Transcript of talk delivered by Sergey Brio at UC Berkley in 2007.
`
`Watch the video at http://www.youtube.com/watch?v=Ka9IwHNvkfU
`
`Interesting quotes bolded.
`
`0:13 MARTI HEARST: We have a fantastic guest sp€:alcer
`0:15 today, Sergey Brio, who is a co-founder of Google.
`0:19 You might have heard of this company.
`0:20 [LAUGHTER]
`0:23 I actually got the Wikipedia article on you, Sergey, in
`0:26 order to give your history. So I could sit here and
`0:27 read things for a couple of minutes but I don't think
`0:31
`I'm going to do that.
`0:32 SERGEY BRIN: I should sit down and do that.
`0:33 MARTI HEARST: Sorry?
`0:33 SBRGEY BRIN: I should sit down that.
`0:36 MARTI HEARST: Yeah, why don't you do that?
`0:38 That'd be great I'm not going to sit here
`0:39 and take up the time;_ We don't have that much time
`0:43 with you so· I'm just going to let you take over.
`0:45 SERGEY BRIN:· OK.
`0:46 MARTI HEARST: So, Sergey Brin.
`0:47 [APPLAUSE]
`0:55 SBRGEY BRIN: So I mostly want to do some Q&A here today,
`0:58 but I wanted to start with a few opening thoughts.
`1:02 And actually you reminded me of o~e of them which
`1:06 is the Wikipedia -- Wlkipedia in general.
`1:10 There are things out there that are very simple and
`1:17 you never think would work.
`1:19 And that's why you just don't do things that you assume
`1:23 they basically won't work.
`1:25 Wikipedia is one of those that it would never occurred to me
`1:29 that something like that would work.
`1:30 And I assume many of you- has everyone here seen
`1 :33 Wikipedia articles? All right.
`1:36 Yeah and it's amazing to think that you can build
`1 :39 an encyclopedia and anyone can edit anytime.
`1:42 I've. gone to Wikip¢ia pages at first when I said, look I don't
`1:46 believe they're getting this content thi.s way.
`1:48 Here, I'll hit the edit button and see what happens.
`1:51
`I go on a random web page:
`1:53 I don't know, if was some artist, 18th century.
`1:55 And I made some stuff up.
`1:58 He really. liked the colors brown and orange,
`1:59 something like that. And I punched it in there.
`2:02 And I said, come on, there's no way this is going to work.
`2:04 And of course, I click submit' and then I view and
`2:06 there's the change there.
`
`EXHIBIT 2032
`Facebook, Inc. et al.
`v.
`Software Rights Archive, LLC
`CASE IPR2013-00479
`
`

`

`2:08 And then I quickly undid it.
`2:11
`I didn't want to pollute it.
`2:13
`2:14 But it does work.
`2:16 And it works for several reasons, many that I
`2:21 don't understand for sure.
`2:23 But one of them is scale.
`2:26 And by virtue of the fact that there are so many people out
`2:30 there that are reading these Wikipedia entries, that are
`2:35 editing-- well, there are a smaller number editing them.
`2:38 And then a still smaller number that really actively
`2:42 monitors all of them. But still, it's a small
`2:44 fraction of a huge number of people.
`2:47 They're able to keep it to be a pretty-- a very comprehensive,
`2:53 reasonably high quality site.
`2:54 Occasionally like some of the stuff! think above
`2:57 me is a little bit wrong.
`2:58
`3:00 But I don't know how it would compare to like normal
`3:03 encyclopedia entries.
`3 :05 I know it not. So I think that they
`3:07 do very, very well and I'm very impressed.
`3:14
`3:16 With internet search as a whole, forget about
`3:21 Google for a second.
`3 :23 That too, which today we take for granted in a sense.
`3:27 But it was a fairly simple idea that you take all the
`3:30 information out there which let's say 12 years ago when the
`3:35 first search engine start being developed wasn't that much.
`3:38 But the computers were a lot less hefty then too.
`3:43 And you just create an index.
`3:46 Even a fairly basic inverted index.
`3:49 In fact, in the earliest days, people didn't really
`3:51 worry about ranking even. It wasn't that big a deal.
`3:54 There weren't that many matches for most searches.
`3:57 And Alta Vista probably made the biggest leap in terms
`4:01 of comprehensiveness and speed and what not.
`4:07 And you just index it and you let everybody query it.
`4:09 And today it's just it's very-- we all take it for granted.
`4: 14 But this was just a short time ago.
`4:17 And it wasn't at all obvious that it would work, that
`4:19 it would be useful or anything like that.
`4:23 And I would extend the same idea to the web as a whole.
`4:29 They were a number of hypertext experiments and systems
`4:33
`that people put up. What was the one with
`4:35 a funny guy, Xanadu?
`4:37 Did you cover that?
`4:40 Yes, Ted Nelson.
`4:41
`
`

