Archives for October 2006

Small bits

A Washington Post article about Google hiring and culture. One nitpick: the article states “Every bathroom stall on the company campus holds a Japanese high-tech commode with a heated seat.” That’s not quite right. Usually each bathroom will have one such toilet, but not every single stall has them.

Then Battelle ponders a Washington Post article on click fraud.

Looks like Ze Frank was offering sponsors “duckies” with cute mouse-over messages and Google Checkout said they couldn’t support that. The Checkout folks apparently studied the details further and decided to allow it. Earth sandwiches for everyone.

Speaking of Earth, Google Earth offers a United States election guide. It looks like this:

Google Earth District

That’s the 14th district. If you click on it, you see the Representatives and Senators from that district. From there, you can click into places like Open Secrets to see what political action committees (PACs) are giving money to a candidate. For example, you can see who is giving money to Anna Eshoo (my representative) on this page. Pretty neat, and clicking around on Open Secrets is addictive.

By the way, it’s really cool that searching in Google News can take you to blog results now. I did the search [google earth election] and got some news results. But at the bottom of the page, you also see the option to run the search on Google Blog search:

Blog search in news

Steve Rubel noticed another blog search link on Google News too. Cool. It’s been a couple months since Yahoo dropped their blog search via News. I wonder if there’s any new news about that?

Happy Diwali!

By the way, Happy Diwali! You can read more about Diwali here. Growing up in Eastern Kentucky in the 80s, I didn’t really encounter much discussion of the festival of lights. 🙂 It wasn’t until I was in college that I heard of it.

I hope everyone has a sweet and safe Diwali.

Interview today with eMarketing Talk Show

At SES San Jose, I noticed Cindy Turrietta asking good questions. When she asked me then if I’d do an interview for eMarketing Talk Show, it sounded like a good idea. Today’s the day; I’ll be talking to the hosts of the show (Cindy, Brooke, and Todd) today at 4 p.m. Pacific.

The archive is live. You can listen to it here.

More webmaster console goodness

It seems as if every 5-6 weeks, the Google webmaster console team rolls out more requested features into Google’s webmaster tools. You can read Vanessa’s post about the latest good stuff. Danny covers it pretty well over here, but I thought I’d show snapshots from my site.

Here’s what my site looks like:

Matt's crawl stats

As you can see, my site is pretty small (I’ve only written a few hundred posts). On average, Google pulls down 438 of my pages each day. And my domain, while not super-speedy, normally returns pages in under 3/4ths of a second.

The other thing to notice is the crawl rate. Google limits how hard it hits web servers with something called “hostload,” which is a measure of how many bots can be simultaneously fetching pages from a web server. Notice how “Faster” is grayed out for my site? That means that hostload and crawl-rate isn’t anywhere near a limiting factor for my site. Heck, even a single Googlebot could fetch pages at a leisurely pace and still crawl most of my site every day. 🙂

But suppose I ran a large domain such as or, or a site with thousands of pages. Then hostload could potentially be a factor. Even if N bots are allowed to fetch pages simultaneously, those bots might not be able to fully crawl a site in the amount of time that we crawl before beginning indexing. If hostload is a factor for your site, the “Faster” option will be available to you. In that case, if you’re willing for Google to crawl your site harder, we should be able to fetch and index more pages from your site.

Of course, you can always opt for slower crawling as well. If you’d like less load on your webserver but don’t want to block Google completely (e.g. with robots.txt), requesting a slower crawl is a good idea.

Let’s see, what else is in this release? Oh, you can opt-in to having your images labeled in the Google Image Labeler game. If you want people to provide free labels for your images, that’s a great reason to try out the webmaster console right there.

The final feature is just a count of urls that we found in a Sitemaps file. As Vanessa said in the official post:

Recently at SES San Jose, a webmaster asked me if we could show the number of URLs we find in a Sitemap. He said that he generates his Sitemaps automatically and he’d like confirmation that the number he thinks he generated is the same number we received. We thought this was a great idea.

