... and we're back! About a year ago I laid the foundation for a website called XMLToday.org, with a vision towards covering a number of issues that pertained to the XML community and those people who used XML related technologies to build applications, create documentation, create messenging architectures and otherwise communicate with people. It's had a rather troubled existence - felled once by an errant database and a second time by a lack of funds, given that the project was started (and continues) in the spirit of open technologies in general (that is to say, underfunded and definitely a labor of love).
The latest setback has had a couple of unexpected side effects. The first of these was the outpouring of support and offers of help in getting the site back up and running. In particular I owe a debt of thanks to David Lee, he of xmlsh fame, who volunteered server space to help me relocate the site off of the existing (quite expensive) server, onto a machine that seems ideal for the task, and to Bill Blondeau, who contributed a significant donation to help in the migration of content off of the existing server. A number of people also volunteered server space or other offers of assistance, and I was both pleased and grateful for your offers.
The second side effect is one that should be fairly immediately obvious upon viewing the site. Drupal is a powerful CMS, incredibly flexible and capable of a great number of things, but that same flexibility can bite you in a soft spot if you're not careful. Even before the site went down, I'd been thinking of relaunching the site with some significant redesign of the architecture, and given the opportunity, I could hardly turn it down whenit came to laying out the new site.
There are a number of new capabilities to the site, some more obvious than others. The first is just simply speed - the old site was painfully slow, and was becoming difficult to maintain. I came to realize, as I was rebuilding it, that the memory had been throttled to well below it's necessary minimum value, which meant that a lot of the processing was being down using virtual RAM, slowing the site down by an order of magnitude. A bit of research (something I'll write up in a subsequent article) let me figure out how to fix it, and now the site's practically flying along.
The second facet, and again one that came about after doing some research on a similar site, was the breakdown of the site by editorial article type. Faced with redesigning the site, one thing that struck me was that much of the manual work that had given into creating divisions by editorial type, and that with a bit of research and intelligence, it should have been possible to rebuild the site so that you could automatically see just news content, or technical analysis, or code snippets, or events, all maintained through the taxonomy. When an author creates a new blog post, he or she is required to specify the type of post from a drop-down selection box.
This in turn updates the "Article by Type" menu on the left side panel, and clicking on the appropriate type will give the listings for that type, sorted by descending date. This should make it much easier for users to find information from the site closest to their interests. In the front page listings (and any time that you see a summary (or teaser, as it's known in Drupal land) for an article, you'll also see a graphic uniquely identifying the article type at a glance.
A third, more subtle change, is the introduction of the Geshi code syntax highlighter, making it possible to create much more visually appealing and legible code output in articles. While Geshi does come with a quite decent XML highlighter, it doesn't yet support XQuery (I may take a stab at writing an XQuery filter for Geshi just to make that more widely available). I'll provide more information about Geshi in a subsequent FAQ (there will be more of these as well on the site) once I have the site itself set up.
Images can be attached to blog posts via the Attached images part of the editor. This makes it possible to upload images that can be used to illustrate blog bosts, though I'm still working out all the bugs with that one. It's usable, just not ideal at the moment for my needs in particular.
Finally, I've switched to the Mollom spam filter service, which provides a much more comprehensive suite of tools (and occasionally human intervention) to foil spammers. This had become a real problem with the older site, forcing me to limit the capabilities of legitimate posters in order to deal with the Turing Bots. From talking with others that have used Mollom, this should go a long way towards solving this issue.
There is still a fair amount left to do. The news and jobs listings that were a big part of the older site will be rebuilt to take better advantage of the redesign, but this will take some time. In the interrim, if you have a job listing that you'd like advertised (or a resume you want to post), then either contact me or join the site and post it as a blog posting to the "Help Wanted" or "Resume" category terms for Article Type.
More significantly, there are a number of articles that are on the other site that I will be exporting and reimporting to this site over the next month. I will also be creating an automated newsletter that will post new articles every week or so to all subscribers (in full this time), under the appropriate channels (again, just a matter of figuring out the mechanics).
If you had a login under the old site, you will need to re-register onto the new one. I'm sorry for the inconvenience that this may have caused - and its possible I may be able to export and reimport users, but until I do it would probably be easier (and just as free) to have you do it yourself.
While I'll probably be posting periodic site announcements such as this, most of the upcoming content should be more technical or business oriented in nature. Please let me know (click on the little messenger icon at the bottom of the Author panel on the left hand side, another new feature of the site, to send me a message) what you think of the site, any problems that you have and any recommendations that you'd make for new or improved features.
- Kurt Cagle's blog
- Add new comment

- Quote
- 682 reads



Re: On XML, Friends and Site Migration
Hi Kurt,
Like the new format and capabilities!
I work with Dorothy Hoskins and we linked from our site at www.novatekcom.com to the post you wrote recently about her paper DITA is Open for Business. So hoping you are able to get the old content over quickly. Any chance you'll keep the old URLs, or create some rewrite rules or someway to redirect the old URLs to the new locations?
Thanks.
Dave Feasey
Novatek Communications
Re: On XML, Friends and Site Migration
Dave,
It's my plan to pull over as many of the articles as I can recover from the old site, with their old URLs. It may take a while to get everything, but my goal is to restore as much of the useful content as possible.