Search in the Wild, Wild Web
I’m glad I was out of town over Christmas, because if I had to read the "Wild, Wild Web" article that ran in the LA Times on Christmas Eve (in the Calendar section, no less) it’s very likely I would have jabbed a ballpoint pen straight through my right eye.
The premise of Reed Johnson’s December 24th article was that the Web, which people once hoped would remain "democratic" and "untainted", has evolved into something a little seedier and it’s basically all our fault. Don’t worry; I’ll let you finish reading before you take your rightful position on the naughty step.
Reed refers to the Web as:
"A global, digital Miss Universe pageant in which the ‘most linked’ and ‘most emailed’ sites get the tiara, even if they won partly by sabotaging their rivals backstage."
While I love the image of my Rand in a ball gown and tiara, I think Reed is confused. He seems to be lumping Google into the same category as sites like Digg, where popularity is determined by clicks and user ratings, not the 100+ plus factors that go into Google’s algorithm. Obviously, that’s not a fair comparison to make. Google and Digg are both important to search engine optimization, but they’re not comparable systems.
The most fun part of the article is when Reed blames the deterioration of the Web on search engine optimizers, even defining our industry for us.
"A cottage industry has sprouted up around “search engine optimization,” more commonly known as boosting Google rankings. The trick: seeding a website with key terms that will show up in text hyperlinks, regardless of the site’s actual importance or relevance. And there’s always the fallback of enlisting actual humans to help click you up the list."
First, I love that search engine optimization is in quotes. That made my entire day. But honestly, you have to ask yourself if Reed even talked to an SEO before penning this article. I have a problem with labeling our profession as simply "boosting Google rankings" because that’s not what it is, at least not for us. We’re not interested in seeding a site with irrelevant terms, we want to uncover the terms that go along with your site’s theme and make sure you’re using them correctly. We want to make sure you’re linking out, that your content is unique, that you’re using Meta tags, that your site is designed correctly, etc. It’s against our code of ethics to trick either the user or the search engines.
Bruce is famous for saying that search engine optimization is not about making a pig fly. It’s about genetically re-engineering a site so that it becomes an eagle. For those that don’t speak fluent Bruce, he’s saying SEO is not about getting an undeserving site to rank for the wrong reasons, it’s about creating a site that’s relevant and expert, and then organizing it so its usefulness is evident to Google, as well as to searchers. There are no smoke and mirrors here.
And this thing about enlisting actual humans, does that mean the fake robotic ones won’t work? Drats, there goes my Sunday.
One of Reed’s other issues with the Web was his declaration that the good, authoritative stuff is being overlooked for the more vapid, popular sites. He uses the fact that a Google search for the term "God" brings up a breezy Q&A video "punctuated by New Age-y piano riffs" (sweet!) instead of the Book of Job or the Bhagavad-Gita. I think his point is valid, a lot of good stuff is getting lost in Google’s crowded index, but how perfect can you expect your results to be for a three-letter, one word search phrase?
Search isn’t perfect. It probably never will be. Right now if you’re looking for targeted information that you’re not finding in the engine’s main index, try using one of the verticals. Use Google’s Book Search for your "God" query. You probably won’t get the Bible on the first page, but your results may be closer to what you’re looking for. With the engines indexing more and more, you have to learn to search smarter. There’s no way around that.
Reed does touch on something interesting when he quotes Oren Etzioni who says a solution for the Web’s current dilemma’ is to create "unsupervised information-extraction systems". Now before you go hurting yourself, I think that’s just geek speak for a search engine that understands sentences beyond basic keywords. Who wouldn’t love that?
According to Oren, Google and the University of Washington have already combined forces to develop such a system to help gives users accurate and authoritative information. I think it’ll be awhile before we see anything like that launched mainstream, but it’s cool to know it’s in the works.
Anyway, go read the article. You’re guaranteed a chuckle or two.