My first Cocoon patch

My first Cocoon patch just got accepted. It fixes the xml2html stylesheet to work with a quirky browser called IE. Gory details: And there was much rejoicing. :) -m

Personal content management and repositories

When you are setting up software for your personal information, say, a blog, what is the single most important design factor, far and away?

It's the data, stupid. I'm still amazed at how many people will fritter their important data--stuff they access many times a day--away in some database, or note-taking application, or proprietary word processor format. Plain or slightly structured text is a big deal.

The one exception, maybe, in certain cases, is open source software with a healthy community in place. Even then, my theory is that over time Danny O'Brien-class Life Hacks geeks will extend and mold the open source software to something that can work with plain text.

My next research question: how does the above fit in with (JSR-170)? -m

I have no key and this is why

One thing that stood out at ApacheCon was my lack of a private key. I'm not saying I'm against keys or key signings or that I will never have one--just that I currently don't for a couple of reasons.

First, I have never needed one. I can happily sign on to IRC with the nick 'mdubinko', and after a few sentences folks know it's me. I can release code and web sites, and so far nobody has needed to question whether it really came from me.

Second, it would be the something that could really hurt me if it doesn't stay private. A half-written manuscript, or some open source code isn't that big a deal if someone snoops it. A private key is.

Third, once it starts getting used, it would be majorly painful to lose. Pretty much everything I'd ever signed would become in-validatable, right? I have a triply-redundant backup scheme in place after my earlier episodes, but not a secure backup plan. (See point #2). A regular copy of my files to my iPod, backup server, and remote repository are great, but hard to lock down.

Maybe somebody will show me the light and I'll get a key soon. Maybe not. Either way, I'm willing to be educated... -m

More XML 2004 goodness: XMLwiki

Rick Jelliffe strikes again. As a follow on from his earlier article on, the Topologi booth at XML 2004 had some fliers for a more productized version of the Wikiml -> XML via SGML concept. This one had an XML configuration file (with a root element named 'psychi' for some reason) to configure all the parsing parameters, making it easier on non-SGML-heads.

Link: I'd love to see a Python port of this, hint hint. -m

Sun demsonstrates comprehensive XForms support in OpenOffice

At XML 2004, Lars Oppermann demonstrated Sun's XForms work, and it is impressive.

The next release of OpenOffice, available as soon as this spring, includes a full XForms editor and engine within Writer. It has all the things you'd expect, like a tree-control XML designer, property sheets to set up all bindings, and an events editor. The standardized file format will include XForms markup. So by next year, both the major open source browser and open source office suite will support standard web forms.

The audience reaction was stunned amazement; apparently I wasn't the only one taken by surprise. I could use more surprises like this! -m

Introduction to XML Events

Now up on IBM developerWorks, my article on XML Events. I'm especially proud of this one, as it contains some research and straight facts that you need to read between the lines of the spec to figure out. Link: Read on for a special hint...

In the graphic, the capture arrows are green, as in Mozilla lizard green, and the bubble arrows are blue, as in spinning "e" blue. There's conflicting information on the internets as to which browser first came out with which, but I took Danny Goodman's JavaScript Bible as the definitive source. -m

W3C and ASF, compare and contrast

At my first ApacheCon, I noticed something: there's hardly any W3C people here. I did see Roy Rielding and David Fallside, and it's not like I could recognize every W3C person on sight, but the observation was striking, and made by others as well. One person I was discussing this with said--completely off the cuff--"theory vs. practice".

I think there's more than a grain of truth to that. "Academic vs. pragmatic," another one put it. (Roy, I think, but don't quote me) Now that I have one foot each in the waters of W3C and the ASF, I'd like to help both sides engage a broader view, at least as far as my tiny sphere of influence goes. -m

Update: lots of W3C people are at XML 2004, so maybe that's a factor in reduced overlap this time.

Random Pre-XML 2004 thoughts

I enjoyed Norm Walsh's musings on a hypothetical XML 2.0. As I head off to a week of travel, a few thoughts of my own.

Norm essay is at and includes a healthy comments section at the bottom. One thing would suggest is geting rid of the biggest problem in namespaces -- the three-part names. Fully spelled out "Clark syntax" gets to wordy, so how about just using the DNS parts that give uniqueness and persistence. Think of how Java uses namespaces. Nobody ever complains about Java namespaces--they're practically invisible. This is the primary quality you want in a namspace spec.

Anyway, like Norm says, just something to chat about will fellow XML-heads. -m

Go get Firefox. Now.

If you're reading this in IE, drop what your're doing, click over to and click on "Free Download". Install it, and run it. It's really that much better. -m

XML 2004, here I come

Same week as ApacheCon. So, yes, I'll be a little busy next week. If you're trying to get in touch with me and don't get through, now you know why. -m

ApacheCon, here I come

Kind of a last minute thing, but I'll be at ApacheCon in Vegas, Sun through Tues. Drop me a line if you want to meet up. -m

UBL 1.0 Ratified

The Press release is out: I've always considered UBL to be a key technology, especially when combined with things like XForms. Combine that with the free UBL transformer at and you're off to a great start. -m

Apache, here I come

I have signed a deal that will allow me to contribute a significant (though less than full-time) amount of work to the Apache Software Foundation, specifically in the area of forms, and very specifically in the area of XForms. I'm pretty excited about this...

Doing this feel a little bit like bungee jumping, though getting into any new codebase feels like that. -m

W3C, here I come

Part of the same deal marks my return to the W3C, joining the XForms and XHTML Working Groups, which I hear are always on the lookout for more helping hands. -m


Terms of use

For external use only. I doubt the enforcability of click-through licenses anyway. Copyright 2004 Brain Attic, L.L.C. All rights reserved.


Older stuff here