`

`4:43 Who's a very interesting guy. But anyway, so he had created
`this thing and it wasn't the quite the same as the web
`4:46
`4:48 but it was-- people have tried that.
`4:52 And yet, with a few simple ideas-- and I won't pretend
`4:56 to know to identify the key features tl}.at really allowed
`the web to grow-- but it's really became a repository
`4:59
`5:04 of the world's knowledge.
`5:07 So anyway, I guess I want to finish that intro just with the
`5:15 point that people who have taken fairly simple ideas, ones
`5:19 which you might not think would work at all really, at a
`5:22 certain scale and after they gain a certain amount of
`5:27 momentum, they can really take off and work.
`5:29 And that's really an amazing thing.
`5:32
`5:34 Let's see, maybe I should try to relate that to
`5:37 Google a little bit. I want to leave time to--
`5:45 At Google we had one simple idea which now seems obvious.
`5:50 But the idea that the ranking does matter.
`5:54 And in fact that was not a high priority in a lot of
`5:58 information retrieval web search research at the time.
`6:03 That the ranking is-- I mean some people worked
`6:06 on it, but it wasn't that important of a thing.
`6:10 And we decided that for queries that really return a lot of
`6:12 results that we could do something more reasonable.
`6: 15 And we sort of stumbled upon a way to do that
`6:19 by studying links.
`6:20 And I don't know if any of you have-- what have we presented
`6:23 here in this class to date?
`6:25 MARTI HEARST: [INAUDIBLE]
`6:29 SERGEY BRIN: OK. So you've covered
`6:30 a lot of stuff. And page rank?
`6:32 Have you?
`6:33
`6:34 OK. I'll go through in a high level.
`6:37
`6:39 We originally developed page rank-- well, I was kind of
`6:41 playing around with studying all the links on the web.
`6:45 And that too was a pretty revolutionary idea though it
`6:51 seems very simple that you could even just collect them
`6:54 and then do anything meaningful.
`6:57 Because as a graph in the computer science sense it was
`7:03 a very large graph compared to computers ofthe time.
`7:05 Or at least compared to our budget of
`7:07 computers at the time.
`7:09 And anyhow, I really credit Larry pursuing that idea
`7:13 that it's even worth collecting the graph.
`7: 17 And then that you could run any kind of processing on it.
`7:21 But soon after we had it, and we had a crawler that went out,
`7:25 and we have to kind of develop our own RAID to be able
`
`

