I discovered that the RSS Validator does not accept relative links in weblog entries. Hmmm, I guess we have to work on our URL-expander in Roller for RSS Feeds. A while back, I found that relative links where a problem in NetNewsWire as well. In my opinion, aggregators should be able to handle some type of base element, like <base href="..." /> in HTML and resolve links from that.
I upgraded Roller to 0.9.7-dev and Tomcat to 4.0.6 (from 4.0.4). Everything appears to have gone smoothly - let me know if I missed something. Also, Roller appears to be incredibly stable since I changed my <session-timeout> (in web.xml) setting from 3600 to 30 minutes. I guess Tomcat can't handle a session timeout of 2.5 days very well!
From my Inbox this morning, Subject: BEA Product News - November 2002.
Designed for server-side enterprise Java applications, BEA WebLogic JRockit, will be supporting the latest J2SE release 1.4.1. WebLogic JRockit 8.0 Beta noticeably improves the startup time for server-side Java applications, and it enables easier debugging and profiling of Java applications through support for both JVMPI and JVMDI. Additionally, JRockit 8.0 Beta introduces support for Intel® Itanium® 2 servers on both Microsoft® Windows® .NET ES and Red Hat® Linux® Advanced Server.
The WebLogic JRockit 8.0 Beta will be available for download on November 11.
More info >>
For anyone who attempts to maintain a valid XHTML site, you know that it's a fair amount of work to make sure your site is valid all the time - especially if you're linking to other sites. You learn to hate ampersands (&). To solve this problem in Roller, I think we need a validator built into the posting of new content. It'd be a sweet feature, maybe we can start with a dirty URL cleaner such as the Hivelogic URL Cleaner. Further comments on this topic can be found at web-graphics.com. Interesting tidbit from Dave:
If you are using XHTML (and you should be!) you could try to incorporate my JavaScript XML parser.
Youâd have to tweak it to wrap the post in a fake "root" element, but
then it could tell whether or not the particular post is well-formed
XML. It could not, however, tell you whether or not your post is valid
(i.e. conforms to a particular DOCENGINE), but it would catch things like
misplaced ampersands, unknown entities, tags that arenât closed, etc.
Better yet, if you use IE for your posts, you could do the same thing using its built-in XML parser.
Yeah I want one, but will I get one? Not this year, my last one is less than a year old. The things I really like about the new ones are 1) DVD Burner, 2) higher screen resolution and 3) faster processor. I might have to get a new one next spring when they come out with the 1.8 Ghz processors - that'll be worth the upgrade!
So far, so fast.
SWT has created lots of love for Eclipse, and I'm willing to bet I'll write a XUL application in the next couple of years. Russell Jones from DevX.com explains how these might very well be the future of cross-platform GUIs.
The reality of a single cross-language, cross-platform GUI interface programming model is in sight, based on an XML description language supported by fast native runtimes. Will Mozilla, Eclipse, or someone else step in and complete the last mile that gives all developers a common way to design and program fast cross-platform user interfaces?
XUL seems to be simply a combination of XML and JavaScript from this example. So the question is, will you be writing HTML, XML, XHTML or XUL in the next couple of years. To stay ahead of the game, I recommend writing your web-based apps in XHTML (which is XML) and if you migrate to XUL, it'll be easy as pie.
The blogging community is great, or at least the guy's blogs that I read. You know who you are. To everyone who's offered congratulations and given our daughter compliments (are you hitting on her already?!) - your thoughts and words are much appreciated. It's always fun to see your name in lights. Here's your names right back at ya (in no particular order) - you guys are great: Dave, Lance, Russell, Greg,
Jeff, Rick, Anthony, Dominic,
and Erik.
If I missed your name, let me know, because I might not be reading your blog, and I certainly should be. Kudos go to Russell for the first post and my friend Brian Burke (no blog yet), both who guessed that we were giving birth from the lack of a Monday posting.
Chimera 0.6 was released the day before Abbie's birthday. The key feature (for me) is the support of Keychain to remember web site passwords. I switched back to Mozilla for a week or so because of this feature, but now I'm back on the juice that Chimera offers. Funny how Phoenix and Chimera have become my favorite browsers in such a short period of time (2 months for Chimera, 2 weeks for Phoenix). I can't wait for the 1.0 releases! I think it'll be tough for them to gain much market share -> folks are just too ingrained in their ways to change. If it ain't broke, don't fix it.
I really like this website for displaying a photo album. I like Russell's Scrapbook as well, and I use Apple's .mac with iPhoto. iPhoto is OK, but the HTML that .mac's "Homepage" application writes is meant for 8 year olds and frustrates me b/c I can't do any advanced HTML editing (although I can download the HTML files and change it that way). I can't beat the ease of iPhoto import/publish, but is it work $100 year when I have my own domain already? I'll stick with it for now, but I'm in the market for alternatives.