Met with Jeff yesterday, and we didn’t have nearly enough time, so we’re going to meet again next week. But the time we did get was quite interesting. This against the backdrop, of course, of Yahoo’s stated intent to shift from Google results to its own native technology. Jeff was coy on when the switch would occur, but extremely enthusiastic about the end result (Yahoo CEO Semel has announced the switch will occur by the end of the first quarter, and that’s not too far away).
I think Yahoo search will be new from the ground up. It’s not just Inktomi in place of Google, it will be an entirely new product. Jeff wouldn’t give me details on what to expect, but he is a man clearly sitting on his hands – he’s proud of the work his team has done. “People don’t realize how scarce search engineering talent is,” he told me. “And we’ve got critical mass.”
On more general topics, we had a robust discussion around the issue of paid inclusion. This issue is almost always painted in black and white – Paid Inclusion Bad, “Pure” Search Good. But Weiner defended the practice against the metric of user value – when sites pay to insure their content is indexed, they also insure it will be available as potentially relevant results to the user. If Yahoo fails to give the user relevant, quality results, and instead spams the user with commercial fare, Yahoo will lose that user. In other words, it’s not in Yahoo’s best interest to value the advertiser over the user’s needs. In fact, it’s in the advertiser’s interest for Yahoo to value the user over the advertiser. This, of course, is Publishing 101.
What Weiner seemed to be saying was: don’t judge us by the past practices of Inktomi, and certainly not by how MSN implements Inktomi. We’re going to do it in a way that delivers value to the searcher.
I get the sense (and it’s just that, a guess, as I have no direct facts yet on what the next rev of Yahoo Search will look like) that in many cases, Yahoo search won’t initially show a list of algorithmic results to the user, but will instead show what the site *thinks* the user is trying to get at, and go from there. This raises a larger question about basic approaches to search results. Yahoo seems far more comfortable making explicit editorial decisions in its approach – intuiting the intent behind a particular search, and delivering results from any number of sources – its commercial deals, its directory, its algorithmic web results, etc. Google, on the other hand, continues, for the most part, to maintain a “purist” approach to search, claiming that its secret algorithmic sauce will deliver the most relevant results regardless of editorial judgment.
I asked Weiner about this idea of “purity” – a concept which is a clear differentiator for Google. Staying on message, Weiner said that the issue is not “purity,” but rather “value.” As in, which site will give the user the most value for the search query. He clearly believes Yahoo can play on that metric. By the end of March or even sooner, we’ll all have a chance to find out.