Matt Cutts: Gadgets, Google, and SEO

This link was cool, but it generates a graph like this:

With a little modification, I made this graph:

I like my picture a little better. It was quite simple to make this diagram, and Google provides a free graph-drawing tool that you can use on your own site with a single url — no account or login is needed! Let’s break down how I made this image:

http://chart.apis.google.com/chart? Loads an image from Google Chart

cht=p Chart type is a pie chart

&chtt=Percentage%20of%20Google%20Chart%20Which%20Resembles%20Pac-man Chart title

&chs=550×250 Chart size

&chd=t:10,80,10 Chart data. This is the only tricky bit. The pie chart starts from the “x-axis” of the pie. The “t:” means that the data is in text format with numbers between 0.0 and 100.0. Going clockwise, the pie chart is 10 units of non-Pacman, then 80 units of Pacman, then another 10 units of non-Pacman. The Google Graph API guide adds “Note: For text encoding, scale your data by converting it into percentages of the largest value in your data set.”

You don’t have to scale your max value to 100, but let’s do it for fun. My maximum value is 80, so to scale 80 to 100 I’d multiply my numbers by 1.25. I tried a value of “12.5,100,12.5″ and it generated an identical Pacman graph. You can even make a funky graph using 12.5 for gray, then use eight yellow slices of 12.5 units each (which add up to 100), and then finish off with a final 12.5 of gray:

I like using the values 10,80,10 more. If your numbers add up to 100 then the data points are just the percentage of the pie chart: 10%, then 80%, then 10%. So 20% of the graph is gray and 80% is yellow. That’s easier for me to remember. Okay, I geeked out a bit there, sorry. Back to the graph.

&chco=FAFAFA,FFFF00,FAFAFA Chart colors. The non-Pacman bits are the color #FAFAFA, while the Pacman color is #FFFF00.

&chl=Does%20not%20resemble%20Pac-man|Resembles%20Pac-man Chart labels

I really like this graph-drawing service because anyone on the web can use it for free without even registering. For example, I used a Google-o-meter graph in a recent post:

I almost wanted to call this post “Stupid Google Tricks.” What fun diagrams can you imagine making with the Google Chart service?

This weekend I did a Q&A session at the Domain Roundtable Conference. It was an hour and a half of answering various questions. Rand Fishkin and John Andrews both did write-ups of the session. Rand and John were both on an SEO panel after me, which I enjoyed.

This was my first domain-related conference, and the vibe was interesting. It was smaller than a typical search conference, but many people seemed to know each other. I enjoyed seeing a few people in the audience that I’ve read about. People weren’t incredibly eager to volunteer specific sites to discuss, but I can understand that. Overall I definitely enjoyed the conference and talking to a different type of audience.

I kept a mental list of the places that I mentioned, and I remember talking about these Google resources:

• Google’s webmaster blog. Someone asked about moving a site to a different domain and we had just done a post about how to move your domain.
• Google Ad Manager lets you sell and manage ads on your site. You can also choose to use AdSense for the ads that you haven’t sold.
• Google’s webmaster portal. Someone bought a domain that they thought might have been bad in a former life. I told them that they should file a reconsideration request using the form in the webmaster console.
• Google’s DMCA Policies. In case you want to report copyright infringement to Google. The person already knew how to do a DMCA complaint to the webhost that was serving up infringing content.
• Google’s quality guidelines. Someone said that they had 800 links on a web page. I recommended that they keep it to under 100 or so, as we mention in the technical part of our guidelines.

One point that I wanted to make is that lots of people seem to buy domains for the joy of finding a “diamond in the rough” — a nice domain name at a good price. And plenty of people are into domaining as a way to make money. But only one person in the audience raised their hand when I asked how many folks really got into the domain business to build out and develop the content on domains. To me, that means there’s some opportunity there. For example, suppose you bought dullest.com for $1000 (full disclosure: I own dullest.com, but I paid about ten bucks for it.) If you’re not doing anything with it, you could make a deal with a blogger or web designer. The blogger could create content for the site, and if/when dullest.com was sold, the blogger would get a fraction of the profit from selling the domain. You probably wouldn’t do that with every domain in your portfolio of course, but if you had some good domains and they were just sitting around empty, it might not be a bad way to demonstrate the value of a domain. It’s Earth Day today, and both earthday.com and earthday.org are parked. If someone could develop one of those sites, that might be worth a cut of the sale. Who knows, maybe this is a stupid idea — or maybe someone is already doing it — but I liked the idea of giving an aspiring writer/designer/programmer/blogger some equity if they could improve the selling price of a domain.

