I want my Interface Agents, right now

At the risk of getting too Google-heavy in my posts here…

What the hell. You have to pay attention to the 800-pound Googrilla, even when they are just speculating. So Marissa Mayer, Google VP for “Search Products and User Experience” has posted to the Google Blog her thoughts on the Future of Search. Quoth Marissa:

The face of search will change dramatically over the next 10 years. Maybe it should contain even more videos and images, maybe it should sharply differentiate the relative weight and accuracy of the results more, maybe it should be more interactive in terms of refinements? We’re not sure yet, but we do know that the one thing that the search experience can’t be – especially in the face of the online media explosion we’re currently experiencing – is stagnant.

Actually, if you read the whole thing, she’s not telling us much. But does it have something to do with how news will reach you in the future? Yes. Because that won’t be stagnant, either.

This afternoon, at a certain moment, John McCain’s campaign announced that he was pulling out of the planned Friday night presidential debate in order to attend to business in Washington. Within minutes, this news was on the Associated Press wire and atop the front pages of the New York Times and Washington Post sites.

But Google News, which is (allegedly) brought to us without human intervention by proprietary algorithms lodged deep in the bunkers of Mountain View, took at least 30 minutes to begin featuring the fairly momentous development. I suspect that’s because the machinery at Google looks for a certain amount of momentum in a story (numbers of news sources publishing it, and the speed at which it is spreading) to decide that it must be important. Yet, earthshaking developments (not that McCain’s decision was one of those) can easily be broken by one or two news organizations that will own them for a while before anyone else picks them up.

In a (future) world where news is brought to us instantaneously on portable electronic devices of some kind, we will expect someone or something to do the sorting-out for us. We don’t want to have to glean, ourselves, the stuff that’s of interest to us from a broad stream of news bulletins coming at us. We’ll need a news search function that’s tailored to our individual needs. And we’ll want it to be instantaneous, just as basic search is today—not delayed by 30 minutes, or even 3 minutes.

In a way, this brings us back to the “interface agents” posited by Patti Maes back in the dark ages of 1993, described in my first post. As envisioned at the time, these “Knowbots” might roam the Internet on your behalf, finding information of interest to you. Given computer speeds and indexing algorithms at the time, this was not seen as an instantaneous function—the internet would be huge, so your agents would need some time, maybe days, to cover all that ground. But as it turns out, of course, no matter how vast vaster the internet gets, or how broad the bandwidth of news coming at us, we want instant delivery of crucial news and information.

Moreover, Maes’s proposed agents had intelligence—they would learn, gradually, what your interests and priorities were, and taylor their deliveries accordingly. Aside from perhaps prioritizing stories according some keywords you’ve specified, that function is still somewhere past the horizon for all online news sites.

So, Marissa, back to the drawing board. The gap is down to 30 minutes, but that’s an eternity. And, along with even more speed, start building some intelligence into my Google News—it should be able to figure out, without me telling it, that I’m interested (among other things) in any kind of news about the Red Sox, the island of Texel in the Netherlands, the politics of Vermont, and deep-sea archaeology.

2 Comments

  1. News agents are here, now, and they are called “RSS”. Feed subscriptions, especially canned searches from the Times or Google News, are the closest approximation I’ve seen to Apple’s Knowledge Navigator agent concept from the late 1980’s.

    The direction I think a lot of this is going is slightly different though. An agent-based approach falls down because of its detachment from your social surroundings. People read the news to be informed members of groups, which means that what is news has to be somewhat uniform within each group. For this reason, I think we’re going to continue to see aggregators expand and replace traditional news sources.

    I followed the McCain story closely because I follow Andrew Sullivan’s dense feed of information. Andrew is not an “agent” but he is a curator and context provider for a particular strain of information. Techmeme fills this niche for a lot of web technology people. Local bloggers are starting to do so for cities like Oakland.

    You’ll subscribe to one aggregator to stay abreast of your industry, another for your hobby, and a third for your town.

    September 25, 2008
  2. Odessa said:

    People should read this.

    November 11, 2008

Comments are closed.