`

`to write to the data to the disk fast enough.
`7:28
`7:30 And it's kind of things that are trivial today, even
`7:34 probably on your laptops, but were hard back then.
`7:38 And then they started to play with it and came up to the
`7:43 notion that not all web pages are created equal.
`7:46 People are but not web pages and some web pages are
`7:51
`inherently not worse than others but at least less
`7:54
`important than others.
`7:56 And we developed this analysis of a graph oflink structure
`8:02 of the web that imputed an importance for every web page.
`8:06 And we use a similar algorithm today.
`8:09 There are many other algorithms that we have to run.
`8:11 And it's evolved a bit over the years.
`8:14
`8: 15 But it is one of the things that we continue to use.
`8:19 And the general concept that not all web pages are created
`8:23 equal is very important.
`8:25
`8:27 The other thing I want to highlight is that when we were
`_ 8:31 studying this, and we actually weren't sure that we wanted to
`8:34 have search as the big application, at some point we
`8:39 realized that this actually worked really well for a search.
`8:43 That if you type Berkeley-- there are a lot of pages that
`8:46 mention Berkeley-- but some like the Berkeley homepage
`8:49 are probably somewhat more important than others.
`8:52 And I guess there's a UC Berkeley homepage and a
`8:55 Berkeley city homepage.
`8:57 Anyway, all the Berkeley pages.
`8:58
`9:00 And we decided that was actually very useful to search.
`9:05 When you had a lot of results and that if you wanted
`9:07 ranking to matter, that was a good way to do it.
`9:10
`9:11 But the other thing we were kind of thinking about at the
`9:13
`time is how would you-- we weren't kind of thinking of
`9:17 this as how would we let millions or hundreds of
`9:19 millions of people use this.
`9:21
`9:22 But how would you even make something anyone, a single
`9:25 person, could use or how could you make a search
`9:27
`that would work well.
`9:29 We had a phrase for it: search for kings.
`9:32
`9:34 No, you're not searching for kings but a search that a
`9:38 king would use or queen.
`9:41
`9:42 But the point was, is given the resources that we had, how
`9:48 would we create really good search engine, not worry about
`9:50 how many searches it could handle or how large a user base
`
`

`

`it could support, but to make something really, really good
`9:53
`9:57 for a small number of people.
`It wasn't that we wanted to make something good for a small
`10:00
`10:02 number of people particularly. But we wanted to get rid of
`10:05
`that constraint that you had to scale it up to a large
`10:08 number of searches.
`10:10
`10:11 But ultimately what we developed we
`10:13 were able to scale. And in fact in subsequent years
`10:17 as a company at Google when we've had sort of projects
`10:21 which say, well throw as much compute power at it as you want.
`10:25 Let's say we just want this to work well for a
`10:26 small number of people.
`10:28
`10:29 We've ultimately always found ways to scale it up
`10:32 and deliver it to everyone.
`10:34
`10:35 Which is kind of interesting. It's kind oflike technology
`10:38 as an inherent democratizer.
`10:40
`10:41 Because based on the evolution of hardware, probably more
`importantly the evolution of algorithms and the system
`10:48
`10:51 software that supports these, you're able to scale sort of
`10:56 almost anything you can think ofup.
`10:58
`10:59 Now it takes-- it's not trivial.
`It takes some hard work and effort.
`11 :03
`11 :05 But I think that's an interesting observation that
`11 :08
`it's-- we'll have to see if in our lifetime if that means
`11 : 15 everybody has more or less tools that are equal power.
`11 :22 And there's not much way that you can really spend a lot more
`11 :27 for the search and get much better results because in
`11:30 a short period of time technologists are able to make
`11:33
`it work better for everyone.
`11:35
`11:36 So anyhow, that said, !just wanted to quickly go over a
`11 :40
`little background and open it up to some questions.
`11:42 MARTI HEARST: [INAUDIBLE]
`11:46 SERGEY BRIN: Yeah.
`11:47
`11 :49 Oh, thank you.
`11 :54 AUDIENCE: Hi. I'm curious what keeps you
`11:56 up most at night from a competitive standpoint?
`12:01
`12:03 SERGEY BRIN: Well I found over the years-- if! may say that
`though we've only had a company for seven years or so now--
`12:05
`12:11
`it's important that those sorts of things not keep
`12:13 you up at night.
`12:15 I mean, we obviously have big competitors.
`
`

