>>
CUTTS: Hi there, everybody from Germany! Whether you're a user or a Web site or a SCO,
very nice to have a chance to talk with you for a few minutes.
And hopefully, I'll get a chance to visit Germany sometime soon, but if not, we have
some really great people who are paying attention to the German market--the German Webmaster
Discussion Group passing that feedback on. And at Google, we'll take that very seriously
and try to make good use out of it, so hope you enjoy it and thanks very much.
[pause]
The very first time I ran across spam-- Whenever I started at Google, I was working
on SafeSearch, which was a porn filter--a family filter.
And we had something in there that said if somebody has enough page rank, they've got
to be clean, because who would be a porn site with tons of page rank?
And a porn site was getting through the filter. And I looked at the site, and it turns out
it was an expired domain which was linked to by the W3C, which is one of the main standards
body that the Web pays attention to. And I remember thinking, okay, link-based
metrics and page rank are fantastic, but maybe there are some ways that dedicated attackers
could try to get around that. And so it was a really big awakening to sort
of realize there are ways that people can target any search engine, and that was the
very first spam site I saw--and it happened to be a porn site.
So it really stuck in my mind more than it would otherwise.
[pause]
I think there's a lot of challenges. A big challenge is the users expect more from
search engines, so-- Udi Manber, who's one of our VPs of engineering,
makes a really interesting point, which is if you look at the queries that were really
hard even a few years ago, we don't consider them quite as hard now.
We do a pretty good job on them, which means users expect to be able to do harder queries.
So they send us queries that are incredibly difficult.
We look at them, and we'll say this is practically impossible.
How are we going to solve this query? So I think we'll continue to get better, but
a big challenge is that users will continue to ask harder and harder queries.
[pause]
I think one of the keys has to be to be creative. Whenever you first start out, you don't expect
that a small Webmaster is going to be able to compete with a huge Webmaster for a very
big phrase or something like that. So if you can target a specific niche, if
you can say, "Okay, maybe I won't go after all types of shoes.
Maybe I'll go after shoes for people with big feet."
I've seen small Webmasters go after that niche. Then you can really target that, and once
you become known as the expert in that area, then you can sort of build out a little bit.
The other thing is that you can be a lot more creative and a lot faster as a small Webmaster
than many big sites. So I've talked to people at large sites, and
they said, "Yeah, I'm leading a multi-year project."
And I said, "Really, what's the project?" And they said, "Oh, we're trying to add one
thing along the top of all of our different Web properties."
And that's the sort of thing where it took them months to get consensus from everybody
in the organization. And so it's really nice because you as a small
Webmaster can run these experiments. You can try out new and different things,
and if they work, you can adjust really, really well very quickly.
So I think the ability to move fast, the ability to be creative and think about new techniques
that a big company might not be willing to try or might not think about trying, and then
the ability to just start with a small niche. Start with something where you can become
an authoritative resource, and then build out from there, once you become known as the
authority in that area.
[pause]
Google Webmaster Central. Yeah, that's-- [laughs]
I may be a little biased here, but it's absolutely the case that you can find out back links
for your site. You can find out the key words that people
type in whenever they're searching for your site.
You can find crawl errors. So if you have broken links and things that
aren't getting crawled, it's really nice to be able to find out, oh, I have these 404
errors. You can find out statistics like how long
is the delay whenever Googlebot tries to fetch your site.
You can see graphs of how many pages are being fetched per day.
And one of my personal favorites is you can see if you have, say, hidden text or Google
thinks that you're spamming or possibly even that your site got hacked--we will show you
a little message. We have what's known as a message center,
and so we'll show it many different languages. We can show you yes, here was some hidden
text, here was the exact URL, and then once you fix that problem, you can go and file
what's known as a reconsideration request. So there's a ton of great tools.
You can even report spam that you see on Google. It's definitely a resource I recommend everybody
to be familiar with, and it's also completely free.
[pause]
Yeah--every market is different, and I think that Google has to be mindful of that.
