Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Eric Schmidt On the Future of Search (googlesystem.blogspot.com)
30 points by Uncle_Sam on Aug 16, 2010 | hide | past | favorite | 20 comments


"I actually think most people don't want Google to answer their questions. They want Google to tell them what they should be doing next."

When I do a search, I really, really want Google to find me a page that answers the question I am searching for. I think they really need to get back to improving their search.


To me, where Google stands to improve search the most is in parsing more complex ideas. Sometimes, I want to search for a specific concept that may be expressible in any number of different ways, but I don't know of (or there doesn't exist) a standardized way of describing it. All I can do is throw different expressions into the field and sift through pages of results hoping my wording may have coincided with someone else's.

I've no doubt there have been many dozens of engineers on this problem in Mountain view for years, but the last time I was truly surprised by the intuition of a Google search result was probably 2002.


I think Eric was not clear in what he was saying. It is not so much that Google should tell you want to do next but that Google should predict what you are going to do next and suggest the answers before you even have to search for them.

Of course I am not Eric so take this with a few grains of salt.


For the first time I'm thinking damn, I don't want to live in Googleworld anymore.


instead of telling me what to do, i'd much rather google helped me make informed choices about what to do.


Your basic disgust at the concept sounds the standard distaste people have upon learning of the Compatibilist formulation of free will[1]. In "telling you what to do," Google would just be acting as an extension of your will—accessing all the same information you would access, and making all the same choices the same way you would make them. Just because the choices of what to do would come from the external world (and thus be determined) rather than from your mind (and thus "feel" more like you are in control of them) is no reason to reject those choices: they are still yours, even if they are determined.

[1] http://en.wikipedia.org/wiki/Compatibilism_and_incompatibili...


Mmm... It's a toughie... maybe you should ask Google what you should do. ;)


I've heard the argument that people aren't free thinking enough as it is. There are many people that are willing to give up the decision of what to do now or tomorrow, maybe it absolves them of the responsibility of their actions. If Google starts suggesting folks do this or that how long will it be till people start blaming Google for their mistakes.

Sounds like a science fiction novel to me.



While I have no reason to doubt the good intentions of Eric Schmidt, I think this makes a lot more sense from Google's point of view, rather than the consumer's point of view.

Being able to suggest things you may like is in a way at the core of what advertising is. Currently Google does a pretty poor job, both because it's hard to predict human behaviour, and because they mostly use (or used?) the search patterns as a guide. But if you want to make money from ads the obvious area to improve is to tell you what you may like before you even know it.

In other words, to me this speech sounds more like an advertiser wondering how to improve the significance of its ads than a search engine wondering how to improve searching.


"We're still happy to be in search, believe me. But one idea is that more and more searches are done on your behalf without you needing to type. I actually think most people don't want Google to answer their questions. They want Google to tell them what they should be doing next."

As Google knows "roughly who you are, roughly what you care about, roughly who your friends are..."

Stop the ride. I want to get off now.


I'm sorry, Dan. I'm afraid I can't do that.

But seriously, I think he's just imagining something like a butler, who anticipates your information needs, or at least your context and where you're coming from. It's a kind of AI, as he says, moving toward do what I mean.

Unfortunately, these Frankensteinian quests have always ended up closer to Office Assistant Clippy, in practice. Historically, mental amplification has been a more fruitful approach, where the user remains in control.


I read somewhere a professor was saying that when you retweet, or vote up articles, or any other of a dozen activities in which you're simply reacting mechanically that you are no longer an individual but part of a hive mind. Sounded a little extreme, but I get his gist.

Years ago I worked on a warehousing system. Used to be that folks would get a list of stuff to pick, then go get that stuff, then get a new list. This became more and more automated. Soon the computer was talking to workers over wireless headsets, giving them step by step directions of where to go and what to do.

I imagine the next step, after "where do I go and what do I do?" will be people writing articles about how much more happy they are now that they have a clear purpose in life. How the system actually knows better at what makes them happy than they do themselves.

I liked the super-brain, immortal demigods sci-fi future much more than I do the borg one we're actually getting.

Hey kid! Get off my lawn! :)


I think I see what he means, but upvoting etc isn't necessarily mechanical. The mechanical conformity he's talking about sounds like social conformity: http://en.wikipedia.org/wiki/Asch_conformity_experiments

That warehousing system sounds like GPS car navigation systems - but this is still amplification (so far), and I doubt it can or will be extended.

I saw a reddit post about how army life can be very appealing, because of the "clear purpose in life": the matter at hand is urgent, so you don't worry about any other problems in your life, and the bigger picture is taken care of by the higher-ups.

aside: I think the ideal life, as organized by a super-brain/demigod (or, from a different point of view, what heaven would be like), would be still have ups an downs (as in the earlier matrixes/ices, people rejecting utopia), but these ups and downs would be more clearly meaningful to the person, like a composed heroic journey, without the meaningless clutter. For people who had grown more, it would include the meaningless clutter, so that part of the "ups and downs" would be to discern the meaning. ie. living in heaven would be indistinguishable from on earth.


Here's that reddit post: Great post explaining a soldier's mindset, written for redditors http://www.reddit.com/tb/d19pt


The single biggest barrier for me is questions that I could google for, but I don't think of it, because in my mind it's just a thought or need, not an explicit question.

For example, issues about cooking, while cooking. Something like a (waterproof) iPad might be one factor in the solution, perhaps with (excellent) voice recognition.

But another problem is that, when I have searched on question like this, the answers are often not there, or buried amongst a lot of inaccurate/unhelpful answers to the question - and it ends up taking an hour, and not actually answering the question. eg. how to stop a rice cooker burning the rice (ans didn't come from google: leave it 15mins after it's cooked + cook a larger amount).


Eric wants Google to be Marshall Brain's Manna?

http://marshallbrain.com/manna1.htm

(warning, the writing gets a little lame toward the end of the book)


This is why I've switched to third-party search engines, and resort to Google when I can't find a result that works.


isn't this what Apple bought when they got Siri?


A technologist is speculating on the future and possibilities of technology, it's nothing to be an alarmist about.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: