like I do for my archives above), present alternative simple ways to get around your site as well that use basic boring run-of-the-mill links. This is not a search optimisation tip – this is good navigation design.
  2. Meta-tagging is not that useful any more. But if you’re going to use it, do it properly. Specifically, if you’re going to put in description and keyword metatags – keep them short (twenty words most), accurate (actually reflecting the content in the body of your page), and don’t put the same metatags on every single damn page of your site! That’s not going to help at all! None of this will affect Google, who don’t pay any attention to meta tags and make up 50% of the searches performed on the web at the moment.
  3. Make it easy to link to things! This means, don’t use frames! This means, try and put discreet chunks of content on clear separate pages. This again is not optimising for a search engine at all – it’s how to build an information-delivery site properly.
  4. Use <title> tags properly! They sit at the top of every one of your pages and they’re designed to make it easy to spot things when you bookmark them. So tell people the title of the page you’re on – and do it honestly! Keep them short and clear, don’t use marketing speech at all, don’t try really really hard to find the right keywords, just use the title that explains what’s on the page best. To help people who bookmark you then you should probably put the name of your site at the beginning or end of the title, and if you’ve got a shallow site hierarchy, you can even put the path to the current page in the title as well. These things are helpful to people! Unsurprisingly, search engines try to use the same criteria as actual people do.
  5. Use semantic content whenever possible. This means when something is the title of a page or a section, stick it in a <h1> tag and use CSS to style it appropriately (and before you say anything, I’m aware that I don’t do that – but there’s a really good reason for that). Also when you’re linking to things don’t use terrible words inside the links like “click here” but actually use read words. This is good for people and helpful for search engines. Don’t lie! People would find it more useful if you linked to a page about sportscar GT with a link that said “We have a comprehensive section about sportscar GT“. Search engines – weirdly – do too!
  6. Bugger link-exchange! Google specifically penalises people for using known link-exchange programmes because they’ve been designed specifically to circumvent Google’s attempt to find quality sites that are well-respected and rated. Don’t try and fool the search engines unless you’re prepared to pay for search engine optimisers to come in and fix your site every two weeks.

God there’s loads more stuff I could say, but the rule of thumb is the same for all of them. Build sites that are easy for people to use, try not to let the technology get in the way of delivering the information and aspire to making things that work the way the web works, and you’ll never have any trouble with search engines.

Addendum: There’s an interesting article on Google over at Salon today in which – yet again – some of the people who try to mischaracterise the usefulness of their own sites by gaming search engine algorithms claim that not being allowed to lie about their site’s relevancy is terribly terribly bad. I have absolutely no respect for these people at all…