And as Google has gotten to be a company that a lot of people pay attention to, I think
it's important for Google to try to be as sensitive as possible and as aware as possible
of every single market. So in Germany, for example, a lot of the times,
rather than seeing a bunch of different words, you'll see people combining words.
In English, you think of glomming them all together and appending them, and you end up
with these really long words. And so you do different techniques.
You do different ways of doing spell checking, you do different ways of retrieving which
documents are going to be the most relevant in any given language.
I think it's great that Google is relatively international in terms of our Webmaster Central
console--supports a ton of different languages from German to Chinese to all sorts of other
languages. So I think we try to be mindful.
Google tries to be mindful and sensitive and think about the different market [sic] and
how to do well in each individual country. And then, hopefully, we also respond to feedback.
So I know that we have people listening to the Google Discussion Groups.
We have free discussion groups where any Webmaster can show up and say, "Hey, I really want this
feature," or "I don't like that you did this." And we pay attention to those in the German
markets just like we do in the English markets, and we get great feedback.
We have very smart people paying attention there.
So I'm really glad that we get good ideas, and then we can find suggestions for how to
improve.
[pause]
[laughs] I do. Yeah, [laughs] it's--
We like to joke around at Google that once you learn how to really see spam, it's almost
like x-ray vision. You can sort of look and guess why somebody
making these links-- The blessing and the curse of being able to
see spam that regular users might not notice is you kind of notice it everywhere you go.
Like, everything you do, every site you're on, you're sort of looking.
What is this site doing well? What is this site not doing well?
So sometimes it interferes with your enjoyment. You can't just surf around the Web and be
completely innocent. But you also get a lot of really good feedback
that way. We have a ton of users who are--who give us
spam reports or who give us feedback. And so the nice thing is even as a lot of
regular people surf around the Web, they look for ways that Google could do better, and
they'll tell us that as well.
[pause]
Yeah. One of my favorite questions is always along
the lines of I have a whole lot of different sites, and I've been cross-linking them together,
and I'm not doing as well in Google--and actually, I'm not doing as well in Yahoo!, either.
So I'm wondering what I need to do to--in order to improve how my web sites are listed.
And I was on a panel with somebody from Yahoo!, and the Yahoo! guy--he got sort of a cynical
look on his face--and he said, "When you say you have a lot of Web sites, how many are
we talking here? Do you have five, do you have 10, how many
is it?" And the person got a sort of sheepish look
on his face, and he sort of looked around a little bit--he was on the front row--he
said, "I have 1,500 Web sites." [laughs] And the whole room went, "Oh!"
Just a ton of Web sites. So that was pretty funny.
Somebody will as a question about a completely innocent Web site, and you're, like, you know
you've got some hidden text on the bottom of this page, or something like that.
The other question that I get, which is sort of funny, is sometimes people will say or
ask a question that's sort of like how much can I get away with?
It's almost like a student asking the professor, "Well, is this going to be on the test?
Okay. Is this going to be on the test?"
And at some point, you want to just say, "You know, make a great site, promote it well,
do _____ ??, and you'll sleep well, and you don't need to worry about how much can I get
away with?"
[pause]
I'm a big sucker for gadgets. So I really enjoy open source, Ubon2, Linux--gadgets
in general. All of those are a lot of fun for me.
Another thing that is kind of weird, but that I enjoy a lot, are plug-ins.
And kind of any type of plug-in--Firefox extensions, Photoshop plug-ins, Firefox add-ons, WordPress
add-ons. And so I think if I had an infinite amount
of time, I would spend a lot of time just writing these little plug-ins and extensions
for all these different pieces of software. It's a lot of fun to see somebody put something
out there and then have people have the ability to hack it or mod [sic] it or tune it or tweak
it or whatever you want to call it--to be able to improve the functionality above and
beyond whatever it would already be. So that's a lot of fun, and probably if I
had an infinite amount of time, something about open source or maybe doing various hacks
and extensions--things like that.