I was interviewed for a piece published today in the New York Times. I hate to sound like a Winer here, but talking to ‘BigPubs’ is just a giant waste of my time. I’ve done this half a dozen times now, and every time, I talk to someone for 45 minutes or so, and they either don’t use anything I say, or use one context-less quote to make me look like an idiot or to help “prove” whatever “angle” they’re trying to take in the story, that I usually don’t agree with.
Not that it will matter, but I’m going to point out what I consider to be the problems in this piece. (Well, maybe it will make me feel better.)
But really, any press is good press, and it’s not every day you get mentioned in the New York Times, so try not to take my bitching and whining too seriously.
"The unlikely electoral battle is being waged through 'Google bombing,' or manipulating the Web's search engines to produce, in this case, political commentary. Unlike Web politicking by other means, like hacking into sites to deface or alter their message, Google bombing is a group sport, taking advantage of the Web-indexing innovation that led Google to search-engine supremacy.
The perpetrators succeed by recruiting a small group of accomplices to link from their Web sites to a target site using specific anchor text (the clickable words in a link). The more high-traffic sites that link a Web page to a particular phrase, the more Google tends to associate that page with the phrase - even if, as in the case of the president's official biography, the term does not occur on the destination site."
First, the word choices - “perpetrators” and “accomplices” - especially in light of the previous paragraph about “hacking into sites” implies that Google bombing is somehow a criminal, illegal, or at the very least unethical activity. It is not. Creating links with anchor text “miserable failure” is no more illegal or unethical than referring to George Bush as a miserable failure in a conversation, stump speech like Gephardt did, or newspaper article. Nor is it a devious, shady search engine optimization scheme that would violate Google’s own rules for inclusion in the index. To imply that it is is unfair and deceitful.
The traffic level of a site has nothing to do with Google bombing, or Google’s results. Google, and any other search engine, does not know how many people visit your site. (1) Google’s innovation in search was to focus on the links that point to a site, and determine “authority” - which they call PageRank - and use that to help rank results. It’s recursive - “good” pages are the pages that are linked to by other “good” pages. That’s an oversimplification, and I certainly don’t know exactly how Google’s algorithms work, but regardless, the important thing is the PageRank - not traffic. This is explained later in the article, so I’m unsure why it’s wrong here.
"The Liberty Round Table, a libertarian group, started a Google bomb that linked the Center for Science in the Public Interest, a nutrition advocacy group, with the term "food Nazis." (As a follow-up, the group is trying to make the Internal Revenue Service site the No. 1 Google result for the term "organized crime.") Other recent Google bombs have sought to associate President Bush, Senator Clinton and Senator Rick Santorum, a Pennsylvania Republican, with various unprintable phrases."
I’m not sure what these unprintable phrases are, since I don’t read “political” hate-mongering weblogs, which is likely where these attempts came from. However, if the Rick Santorum one is a reference to santorum - the frothy mix of lube and fecal matter that is sometimes the byproduct of anal sex - I’m not sure what part of that is unprintable. Is ‘lube’ unprintable? (2)
"Some Google bombs may have been accomplished with as few as 20 links. What is important is not the number of links, but rather the popularity of the sites doing the linking and the relative obscurity of the search term."
I think the author is referring to my original “talentless hack” Google bomb here, as we discussed how many links were required originally, and I think I said I thought it was around 20, but it was hard to know since I didn’t think it would ever succeed and didn’t pay much attention to it until months later. So 20 may or may not be accurate, and I made that caveat in the interview, but regardless, it was years ago, and I think the more recent Google bombings had many, many more links than that before success. But since it’s not attributed to me, or to somebody else’s Google bomb, or to anybody else, or backed up with any evidence, it’s merely a guess, rather than the statement of fact that it purports to be.
"The first Google bomb exploded in the fall of 1999, when a search for the term 'more evil than Satan himself' returned Microsoft's home page as the first result. At the time, Google denied that its search algorithm had been a victim of a prank. Rather, the company insisted, the ranking was an accurate reflection of the Web's many Microsoft critics referring to the company, independently of one another, as being more evil than Satan himself. But subsequent bombs made it clear that the Microsoft result was probably no accident."
