this week brings more releases than one of my dads fishing trips
What, two weeks since a roundup? Say it aint so! Yes folks, last week was a holiday for us, and I was too busy Friday to set one up beforehand. Don't worry though, I'm a slave to my job every monday from now until some time in the autumn. Or until I successfully lobby the BC government for a four day work week.
This week we have the release of Chrome 2, WolframAlpha, new features for Google Suggest, usability considerations for SEO, and more…
- We start the week with some pretty major news. A lot of people are talking about it but I'll cite searchengineland on it: Google is adding ads to suggested search. Oh dear, I don't know whether to love it or hate it, since it means more advertising directed at me, but at the same time easier promotion for PPC.
- Next up, David Mihm and a framework for thinking about local search campaigns. He argues that we need to stop thinking about local search as a part of SEO, and thinking about it similarly to how we think about PPC. His advice for setting up a local search campaign is pretty good too, providing a good early checklist.
- Finally, Bruce Clay is linking two of our favourite topics with “Adding user behavior to SEO considerations”. Really his point is that value doesn't come from traffic but from targeted traffic, something we couldn't agree with more.
- It's out! It's out! And it really isn't all that exciting. Google Chrome 2 has been released. Wow. Ok. Now wheres my Firefox 3.5?
- As I am sure you know by now, WolframAlpha is out. How exciting! Now if only it could understand a single command I give it. Yeah its buggy right now, but it will get better, right? Perhaps when its out of Alpha, and with a heart appended…
- And with WolframAlpha's release, Stephen Spencer is saying that an age of computational engines is upon us. Well, ok my wording was bad, but he does talk very positively about them and tha… oh god I feel a rant coming on…
See, I am perhaps the only person on the internet that doesn't like WolframAlpha. In fact I don't think its going to get anywhere unless they do some major changes to it. Why? Well… it's so 1990's (or 1980's, if you remember Cyc). Spencer thinks its a powerful new way of interacting with large repositories of data, but honestly we've had similar systems for interpreting data around for ages and they've never gone any farther. Why? Well because they are complex and difficult to manage.
The problem with Wolfram Alpha isn't something that can be fixed by a dev team; it's too closed. Unlike Google, it's trying to find focused information, and unlike wikipedia it isn't allowing the user manipulation and content submission that would fix errors as they occur. Wolframs focus is in building centralized (or in this case authorized) data bases of links based on dev selection, instead of user interaction. It is designed as a top-down, instead of a bottom-up information system. Thus its potential knowledge base will always be substantially smaller than that that Wikipedia or even Google could provide, and due to its complexity (and frequent lack fo content) likely more troublesome to use. Its only strength is in its qualified answers, but thats meaningless to people either too limited in knowledge about wikipedia/google to be wary of it, or knowledgable enough about both to be able to gather proper research with them.
So what could fix it? Well, its a great interface. Really great. It also has a neat parsing concept. So why not expand it? Why not make it an interface for, say, wikipedia. Then give people greater control over keyword selection and preference (like, say, being able to choose from a drop down menu other options for that keyword—like Alberta, Canada), and maybe even connect a result to the keywords they're looking for, should they not find it immediately on page (this would immediately resolve issues like “Alberta, Canada” returning no results)? This would also give it a HUGE database of information to work from. Finally, it would be integrating directly with the tools people are already using, meaning adoption would likely be much faster.
- In analytics this… err, last week I guess, Avinash is at it again with a 6 step process to evolve your mental analytics model. This cathartic piece from everyone's favorite analyst takes on the old motto of GIGO (Garbage In Garbage Out) as it applies to analytics.
- Our PPC guy recently turned me on to Copyblogger, which, this week, has a post on landing page makeovers. An odd topic for a site normally focused on content writing, but still damned good.
- And finally, GoodUsability on FAQ usability. As normal good stuff.