Dylan Schiemann

NetWindows, an open source api project started by Alex Russell, is much improved. I took a second look at it recently and am very impressed with how well it handles a number of application like interfaces. It uses a signals and slots event mechanism, and despite there not being much in the way of updated documentation, I was able to get working what I needed for a client project within a few hours.

New Architect magazine is no longer being published. Under its original title of Web Techniques, it was a useful magazine for beginning web developers with code samples. The more recent rendition was more appropriate for business types (no code) making broad decisions related to web technologies. Except for the Lawrence Lessig articles related to the DMCA, the new version was not very interesting. I stopped subscribing, mostly because the content was becoming less useful, and partly because all articles were available through their web site.

I had never considered using XQuery for generating html and xml. Until now.

LawMeme summary of SearchKing v. Google case, an interesting read even if you have little to no interest in law.

SitePen client West Coast Internet is offering a replacement DSL service for former DirectTV DSL customers, a service that is being discontinued.

“The Checky plug-in is an simple Interface for web content and resource developers to free and commercial online Validator and Checker services.”

Tom’s Hardware article on building stable computers. Great information on choosing components (where and where not to cut costs), assembling the system, installing software, and testing. The first really good article I’ve seen in a while on building a reliable computer.

An article describing the benefits and shortcomings of using DOM mutations events in SVG. Much of the recent innovation in JavaScript is closely coupled with SVG. Unfortunately, mozilla and IE do not yet come by default with native SVG support.

I’ve been using mozilla1.3 alpha for a while, and I couldn’t figure out why the spam algorithm would stop working after a day or two. Well, it turns out in addition to having to manually mark messages as spam, you also have to explicitly mark messages as not spam. In retrospect, this sounds obvious, but it isn’t because mozilla uses identical looking icons for messages that are not spam and messages that it couldn’t determine, so I thought it was saying messages were not spam when really it was saying that it didn’t know. Sure enough, after labeling a few hundred messages as not spam, all spam messages are now labeled as spam once again.

A Washington Post article describes mozilla as the best hope for better e-mail software in outlining mozilla’s addition of spam filters, palm address synchronization, and saved searches.