`

`12:17 And you may have heard of some them: Micro ...
`12:20
`12:22 soft?
`12:22 [LAUGHTER]
`12:25 SERGEY BRIN: And anyways, some others nearby.
`12:27 But you really want to make sure you don't-- that's not
`12:35 what you should be spending most of your time
`12:37
`thinking about. Though inevitably in a company
`that happens and lot of people worry about this company doing
`12:39
`12:43
`this thing or the other.
`12:45
`12:47 What I think is more valuable to spend time thinking of is
`12:50
`that we have this really incredible opportunity and we
`12:55 have a lot of computational resources, a lot of great
`13:00 people that worked there.
`13:02 We have the ability to analyze lots of data.
`13 :05 And there are a terrific products and services
`13:09
`that come out of it.
`13:10
`13:12 I'll just-throw out one example: we developed
`13: 15
`this advertising network.
`13: 1 7 And a real issue for the web is you can't find something if
`13:22 it's not there, even though it's very comprehensive.
`13 :25 And the high quality sites, they do have economic issues.
`13 :31 Like they have to stay in business.
`13:33 And a lot of sites during the boom, the bust ofthe boom,
`13:3 7 went out of business. And we did have-- we wanted
`to make sure people could get their content out there and
`13:42
`13:45 could afford to pay for it.
`13:46 So we develop this program called Adsense which
`13:50 allows us to put ads. And you probably see them kind
`13:52 of on various parts of the web now, not just on search
`13:55 sites but on other sites.
`13:56 Sometimes you might even annoyed by them.
`13:59 But we like to think they're better than the
`14:01 kind of flashy, crazy ads.
`14:02
`14:06 Anyhow, they also create an income stream for a lot of
`14:08 people who are developing great content.
`14:10 And that's something we can do by virtue of the size of our
`14:14 advertising network, the information retrieval
`14:17 technology we have to target the ads, the scale of our
`14:22 brand, various other things.
`14:24 And that's just one example.
`14:28 But there many things that we can apply, the different things
`14:32 we have in our company, both technologies as well as reach
`14:36 of various kinds, and create new things.
`14:39 And that's exciting thing to lose sleep over.
`14:42
`
`

`

`14:44 AUDIENCE: My question is, at the time that you developed
`14:49 and released Google there were other search engines that had
`14:53 been accepted as the standard and were considered to
`14:56 work fine and so on.
`14:57 What is it that inspired you to go back to the drawing
`15:00 board, create a new search engine, and release it?
`15:02
`15:04 SERGEY BRIN: Great question. So at the time there
`15:06 were probably five major search engines or so.
`15:11 And you might not even remember some ofthem.
`15:14 I don't know. People here are pretty young.
`15:17 But what we found was we-- kind of by accident almost-- we
`15:24 found that this processing of the link structure of the web,
`15:28 we could create a search that was better in important ways.
`In ways that these search engines had ignored.
`15:33
`15:35 And also th~y had decided that the search wasn't that
`15:41
`important at the time.
`15:43 All the search engines wanted to be Yahoo!
`15:45
`15:47 which at the time, and actually until relatively recently,
`15:51 didn't have a search engine at all except one
`15:53
`that it licensed.
`15:55 So we found that it was actually-- we thought
`16:01 we'd give it a try.
`16:02
`16:04 That's another I think important thing.
`16:05 There was relatively little downside to trying.
`16:08 I mean, we thought we have something that was pretty good
`16:12 and we were able to get a little bit of funding for it.
`16:15 We didn't need huge amounts.
`16:16
`16:18 And my adviser assured me that I could give it a try and if it
`16:20 didn't pan out, could come back to the PhD program, which
`16:23 I still need to finish. I'm technically on
`leave of absence.
`16:26
`16:27
`16:29 So we gave it a go.
`16:30
`16:35 AUDIENCE: I was wondering, at the end of John Battelle's
`16:37 book on The Search, he starts discussing the project at IBM
`16:42 which is systematically creating a tagged or semantic
`index of the web and kind of approaches a very slow
`16:4 7
`16:52 but nearly perfect search experience.
`16:55 And I was wondering how do you feel about the future of
`16:59 tagging and semantics in the role of search in the future.
`17:06 SERGEY BRIN: Great question. Actually I have not finished--
`17:08 I may have seen the book but I haven't-- you guys had
`17: 11
`it as reading, I guess?
`17:12 Well you can tell me about it.
`
`

