I just released the 0.01 (pre-alpha) version of Alexandra , a roundtripping forms system that uses RNG (Relax NG), XSLT pipelining (using AxKit of course...;-) and presents in HTML forms. There's an interactive demo to play with as well.

Just for the first time, I saw a posting on slashdot signed with a female name. (It was Allison.) This is the first time I can ever remember seeing that.

If you're coming here from slashdot you might want to look at this other site I'm involved in OpenICT.net .

Now read this ... Steve Deering gave a brilliant talk at the 51st IETF meeting called Watching the Waist of the Protocol Hourglass . (PDF) Fundamentally it's about end-to-end networking, which relies upon a very simple, ubiquitous protocol to run the internet. A protocol that's called (believe it or not) the Internet Protocol. IP is very important. Then, when you're done reading the incredible slides, go over to this page and watch the video! 22 minutes, Real , MPEG-1 .

Heh... this site was static content for like a week there because somehow AxKit got hosed when paul tried to upgrade it using debian apt or dselect or whatever. Man! What a nightmare. The only reason the site was still here was it was delivering from cache -- content generation was totally broken. Now I've got all the code loaded into my home directory and all of the libraries and perl modules are under my home directory so now I have total control and I was able to hack it back into working shape after a few days and some fudging with the server .conf with paul. It's back!!!

(and the wiki's on openict.net too ;-)

If you don't know me very well, you will probably benefit from reading World of Ends .

Many experts like, for example, Bob Metcalfe, the inventor of Ethernet, thought the Internet would collapse. Why does the Internet not collapse? Because everyone has an incentive not to go to a court, and not to go to a government when someone is preventing their packet from getting from A to B. If their packet doesn't get from A to B, they have an incentive to route around it. The Internet treats these things as damage and routes around them. If we decentralize the decision-making process to the participants in the market in real time, as opposed to assuming that ahead of time they can decide what the value is and transact.

Kevin Werbach speaks at the Stanford Spectrum Conference.

Some recent stuff. The wireless-longhaul mailing list hosted at OpenICT.net (a new site). The Longship project to do community development on Chimera. If paul ever gets AxKit 1.6.1 installed :-P then I'll be able to update openict.net with some nifty new axkit features.

It's a bit of a ramble, but The Realities of Online Reputation Management gives a very, very intelligent cross-section analysis of why the internet changes the world.

Ceci n'est pas une pipe -- enfin je le comprends. C'etait vraiement quelques semaines que j'ai venue de comprendre le message de cette image. Voila un collection brillant des peintures Magritte, debutez-vous ici mais prenez soin de faire voire toutes les images.

I finally understood the painting "Ceci n'est pas une pipe" (This is not a pipe) a few weeks ago. Now check out this awesome collection of Magritte paintings. Start here but make sure to look at all of them.

Wise words about UE :

But here is the thing you will learn from really using an OS X Macintosh, and must somehow accept on faith if that's what it takes to get you to Switch: Apple makes design decisions based on a sincere desire to make your life better. Maybe they always did, but they've gotten better at it. OS X is not just less-bad than Windows, it's Good. Yes, Apple also have sleazy marketing weasels, and the salesdrone at the Apple Store in your mall may be the same woeful grade of maladjusted cretin as the one at Best Buy that tries to sell you $49 monster cables for a $59 VCR, but somewhere in California, in the back corner of some office building where they're deciding what should appear on the screen when you click the next button, somebody is asking themselves not only what could appear on the screen that corresponds vaguely with what you nervously hoped you hit the right button to make appear, but what could jump a couple steps forward and startle and delight you.

Yeah... just made some fairly dramatic changes to the XSLT for the blog... I managed to get rid of all the custom XML tags that were being spewed as part of the HTML output, so we're a little closer to validation here. Now I just have a small problem of IDs not being well-formed.

Just successfully moved this whole site into CVS, canned the site, checked it out and here it is. Now I can access the cvs from home and it'll be so much easier to do updates and also I don't have to be paranoid about losing stuff anymore. Cool. Also, I added the Creative Commons license (Attribution) to my ICT essays . Also, yes I know this site doesn't verify as HTML right now, because some of my custom XML tags are still showing up. Keep bothering me and I'll fix it eventualy.

Some very smart stuff about software and infrastructure .

Two new chimera mockups, "Mockup5C1" and "Mockup5C2" ... they are both the same, the second one shows how the "Others..." QuickMark works. Comments welcome!

Update: Are you ready? for ...

What do you think of this hypothetical user interface for Chimera ?

Some thoughts on Using the open source model for (ICT) development . That would be information and communications technology, and the development in this case is the "international" kind.

Looks like I'm not the only person "switching" back to HTML 4.01. Here's why I did it. Oh yeah, and because actually everything's XML on the back end now and it just gets converted into HTML 4 at the very last minute by an automatic process that's part of XSLT. Actually on the server side, all of my markup being served by AxKit is either my own custom XML or XHTML right now. If you care.

I just switched the IT@UW weblog over the AxKit . So now it has the three most recent entries on the "front" page and the "archives" page has the complete set of entries. The data is being stored in the archives page and pulled from there to the front page using XSLT .

I'm starting to see the benefits of this AxKit stuff. Now I have all of my site updates in one file ( oldnews/(index.html) ) and this index page is pulling out only the last 5 (currently) entries automatically. The code is pretty simple: <xsl:apply-templates select="entry[5>=position()]"/> in XSLT. If you go to the Old News page you see if without the select= conditional so it shows you everything.

Another benefit I get is that the code has to be valid in order for AxKit to handle it, so my pages that use AxKit will be valid HTML 4.01 as soon as I work out a couple of little bugs.

I finally grokked the difference between XSL and XSLT. XSLT is a subset of XSL that is just for transforming XML into different XML. That's basically what I'm using (XSLT). XSL adds a lot of other stuff for styling documents. I'm not using any of that (right now anyway...) and plan to continue to use CSS for presentation purposes.

You are looking at this index page generated by AxKit. This What's New entry is in XML. You can look at the what's new stuff all by itself , but you'll see HTML there too right now. I'm typing this into my own handrolled XML format that's parsed by three XSL style sheets in a pipeline before being display, all processed courtesy of AxKit .
By the way, the output appears as HTML 4.01 transitional but I'm moving to an XML backend with XHTML for content for now. I'm just converting it to HTML in the output.

Have you heard of AxKit ? Neither had I. But I'm looking at tools for developing web applications. I'm lazy so I want to do as little work as possible. AxKit seems like a possible answer. Oh yeah, and it has to use perl ;-) No php or java for me. In fact, this site is all php, but really I'm just using it for SSI. Funny thing is that my server has PHP enabled but not apache SSI. Oh well.

Another spec for my hypothetical web application is that I don't want to use SQL. OK, I realize that's completely stupid, because my web app is going to need a database. But frankly I don't like SQL, the syntax is horrid etc. etc. etc.. (thank god I discovered MyPHPadmin). So I'm looking for a way to replace as much SQL with XML as possible... maybe even to the point of writing all my schemas in XML, using some kind of XML query language instead of SQL, etc.

Another reason I'm looking at AxKit is that it's fully buzzword compliant. I'm slowly grokking XML and XSLT (as usually, they are turning out to be pretty obvious once I figure it out) and they do seem like powerful tools. EmbPerl was the other tool I was looking hard at... but there's no HOWTO for OS X, and there is one for AxKit.

In other news, I've made some discoveries about CSS that are making my life a lot easier. Instead of having to put class="foo" in each

tag, I set up a div class that applies certain styles to it's

children. And here's where I finally figured out where the "cascading" part of CSS comes from. It's not in the .css files, but in the actual content where the cascade occurs. Am I the only person in the world who didn't find that obvious from the CSS documents all over the place. So, check out the source code to this document to see how much easier my life is with this new system.

I made a mistake in the final copy of the IT Review Letter, so I corrected it and re-sent it to the Review board. Fortunately just ahead of the deadline!

Just posted the final copy of An Open Letter to the IT Review board at the University of Waterloo . It's pretty long, but apparently it's pretty interesting too... I've been getting emails about it, apparently it's making the rounds even before I finished it. In case you're wondering what it's all about, it's about the IT Review at waterloo going on pretty soon. I had some things that I wanted to say, and this seemed to be a good chance to say them to someone who might listen.

Apparently people are having trouble viewing this site on windows IE. What can I say? It's web standards compliant. Anyway, I added a little blurb to the top of the page that only shows up for me on IE for OS X, but is hidden in Mozilla/Chimera (i.e. webstandards compliant browsers). Hopefully it will show up for IE win people too. Also, I added a 404 page, that you might like. Or you might not. Anyway, you can see it by entering gibberish into the URL bar after the simonwoodside.com/

I've revised the letter to the IT review board a few times.

Draft of an Open Letter to the IT Review board at the University of Waterloo. You can print a nice copy directly from the page if you have a web-standards compliant browser (try it) that can handle CSS2.

Now I've added another style sheet to support printing. Try printing this page or just use print preview. Oh yeah, also I installed a hack on the left-hand navbox sidebar, so that it won't leave trails behind in Chimera (it's called hackbox if you care). Oh yeah, I looked at this page on a windows with IE 6 (on XP) and it was horrible. If you're on windows, check out Phoenix which is a Mozilla variant.

What the heck happened to this website? Well, if your answer is nothing, then congratulations, you are using a standards-compliant browser (XHTML and CSS2). If your answer is anything else, get a new browser. I redid the site in CSS2 and XHTML, and, wow, isn't it nice to get rid of all that tables crap? The only thing I don't like about XHTML is that I have to close all of my paragraphs. Oh yeah, I only did the root pages, so the FTX and D&D pages won't verify until I get around to them.

Added a few Imprint articles that I wrote but weren't attributed to me by name. Mostly co-op related editorials.

I just updated the flow of the site a bit by collapsing the Imprint section into the "writings" section. I also updated the Imprint section so that it's more complete and in chronological order -- so those of you interested in reading the complete archive of S.W. Imprint stories will be very pleased I'm sure.

I just had this crazy idea. Could Mozilla be the next great word processor? I mean, what is there to a word processor? There's a typing interface, a data format, a display engine, and various support tools. Mozilla is a display engine that follows industry standards very well. Those standards -- especially with CSS2 support -- make HTML a fully qualified data format for a word processor. Mozilla has an HTML editor that could be modified to make a good typing interface, so all that's needed are the support tools. The documents are HTML pages with CSS, so that they can be viewed in any browser, published on the web instantly, and edited in a wide variety of editors as long as they support the standards.

D&D fans rejoice ... I just updated the Edion Sourcebook as well.

Well, the transcript project seems to be dead after getting a good 40 minutes down on paper. However, today I bring you an updated D&D subsite ... ye-haw! OK, there's no new content, but the Edion Handbook is substantially easier to navigate and read online now, and the pages are streamlined.

Microsoft / University of Waterloo EngSoc Forum transcript project . A project to transcribe the mp3 recording made by the Feds.

Updated the layout of the site and a bunch of the miscellaneous pages. I discarded some of the really old stuff. You might also want to check out Semacode . Or, you might not.

Updated the ol' site design. Yeah, it looks exactly the same in most browsers, but it's now fully good-looking in Mozilla (on OS X) and it's better designed. Most of the pages meet the W3C spec so I added that to the bottom.

FractalTrees X 1.2 is now available .