Matthew Gifford

The Sacramento Bee has posted a story about a study that shows that support for our free speech rights is waning.

I strongly disagree with Jeff Zeldman about returning to the use of transitional markup. Giving in to bad browsers will only slow the correct implementation of XHTML, CSS, DOM, and other standards. I believe that it would be more constructive to educate users about the specific shortcomings of the software that they use and tell them what they can do about it. Hopefully, increased bug reporting and changes in market share would be enough to convince browser developers that changes need to be made.

I know that this isn’t a very convincing argument, so I’m going to take some time this weekend to expand on my ideas.

I’m pleased that the players won’t be going on strike this year. However, I fear that the agreement will do nothing about the fact that the players’ salaries are slowly choking baseball to death. I don’t think things will be getting any better until salaries are capped.

Apple has announced that they are going to be releasing the source for Rendezvous, their implementation of Zeroconf. Of all the new stuff in Jaguar, I think Rendezvous is the most interesting.

Spring is an alternative desktop for OS X. It’s organized around people, places, and tasks, rather than files and folders. It might be something to keep an eye on.

…the RIAA site is still down. And no one misses it.

The New Republic has posted an excellent article about the sorry state of American broadband service. Based on what a friend told me when I first moved to Illinois, I can’t get decent broadband for a reason not mentioned in the article. According to him, state law requires the cable and telecom companies to share their lines with other providers. Their response was to stop building infrastructure altogether. Can anyone verify this for me?

The RIAA’s site was cracked overnight. I wouldn’t be surprised if this becomes a regular occurrence if the Berman-Coble bill becomes law. (via Dave Winer)

My new blog software generates pretty URLs using a technique adapted from an article at A List Apart. In a nutshell, all requests for PHP files are redirected back to index.php using mod_rewrite commands in the .htaccess file. If the URL points to an actual file, the file is included. If no such file exists, the directories in the URL are converted into variables and run through select statements. Overall, I’m pretty happy with how it works. However, I’ve run across two problems:

1. The article says one of the conditions for deciding whether or not the URL points to a file should be:

   file_exists($DOCUMENT_ROOT . $REQUEST_URI)

   If you don’t use query strings anywhere, that would be fine. But I like to use query strings for my non-public administration files. It’s just easier that way. To scratch my particular itch, I changed the condition to:

   file_exists($DOCUMENT_ROOT . ereg_replace("\?.*$", "", $REQUEST_URI))

   The next line should then be changed to:

   include($DOCUMENT_ROOT . ereg_replace("\?.*$", "", $REQUEST_URI));

2. .htaccess and the mod_rewrite commands in it are being applied to all directories below the document root. To exempt a directory from being affected by mod_rewrite, I have to add this to .htaccess:

   RewriteCond %{REQUEST_URI} !^/directoryname/*

   This gets messy as I add more directories. What’s the correct solution? I’m sure it’s totally obvious, but I’m rather new to using .htaccess.

I have been approached to join the Quixtar pyramid scheme again. Their explanation of how it works just doesn’t make sense.

Beware of people who want to talk about business opportunities and have a bww.com e-mail address on their business cards.