Tag Archives: google

The Semantic Web just hit the big time

Ever since I got involved in the Semantic Web, people have been asking when the world will start using it. Apparently quite a few have been asking that question for over a decade prior. Well folks, wait no longer.

Facebook, arguably the #1 site in the world, is powered by the Semantic Web.

For the past year, Facebook’s Open Graph Protocol has been my favorite API. Even without talking to Facebook, you can leverage OGP by pinging the large number of sites tagging themselves for the sake of Facebook.

At this year’s F8, the annual Facebook event, OGP just grew by leaps and bounds. Facebook needed to gather about a year’s worth of data to enable some of what they’re doing now.

But with the latest changes, Facebook is poised to challenge Google with a search engine of their own. Give them a year or two of data with the new and improved OGP, and you have a powerful semantic search engine.

Exciting times my friends.

Why would Google be Traffic Shaping?

There’s an interesting thread over on Webmasterworld about Google Traffic Shaping. It may seem overbaked a bit, but if you know some of the thread participants personally, you’ll realize that if their focus is on it, there’s a good reason.

Secrets abound and on these threads you need to read between the lines a lot or else you’ll be scratching your head.

To summarize the thread, some people have noticed Google tuning their knobs on a site by site basis with respect to traffic that converts. One month traffic may go up while conversions go down…and vice versa.

We all know Google is a data analysis monster. You kind of have to be when your products are “free” and you have to make money through the back door, if you know what I mean.

So how do they make their money? Selling adwords. Lots of them. The more money their advertisers make, the more they’ll advertise. And the more Google will make.

So while they may repeat the mantra that they want to provide the best possible results to the user…as a public corporation, their DUTY is to provide the best possible return on investment to their shareholders.

Therefore, and I say this without emotion or casting aspersions, you have to assume that their dream scenario is a bit different than the PR….

For them, the shopper is the most important Google visitor. Give your customer (ie advertiser), a shopper who converts, and now you have a product with a metric that sounds like music to advertisers.

Imagine if you will, you’re a local shop, independent or outlet of a large chain. You want to pay Google for shoppers and shoppers alone.

So in Google’s ideal world, shoppers will type search terms in Google. All organic results will be of no interest to the shopper and only the paid search results will entice them to click. And once they click, they will convert.

THAT’s why they are experimenting with traffic shaping, particularly in respect to conversions. And they seem pretty far along.

The take home point for webmasters is, to score well online, at least with Google, you need an online presence that converts…and you need to pay Google for that traffic.

I really don’t want any of my friends at Google to be upset by this post. So I’m going to repeat. Google is a public corporation. PR is PR. Business is business. And if Google AREN’T doing this, they would be breaking their fiduciary responsibility. Like in the Godfather, everyone needs to put bread on the table. It’s not personal.

Web roundup

  • Ian Bullard asserts that you should never use the RAND() function. The Ayn Rand Google Ad is pretty funny.
  • The new Zend Server is coming out as a web stack. Although it can be easier to install a Xampp style web stack on your development machine, if you’re a web developer, you’re better off installing each component separately. In the long run it’s less buggy in my opinion, and what you learn in the process of doing it yourself is priceless.
  • With the Zend Framework 1.75 release, Zend View has an important security fix in it, that is somewhat incompatible with previous version. Check it out if you are upgrading.
  • Google has finally lost some market share…to Yahoo!. Despite all the bad press Yahoo! has gotten lately, they have some great Semantic Web products. Including Search Monkey, which I plan to post about soon in anticipation of our Semantic Web Meetup next week with Peter Mika presenting.
  • Eric Meyer provides feedback on some long awaited CSS3 features. It will be very cool to use shapes in css.
  • Congratulations to WordPress on their new datacenter.
  • Interesting to learn (via Baron) that there’s a MySQL Federal Migration Bootcamp. Good idea. I hope to soon see a “Don’t Waste Taxpayer’s Bailout Money Bootcamp” too :)
  • Sad to hear that Ma.gnolia lost all their data. A good DBA and backup strategy is vital to the success of any startup. Very impressive that it was basically a one man operation though. Kudos for getting as far as he did…