This is the perfect example of how things should work to me. The webmaster console folks have their own ideas on what webmasters will find useful. But talking to webmasters is the best way to hear what people really want (this particular idea came from Tim Jackson at Plumber Surplus, for example).

So what do you want to see from future versions of the webmaster console? Philipp put out a call a month ago and got over 60 comments. I know the webmaster console team will read the feedback here as well. So what should we do next? And if you haven’t tried the Google webmaster console yet, please give it a test drive. It may help you find problems with your site, and it will offer more and more information over time.

Firefox 2.0 or IE7?

I was going to try out both IE7 (Robert Scoble says it just came out) and the Firefox 2.0 release candidate. You know, a little compare and contrast piece. Then I backed out of trying IE7. 🙂 I’ve talked to two friends recently who tried it and had to uninstall it shortly afterwards, and Danny’s experience scared the bejeezus out of me.

But I tried out Firefox 2.0 (release candidate 2) last week, and here’s what I found.
– FF2.0 will automatically run spell correct for you in the text boxes of web forms. On one hand, it’s a little annoying at first to see red underlines under stuff (I’m a good speller, so all my suggestions were for real words like “webspam”). 😉 On the other hand, it’s easy to right-click and add a word to the dictionary, so if you have a single computer then in a few days most of your weird words will be added. On the third-hand, if you have multiple computers, I hate the idea of adding a word to my home computer’s dictionary, my laptop’s dictionary, and my work computer’s dictionary. Let’s hope that Google Browser Sync or some other extension gives you a way to sync dictionaries on different computers. Oh, and by the way: Google Browser Sync seems to work great with FF2.0. I haven’t had any problems keeping bookmarks in sync between 1.5 branch versions and the 2.0 version.
– The first time a new window popped up (e.g. manage bookmarks or manage search engines), Firefox would hang for 3-4 seconds. It never paused after that first time though. Maybe it was just building some XUL as a one-time thing?
– Firefox2.0 is better on memory leaks than the 1.5 branch, at least for the sites I visited. You’re still looking at 100MB or so if you open a bunch of tabs, but I didn’t see memory usage growing and growing.
– I haven’t had any crashes yet. Again, I might be lucky, but I tend to hit some scuzzball sites as a member of Google’s webspam team.
– I think the 2.0 release candidate handles virtual private networks (VPNs) better? When I hop on the Google VPN with 1.5.0.x, I normally need to use SwitchProxy and change a proxy to kick the browser a little bit. In 2.0, things just switched over seamlessly for me.

Here are some nice things that I missed but noticed in the release notes:
– The History menu keeps a list of recently closed tabs, so you can undo a closed tab.
– If the browser crashes, Firefox will restore not only windows and tabs, but even words in a textarea box.
These two features are great, and pull in the best features of the SessionSaver extension.

This version isn’t perfect though. There are two things I’d like to change (one new issue and one longstanding one):
– They put a “close tab” button on each tab. I don’t mind that, but my muscle memory goes to the top-right to close a tab. Instead, it’s now some doohickey to show me a list of what’s in each tab. Dude, I don’t need that–I opened the tabs myself, so I don’t need some table of contents. 🙂 I hope they’ll provide an about:config flag to give me my “X” back in the top-right corner.
– Firefox does not store favicons for urls that redirect. I *hate* that. I hit some of my bookmarks in the Personal Toolbar Folder 30-40 times a day, but the favicon never shows up beside them? I think it’s for urls that redirect. Come on, if my bookmark has url A and when I click on it I go to url B or C and there’s a favicon on the final destination url, go ahead and store it for url A. Or give me an about:config option for it.

But those are minor nuisances. Overall, version 2.0 of Firefox looks very stable to me.

Update: The answer to my close-tab preference is found in the comments. I love my commenters. 🙂