`

`17:13
`17:16 But I spoke to him at length about it so hopefully some
`17:18 of that was in there.
`17:21
`I think that the tagging and semantics are great as long
`17:25 as the computers are doing the tagging and semantics.
`17:29 Because if people are doing the tagging and semantics for the
`17:33 computers then there's something a little bit inverted
`17:36 about the relationship with man and machine there.
`17:39 So I'm a big believer of creating lots of innovative
`17:45 algorithms that can extract this kind of structure
`17:48 knowledge from lots of the text that's out there and created
`17:52 by people all the time.
`17:53
`17:54 But I'm not a big believer that you're just going to have lots
`17:5 8 of people that enter the data very carefully so machines
`18:01 can then process it.
`18:02
`18:06 AUDIENCE: I was wondering if you can give a brief overview
`18:08 of your China strategy.
`18: 10 Also from the standpoint of technical barriers as well
`18: 14 as business and governmental barriers that you foresee?
`18:17
`18:19 SERGEY BRIN: That's a good question.
`18:19 Yeah in China there are some technical challenges.
`18:26 In particular the CJK languages as we call them: Chinese,
`18:29 Japanese, Korean.
`18:32 They're double byte. They don't have sort of
`18:33
`the same idea of words and sentences and what not.
`18:36 And anyway, there's some special technology that
`18:42 goes into making the Chinese search work well.
`18:46 But we've been developing that pretty well and I'm pretty
`18:49 comfortable with that.
`18:51 There are a number of complicated government
`18:54 relations issues.
`18:56 And anyway there are a lot of trade issues.
`19:01 There are a lot of policy questions that haven't really
`19:05 played out I think as of yet.
`19:07 And I think that all I can know is that I don't think we can
`19:12 predict today exactly what will happen with China in the future.
`19:15 The approach that we have taken is to make sure that we stay in
`19:22 touch with both of the representatives of various
`19:27 people knowledgeable about China both in China and out,
`19:31 as well as human rights organizations which care about
`19:35 some of the censorship questions and things like that.
`19:38 And what we've deduced out of those conversations with both
`is that our participation in the Chinese market is probably
`19:41
`19:47 positive, not just for us as a business to participate in
`19:51 China but also for the people having access to more
`19:56 information, more kinds of communication services,
`
`

`

`19:59 and things like that.
`20:01
`20:02 And in some cases, just like we do in Germany and the U.S.,
`20:08 we comply with laws which we may not necessarily support.
`20:12 But we do need to be law abiding.
`20:14
`20:15 So I hope that answers the questions.
`20:17
`20:43 AUDIENCE: Both the background from academia as well as your
`20:45 current opportunity having access to resources,
`20:49
`infrastructure, and data.
`I'm wondering your views in terms of doing research, coming
`20:51
`20:56 up with great ideas in the corporate environment versus
`21 :00 say in the university.
`21 :02 And what opportunities exist today you think in university
`21:07 research within the context of search or marginally in terms
`21 : 11 of distributive systems design?
`21:12
`21:13 And what might some things Google could do to perhaps
`21 :20 enhance-- make innovation happen in both big corporate
`21:27 environments as well as academic research?
`21:29
`21 :30 SERGEY BRIN: Great question. I've had some experience now
`21 :34 with both academic research as well as within our company.
`21:39 I think that you can-- we have a lot ofuseful assets and I
`21:46 guess technologies, things like that, a Google.
`21:50
`21 :52 And I gave the-- I illustrated with one example.
`21:55 You can look at other things.
`21 :58 Google maps takes a lot of computers and some
`22:01 data that we license. Things like that.
`22:03
`22:04 But that said, there are plenty of interesting things were you
`22:06 don't need any of those things that we have.
`22:08 Like you might-- especially with today's computers, you
`22:11 could easily get by with just your laptop perhaps.
`22:15 I've certainly done experiments even while with Google
`22: 18 where I didn't use any more computer power than that.
`22:21 Or maybe we'll use a kind of small cluster of machines.
`22:24 You probably have things like that around here.
`22:27
`22:28 I think there are many interesting things that can be
`22:30 at the very least tested out without the particular kinds
`22:35 of resources that we have.
`22:38 There's some things that can't.
`22:40 Which is why it's a good idea for us to focus
`22:42 on those problems.
`22:44 And in some cases we've actually collaborated
`22:46 with researchers in different places.
`
`