My understanding of the 1999 Microsoft-satan result (and I could be wrong) was that it was random - a few pages had independently used that phrase to link to Microsoft. I’ve never read anything that suggests a single person or group encouraged people to use that phrase. This is different from Google bombing. Normal people, independently creating links to sites is way the web works normally. Google bombing is a concerted effort by people to link to things with the same link text. It is not the case that one day 100 people woke up and decided to link to Andy Pressman’s site with “talentless hack.” It was my idea, and then I coordinated with other people and they also linked. Since I coined the term, I feel some measure of confidence in this definition.
"Adam Mathes, a blogger and computer science major at Stanford, is generally credited with having coined the term "Google bombing" almost three years ago to describe the practice of manipulating Google results through seeding the Web with links. Mr. Mathes started a Google bomb as a joke at the expense of a friend and graphic artist, Andy Pressman, managing to get Mr. Pressman's blog listed as the first result for the phrase "talentless hack." Mr. Mathes later interviewed for a job at Google and felt compelled to confess his campaign."
"It was definitely a big thing for them," Mr. Mathes said. "They told me, 'Yes, we've had many meetings about Google bombing.' I don't think that's why I didn't get the job, but it probably wasn't the best career move."
First of all, I specifically asked Tom McNichol not to refer to me as a “blogger.” Apparently this meant he was going to refer to me as a “blogger.” It’s also unclear that I was formerly a computer science major at Stanford, and not currently one. The “generally credited with having coined the term” is accurate, but I’m not sure why the New York Times can’t just say I coined the term. While there may be a valid case that I didn’t “invent” the technique (3) there really isn’t any dispute over my invention of the term “Google bombing.”
As I said in the interview with Tom, it was curiosity that compelled me to ask my interviewers from Google about Google bombing. I wanted to know if this had actually been a big deal internally, or was merely shrugged off as an innocuous stunt. I would hardly call it “confessing” - at the time a Google search for my name returned a BBC article on Google bombing in the top five results. It was not exactly a secret, and surely would have been found anyway. I also mentioned that I felt like it would have been too dishonest not to mention it, but the article leaves out all the context around that quote that makes me not seem like quite an idiot. (Or maybe it’s impossible to convey the story without seeming like an idiot.)
If it’s not clear, I’m a little upset that out of the 45 minute interview, the only part that was included was the part where I don’t get a job at Google.
Finally, the end of the article baffles me.
"Google maintains that such activity still is not hurting the overall quality of its service. The company says it expects Google bombing will soon go the way of most Web fads.
'It's the kind of thing people enjoy doing once because it's fun to be able to put up a page that can have a powerful effect,' Mr. Silverstein said. 'But it's not something people are going to want to spend their lives doing.'
Clearly, anyone who goes through life trying to manipulate search engine results would have to be called a miserable failure. And how many of them can there be?"
So, even though the entire article is about how this has been going on for years, and as it is stated in the article, “The company’s success, to a large extent, has been built on its search algorithm’s ability to return relevant Web pages and weed out irrelevant or outright bogus results,” the article concludes by giving Google the last word saying it doesn’t really matter. Wonderful. And I don’t think the last paragraph is funny, or clever, and just seems to sort of be mean. But maybe I’m just in a humorless mood today.
This is not entirely true anymore, if you use Google’s advertising program to put ads on your pages, then they do have detailed information on a site’s traffic, but my understanding is this information is not used in determining page relevance. If it were used, it would favor the relatively small portion of the web that uses their advertising program, and Google has repeatedly said that buying advertising will not enhance your search ranking.
Ok, I know, this isn’t an actual error in the story, and I find Dan Savage’s “spreading santorum” crusade a little disturbing - is associating a homophobic senator’s last name with anal sex really helping gay rights? - but I’m including it anyway.
I’d argue that I did invent the technique, since I was encouraging legitimate sites to use link text to raise the ranking for a site without those keywords, but there may be some search engine optimizer that used legitimate sites and link text that I don’t know about for the same purpose.