Archives for August 2005


I use Bloglines all the time. Here’s what I love about Bloglines:

  • It’s an easy way to keep an eye on the blogosphere, or at least the parts of the blogosphere that I care about.
  • I can access it from anywhere and it saves my state. My browser knows the password, so it’s one-click from work/home/laptop to start reading Battelle, the Spam Huntress, Mapping Hacks, martinibuster, or even the wet/burnt dog hair guy at Yahoo, and I can always pick up where I left off.
  • I can choose whether to read a lightweight comic feed or tackle the voluminous-but-useful set of posts that is the Search Engine Watch blog.
  • I don’t have to worry about funky spelling, capitalization, or a name that sounds close to “Google” (pet peeve of mine). If the name were BloogleLyNes, I’d probably like it less. TiVo gets a pass on capitalization because they rock so hard, but I wish more sites chose easy, memorable spellings. I’m looking at you, kuro5hin and The Cap’n’s Log 😉
  • Bloglines’ favicon is clean, distinctive, and memorable. Easy to pick out from my personal toolbar folder.

Here is what I wish Bloglines would change:

  • As far as I can tell, you can only accumulate 200 posts from a feed before posts start dropping on the floor. Mark Fletcher, can you boost that to 400 or 500? I don’t want to have to check ThreadWatch that often; sometimes it puts me in a foul mood, so I’d rather save it up for a while and then read it all at once. 😉
  • Sometimes posts appear as unread more than once. I know that usually it’s because of an update to the post, but sometimes it doesn’t look like the post has changed. Maybe a different color instead of bold/black for updated posts?
  • When something has to go down for an upgrade, you see the [bloglines plumber]. Less plumber time is always good. 🙂

Dashes vs. underscores

I often get asked whether I’d recommend dashes or underscores for words in urls. For urls in Google, I would recommend using dashes. Why? To find out, let’s take a trip in the Google Time Machine. Set the dial for 1999, the year Matt first discovered Google. Matt was using, I dunno, maybe HotBot at that point? The curtain rises:

Matt: Hmm, this search for [FTP_BINARY] didn’t turn out the way I wanted. I got a couple scuzzy looking urls, and the other documents just have the words “FTP” and “BINARY” but the term “FTP_BINARY” doesn’t actually appear. (Note: Matt was a bit of a nerd, as you can tell.)
Some Random Person That I Don’t Remember: Have you tried Google?
Matt: What’s that?
SROTIDR: It’s a search engine written by nerds for nerds! They index numbers! Sometimes they even index punctuation, like “C++”. Try your underscore search there.
Matt: Okay, here goes. Whoa! They actually return pages with the literal string “FTP_BINARY”! That’s wicked cool! (Did I mention Matt was a nerd? Big-time nerd.)
SROTIDR: Yeah. The wild thing is that they wrote a paper about how they crawl the web and rank pages.
Matt: Well, now that’s just silly. I wonder why they didn’t keep it a secret? I bet those papers will make great reading for my information retrieval class.

I’ve stylized the conversation quite a bit, but I remember how impressed I was that Google indexed numbers and some punctuation (come to think of it, search engines have come a long way in five years). With underscores, Google’s programmer roots are showing. Lots of computer programming languages have stuff like _MAXINT, which may be different than MAXINT. So if you have a url like word1_word2, Google will only return that page if the user searches for word1_word2 (which almost never happens). If you have a url like word1-word2, that page can be returned for the searches word1, word2, and even “word1 word2”.

That’s why I would always choose dashes instead of underscores. To answer a common question, Google doesn’t algorithmically penalize for dashes in the url. Of course I can only speak for Google, not other search engines. And bear in mind that if your domain looks like, that may still attract attention for other reasons. 🙂

Seeing nofollow links

Update: The ChromEdit extension I mention below is stale and doesn’t work with Firefox 1.5. Someone has made a version for FF1.5 though; it’s called ChromEdit Plus. Follow the directions at the top of that page. The short version is 1. save the .xpi file to your desktop, then 2. drag the .xpi file over onto your browser to install the .xpi file. 3. Restart Firefox to install the extension. 4. To use, select Tools->ChromeEdit Plus->ChromEdit. Select the userContent.css tab and paste the code

a[rel~="nofollow"] {
  border: thin dashed firebrick ! important;
  background-color: rgb(255, 200, 200) ! important;

into the userContent.css area, then click Save. You might need to restart Firefox once more, but when you do, this link should have a pink background, and a dotted red line around it.

Old post follows:
The blessing/curse of learning more about SEO is that it’s often in the back of your mind; it affects how you look at regular web pages. Here’s a simple tip that will help with your X-ray vision: make nofollow links visible.

When you’re using Firefox, you can make a file called userContent.css to override the css on a given page. First you have to find the right directory; all your Firefox settings are stored in a profile directory. This page tells how to find your profile directory. Once you know the profile directory location, go into the “chrome” directory and look for a file userContent.css. If no such file exists, there may be a file called “userContent-example.css” that you can rename. Here’s what you can put into userContent.css:

a[rel~="nofollow"] {
  border: thin dashed firebrick ! important;
  background-color: rgb(255, 200, 200) ! important;

(Note: you may need to restart Firefox as well for the changes to take effect.)

Then if you want to link to a blackhat spammer without it counting as a vote in Google, just add rel=”nofollow” to the hyperlink, and you’ll be able to tell the difference between normal and nofollow links.

When you view a page, you’ll see something like this (please pardon the self-referential snapshot):

Example of visible nofollow CSS

I highly recommend adding code to your userContent.css file so that you can see nofollow links easily. Note: ChromEdit is a nice extension that makes editing your userContent.css file really easy, but it currently (Jan. 14, 2006) only works with Firefox versions less than 1.5.

Smart is sexy

I’m happy about our new desktop search, but this is awfully cool. The more you can understand about the language on a page, the more you can do with that meaning to improve relevance. World-class machine translation makes me hot. 🙂

I do not wish my screensaver to lock my computer, thank you.

In Windows XP, it really annoys me when my computer idles for a few minutes and then the screen locks. I try to be careful not to leave my laptop lying around, so I prefer my screensaver not to be password-protected.

In order to make it so that your computer won’t lock itself after a few minutes, do this:

  • Run ‘regedit’ to edit your registry
  • Navigate down the tree to HKEY_CURRENT_USER > Software Policies > Microsoft > Windows > Control Panel > Desktop and look for an entry ScreenSaverIsSecure. You want to make sure the value for ScreenSaverIsSecure is 0 (zero).
  • Exit regedit; you’re done!

Now just remember not to leave your laptop sitting around at your local den of identity theft.

More details for the terminally curious.