`

`22:50 For example, we were just talking to the.Venter
`22:54 Institute-- I don't know if you're familiar, they do
`22:57 DNA sequencing-- about using some of our compute farms
`23:03
`to run their processing.
`23:05 And we've run experiments with that.
`23:07 And other kinds of things.
`23:11
`23:12 There is always a cost, both to us as well as to the
`23: 18 researcher of that infraction.
`23 :21 There's communications, there's overhead.
`23 :24 It does take some effort.
`23:26 So we try to apply it in the areas which appear the most
`23:31 promising at the time.
`23:34 At the same time, we also want to make public a lot of-- we
`23:3 8 actually publish a fair amount about our systems designs,
`23:41
`the algorithms that we use.
`23:44 We also have published some software tools and some of our
`23:50 libraries and things like that.
`23:52 It's still a fairly small fraction.
`23:54 But that's the direction we found to be helpful.
`23:57
`24:09 AUDIENCE: [INAUDIBLE]
`24:09 SERGEYBRIN: We can definitely hear you.
`24:10 AUDIENCE: [INAUDIBLE]
`24:38 SERGEY BRIN: I don't know.
`24:39 I guess I feel that human language is
`24:44 pretty well evolved. And it's certainly not perfect
`24:48 but attempts create better ones like Esperanto haven't
`24:51 been that successful.
`24:53
`24:54 So I think that professional researchers and authors-- I'd
`25 :00 rather they spend more of their time researching and writing.
`25:06 And I think it's our task as well as all of your tasks to
`25:12 tum those words into semantics which could be used for
`25:16 a variety of purposes.
`25:19 I just don't think it's good use ofhuman effort if you
`25:21 kind of compound all the hours out there.
`25:24 I think the AI type of problems are actually very solvable.
`25:28 So that's my view.
`"25:30
`25:35 AUDIENCE: I have a question on the-- to follow-up
`25:37 the China question.
`25:38
`25:39 So Google's mission is to organized the world's
`25:43
`information and make it universally available.
`25:46 So the fact that Google is censoring, does it violate
`25:51
`this mission in not making it universally available?
`25:54
`25:55 SERGEY BRIN: So, excellent question.
`
`

`

`25:57 First I'll disclude the premise. We're actually are
`25:59 not censoring. In fact, our search is the only
`26:02 search engine I know of-- the only major one operating in
`26:06 China where we do no filtering.
`26:10 There are some properties such as news.google--
`26:14 I forget the code--.
`26:15 There are some other sites-- I guess-- I don't remember the
`26:20 exact URL where we've had to have tailored experience.
`26:23 But the web search is actually not censored by us.
`26:26 It is, however, censored by the Chinese firewall itself where
`26:31
`they're able to take out certain searches at the border.
`26:34 And there's nothing we can do.
`26:37 So I think we've taken a pretty good stand there.
`26:40 But in general with respect your question in terms of
`26:44 making information universally accessible and useful,
`26:50
`it's not very useful if it's not there at all.
`26:53 And in fact we have had our side shut down
`26:56 multiple times, altogether blocked out of China.
`In fact had the news site shut down in China and Blogger.
`26:58
`27:02 And things kind of go up and down.
`27:05 And we have to make a judgment call combined with people
`27:0·9 who really care about the political situation there
`27:13 and human rights there.
`27:14 And I've spent a lot of time talking to.
`27:17
`27:18 We have to use our judgment to sort what the best
`27:20 outcome will be. We've been doing that.
`27:23
`27:25 AUDIENCE: Baidu? Baidu, the Chinese
`27:28 search engine. Do you guys think it's
`27:31 a big competitor?
`27:33 SERGEY BRIN: Well, it's been a good investment so far.
`27:35 Yeah we actually invested in Baidu, I forgot, a
`27:37 couple of years ago or so. And that investment has
`27:40 done very well from a financial point of view.
`27:43
`27:44 But in general, in this space we have-- there a lot of
`27:47 companies that compete with us in different local geographies.
`27:50 In China there are some.
`In russia actually there-- Russia and Korea have ones with
`27:51
`27:56 much more market share than us.
`27:58 That's compared to China where we have pretty
`28:00 good market share too.
`28:01
`28:03 I don't think we fear them. They were actually-- I mean,
`28:05 I know the founders ofBaidu and I know the founders of
`28:07 most of these companies.
`28:09
`28:10 It's a pretty good relationship but obviously we do compete.
`
`

