Month: April 2009

The other day, I attended a launch party where Jeff Jarvis talked about his now best-selling book, “What Would Google Do?” I came back and assembled my reactions in a post entitled “What Would Google Do? / What Does Google Do?“. One of my strongest objections to Jarvis’s shtick is that he extols transparency as a particularly “Googley” attribute.

But now, in an interview with Steve Rubel from Micro Persuasion, Jarvis seems to accept that this really isn’t the case:

Mr. Rubel: You also talk a lot about transparency. Google, however, isn’t the most transparent company. What does the ad industry need to change here?

Mr. Jarvis: Google is not perfect. It expects us all to be transparent — so we can be found in search, so we can benefit from our Googlejuice. But Google is not sufficiently transparent about its ad splits or its Google News sources. So, as our parents would say, this may be a case of doing what Google says more than what it does.

Amen! I’m not saying Google is evil, or trying to incite a round of accusing Google of not living up to its ideals. But I am glad to see Jarvis coming clean on the point I found most objectionable in his presentation. Now I’m at least open to finding out if I should actually get the book and read it.

I wandered over to Techmeme a bit late today and was greeted with a blaring headline from Twitter co-founder Biz Stone: “The Discovery Engine Is Coming “. Twitter, search, and discovery–that’s enough to earn a click-through from me! But I’m not quite sure what they have in mind.

They say that their integrated search will look something like this 2008 sketch:

Hmm. That still feels like their current search to me, only with a different layout. Am I missing something here? When I hear “discovery”, I expect something a bit more exploratory. What I’m reading leads me to believe that they’re working on tightening up the interface, but that they’re not making any significant changes to the functionality. I don’t mean to sound ungrateful about improvements to a service I’m not even paying for, but I am a bit disappointed.

Is there more to this? After all, the blogosphere seems really excited about it, and I’d like to be too. Can anyone help me out?

It wasn’t so long ago that Cuil launched and I offered a pretty cold appraisal. Of course, I wasn’t the only one to be underwhelmed after they under-delivered on their pre-launch hype. You’d think start-ups would have figured out by now how to better manage expectations. After playing with the site for a couple of days last summer, I wrote Cuil off as dead on arrival.

But, to my surprise, Cuil did not just roll over and die. In fact, it has improved a lot while I–and most of the world–ignored it. And, just the other day, Stephen Arnold wrote a post entitled “Cuil.com Gets Better” that inspired me to take another look, even before seeing today’s post on Search Engine Land about their new timelines feature.

Initial reaction: the folks at Cuil have definitely stepped up their game. But, compared to when they launched a year ago, I’d say the bar has gone up too. I’d compare Cuil to Kosmix, as they seem to  have similar interface goals. Both have warts; I’ll need to play with both for a while to decide which I like better.

Here are two queries that I tried, inspired by Arnold’s post at the Beyond Search blog:

• http://www.cuil.com/search?q=beyond+search
• http://www.kosmix.com/topic/beyond_search

From an architecture point of view, it may not be fair to compare Cuil, which does its own crawling and indexing, to Kosmix, which federates results from other services. But they seem to be aiming for similar experiences, which is all that users care about. Though one advantage Cuil has is that, because it uses its own index, it is much, much faster than Kosmix.

In any case, I’m happy to see more attention going to exploratory search, and I am glad that a few companies are bold enough to try to make it work on the web. Perhaps Microsoft will try too, with Jan Pedersen and Stefan Weitz tilting Kumo-emblazoned banners at the grand challenges cited by their co-workers. These are early efforts, but they are promising.

And, to anyone whose technology I’ve criticized harshly in the past, please see posts like these as a sign that I do notice improvements, and that I am more than willing to reconsider my assessments in the face of new evidence.