Next up, I’m speaking at the Web 2.0 Expo this week. Tomorrow (Wednesday) I’ll be participating in a Speed Q&A: 5 tables, 5 experts, and the experts rotate to a new table every 10 minutes. The Web 2.0 Expo costs money, but I think you can get into the Speed Q&A part of the conference for free. If you decide to stop by, please bring questions that everyone would be interested in, not just “Can you critique my site for me?”

I’m also doing a short keynote (ten minutes) on Friday speaking about “What Google Knows About Spam.” I’m struggling with what exactly to say. On one hand, Google knows a lot about spam, as illustrated by this graph:

On the other hand, I don’t want to disclose things that would benefit people that try to spam. I’m sure I’ll come up with something by Friday. By the way, that danger dial-ish diagram above is a Google-o-meter graph.

There are a ton of Interney Relay Chat (IRC) clients for Linux/Ubuntu, e.g. Gaim (now called Pidgin). One IRC client that makes recording an IRC conversation pretty easy is Irssi. Install Irssi on Ubuntu with a simple command:

sudo apt-get install irssi

Then run the program “irssi” from the command line.

Now suppose you want to record what’s going on in the #iphone channel, which runs on the IRC server irc.osx86.hu. Pick a nickname for yourself such as “notanewbie” and run the following commands:

irssi
/set nick notanewbie
/set autolog on
/connect irc.osx86.hu
/list
/join #iphone

Then just leave the terminal running or type “/quit” when you’re ready to exit. The IRC chat log will automatically be placed in ~/irclogs/osx86/#iphone.log . The chat log includes messages when people join/leave the channel, so you can use this command to peruse what people are actually saying:

cat \#iphone.log | egrep -v ‘has quit|has left|has joined’ | less

That should get you recording an IRC channel. For example, if you wanted to record the Webmaster Radio channel from webmasterradio.fm, you’d type “/connect irc.webmasterradio.fm” in the “/connect” line above, and then use “/join #webmasterradio” to join that channel and record it. So the commands would be:

irssi
/set nick notanewbie
/set autolog on
/connect irc.webmasterradio.fm
/list
/join #webmasterradio

If you join multiple channels, you can switch between them with -#, where # is a number like 0-9

Google just announced a cool addition to Google News. If you search for a person’s name on Google News, you can see statements where that person has been quoted by a news source. For example, search on Google News for [Arnold Schwarzenegger] and you’ll see

It’s like how smart the feature is. It can correctly handle hard cases like “she said” or “he said” quite well. Note that the mention of “True Lies” in the first quote didn’t throw it off, even though it was in quotes too. Once you’re looking at the news quotes of a person, you can also search within their quotes. So you could see what Arnold had to say about California’s hands-free cellphone law by searching his quotes for [cellphone].

I like that Google is doing this for a couple reasons. First, it’s an additional way to slice and dice news information that wasn’t really available before. In that sense, it reminds me of taking geographic mentions in books and plotting that data on a map of the world or over time. There’s no way to do that scale of research or analysis by hand.

The other reason that I like this feature is that it fits in with a basic Google philosophy, which is “you enter whatever interests you in the search box, and we’ll try to do something smart to help.” I really like that we have features like a smart calculator. Or that you can type in [whois example.com] and we’ll give you some information about the domain. Or that you can type in an airline flight like [aa 125] and get up-to-the-minute flight status. Users don’t have to do anything special, but we work hard to show something helpful for each search.

I communicate with people in lots of ways: face-to-face, email, via my blog, leaving comments in the blogosphere, conferences, etc. At SMX West a couple people asked “I sent you a friend invite on service X but you haven’t responded. Do you not like me?” Please don’t feel bad, because it’s not that. I’m letting a lot of requests drop on the floor — even requests from other Googlers to chat on Google Talk. I did a quick check of various social services and here’s what I found:

LinkedIn: 176 invitations to connect
Twitter: 671 requests 1060 requests
Google Talk: 27 chat requests
Facebook: 190 friend requests
MySpace: 35 friends, and it’s a fake account that someone else set up in my name (I’m not 42 years old, thank you very much ).

At this point, managing friend invitations feels more like work than fun. Many of these services have really poor interfaces for mass approving, and a while ago I discovered that if I stopped responding to friend requests, very few people got angry with me. So if I haven’t responded to a friend request from you, please don’t take it personally — I’m just a little socially exhausted.

By the way, I have a precise measurement of being Calicanissed. He told his twitter following to add me, and I got almost exactly 400 additional twitter requests. Jason didn’t know it, but I had my twitter set to the private mode that requires each twitterer to be approved. Thanks, Jason.