`

`28:13
`28:15 AUDIENCE: My question is also about China--
`28:17 Google China strategy.
`28:19 So I know most of the company, they face two big challenges
`28:21 when entering China market.
`28:24 One is to find the right person.
`28:26 So it sounds like Google did a great job in this side.
`28:29 The second biggest challenge is to figure out what to change to
`28:35
`fit the local market and what to maintain to keep consistency
`28:40 as a whole company.
`28:41 So what should Google do from the product perspective to
`28:45 adjust the second challenge.
`28:47
`28:49 SERGEY BRIN: That's a great question.
`I think there-- well, I think many of you are well-versed
`28:51
`28:54 and I should get your input on this question.
`28:57 But yes we have been hiring in China.
`28:59 We're very excited to have Kai-Fu Lee join us.
`29:02 Who, I don't know if any of you saw in the press, we
`29:04 had a little bit of a legal skirmish about that.
`29:08
`29:10 But he's on board now and busily hiring all
`29:13
`the other people who are going to help him.
`29:16 And most markets I find that it's actually sort of 90%
`29:23
`is the same as the U.S.
`29:26 and other markets were serve in 10% has to be unique.
`29:30 China's probably a little more towards the 70-30 mark
`29:33 or something like that.
`29:34 And I think we're going to have to deal with
`29:35 a number of issues.
`29:37
`29:41 Just for example, the connectivity issues
`29:44
`there-- the network.
`29:46 We found that most university students have to pay or
`29:49 couldn't access Google at all because you have these-- you
`29:52 pay these fees for going outside the country like 10
`29:56 cents a megabyte or something. I don't remember what it was.
`29:58 But that's a lot of money.
`30:00 It's even a lot of money for here. And to a Chinese
`30:03 university student, it was just unaffordable.
`30:07 And so we've had to work on our networking situation.
`30:12 And on top of that, there's also latency issues,
`30:14 things like that.
`30:15
`30:17 And these things aren't true in-- well certainly
`30:20 the charging it's true no place that I'm aware of.
`30:23
`30:25 A lot of the unique kind oflicenses you need to get,
`30:28
`things like that just aren't true in other
`
`

`

`30:29 parts of the world.
`30:31 So yeah we would have to address that.
`30:35 There are other markets that have their uniquenesses.
`30:37 Korea is very unusual in a lot ofways.
`30:42 Far more broadband adoption.
`30:44 And a lot of things about the end-user behavior
`30:48 we don't understand.
`30:49
`3 0:5 0 So I think all these things are going to take work.
`30:53 I would say on average though 90-10 might be typical.
`30:58 Maybe based on these other countries that gets
`31:00 pulled down to 80-20.
`31:01 80% of the work we do is the same across all countries.
`31 :06 And I think that's been one of the reasons why we've been
`31 :08 actually very successful internationally.
`31:1 0 Because there many markets which our competitors never
`31 : 13 paid attention to at all.
`31: 15 And we've had a pretty scalable approach.
`31:19
`31 :20 We have interfaces in well over a hundred languages right now.
`31 :24 And you can search in probably hundreds more.
`31:28 It's kind of hard to test. I don't speak most of those.
`31:30
`31:32 But I think it makes sense to make sure that you continue
`31 :40 to put the core efforts into technologies which will
`31 :43 work across the board.
`31 :46 And then don't forget that you're going to have to put
`31 :49 some specialized technologies in each market., that you're
`31:52 going to have to budget somehow.
`32:00 SERGEY BRIN: Hi Sergey. Today is actually my birthday,
`32:02 but I think I speak for all of us-- SERGEY BRIN:
`32:03 Happy Birthday.
`32:04
`32:05 AUDIENCE: -thanks. I think I speak for all of us
`32:07
`in saying I think this is a great surprise present
`32:08
`to hear you talk.
`32:09 SERGEY BRIN: Well, it's my pleasure to visit.
`32:11 AUDIENCE: My question is about the recent announcement of
`32:14 partnering with NASA and the implications of that in
`32:18
`terms ofbranching out into different technology areas.
`32:22
`32:24 SERGEY BRIN: Yeah, well the NASA announcement's probably a
`little-- actually even a small portion of kinds of things we
`32:28
`32:34 do to collaborate with NASA and other research institutions.
`32:38
`3 2:3 9 NASA for a long time at Ames ran one of the largest
`internet connection points. NASA Ames-- I'm not sure
`32:43
`it's still such a big hub.
`32:48
`32:50 But they pumped a lot more bandwidth than anyone
`
`

`

`32:53 else for a long time.
`32:55 And they also have some of the largest computers.
`33:01 Just sort of the kinds we use, clusters ofPCs or
`33:05 similar kinds of things. And they've been experimenting
`33:08 with those for many years,
`33:11 Let's see, what else do they do?
`33:12
`33:14 They've had their share of research in AI and
`33:17 things like that. And we've had collaboration on
`33:19 the Google Earth-- if any ofyou've tried Google Earth?--
`33:23 with some of the NASA folks.
`33:26 In fact, Keyhole, which is the company that became
`33:28 Google Earth worked with NASA before too.
`33:32 So there's a lot that we work together in.
`33:36 Maybe someday we'll get exotic, we'll build space tethers
`33:39 and things like that.
`33:40 But that not in the-- the short term plan is just for some
`33:44 office buildings to start.
`33:46
`. 33:49 AUDIENCE: Hi. My question is regarding
`33:50 the future of Go ogle. I know Google started off as a
`33:53 search, but in the recent years they have been rolling out
`33:56 more desktop software like Google desktop search, like
`33:59 Google Earth or Picasa.
`34:01
`Is that where the company is headed towards a more desktop
`34:04 environment rather than focusing on the search?
`34:07 SERGEY BRIN: That's a good question.
`34:08 Yeah, initially we didn't have desktop software because
`34:12 we didn't really have the expertise for writing it.
`34:17 And there's some cross-platform issues and we're not-- our
`34:19 desktop software is not on as many platforms as I'd like.
`34:22
`34:25 It's also a little bit more work for users to get to
`34:27 because you actually have to download and install
`34:29 an application. But over time we've had things
`34:32 that we felt required desktop software to have a really
`34:36 great user experience.
`34:37 And you couldn't really have Google Earth in
`34:40 today's browsers. Maybe in tomorrow's browsers
`34:42 you can build Google Earth. But fortunately more and
`34:45 more, the browsers are becoming more capable.
`34:47 So we can do things like Google Maps, which is actually pretty
`34:50 close to Google Earth, and Gmail which because a pretty
`34:56 well-m featured email client just in the browser.
`35:00 I think we're going to continue to use both.
`35:04 I'm actually hoping we won't have to have as many clients
`3 5:06 because of those install issues and things like that.
`3 5: 10 But we'll do whatever it takes in order to deploy
`35:13
`the services we want.
`
`

`

`35:14 MARTI HEARST: I'm think I'm going to