What’s your favorite way to assign identity to XML elements?

Related link: http://edition.cnn.com/2004/TECH/ptech/04/28/dvd.censor/index.html

CNN is covering a DVD player that automatically edits DVDs to remove “potentially offensive content”. Apart from the obvious howls, I’m thinking that something different is happening to culture today. People on all kinds of levels are actively filtering the contents of all kinds of media. For some people it’s violence in movies, for others it’s radio and television, and for some of us it’s spam of whatever variety. Caller ID gave people the chance to filter their phone calls.

I don’t look at all kinds of filtering the same way. I can’t, for example, imagine using the ClearPlay DVD player myself, though I use spam filtering and non-destructive email filtering all the time. Filtering is a common coping strategy for dealing with all kinds of aggravations, including the sheer volume of information flows we have to deal with today.

While it’s certainly a coping strategy, I have to wonder what it means that there is so much out there that we feel we have to cope with. The ubiquity of filtering in all its forms suggests that a lot of information sources, intentionally or not, are perceived as hostile by the people receiving them. Spam comes with having an email account, while sex and violence come with movies. (Heck, DVDs won’t even let you skip the ads!) Junk mail comes with getting the mail, and bills of all flavors now include advertising flyers.

I don’t sympathize with the ClearPlay agenda, but I do think it’s worth pausing for a moment to ask why so many people are putting so much effort into filtering away parts of the world they find unpleasant.

Is filtering a cure or a symptom?

Related link: http://www.zoomon.com/files/Zoomon_SonyEricsson_PR_final_19_April_2004.pdf

While I don’t want to steal my own thunder — I’ve got two XML.com articles on SVG Mobile coming up in the next month — I’m happy to relay the news of SVG Tiny finding its way onto the new Sony Ericsson K700 phone. This news comes to us hot on the heels on previous announcements of other phones shipping with SVG Tiny support: the Siements CX65 and the Sharp V601SH (in Japan). This is excellent news for SVG and mobile applications in general, and it’s only getting started, expect many, many more such announcements during the year. Cheers!

Excited about SVG Tiny finding its way onto major mobile phones?

Related link: http://simonstl.com/dryden/localBlogs.html

While weblogs’ ability to connect people regardless of their geographic location has produced some fascinating stuff, I’m looking for weblogs that focus squarely on a particular area.

Jon Udell had some kind words last week for local weblogging, mine in particular, and I brought up local weblogging at the Digital Democracy Teach-in back in January, where it seemed pretty popular.

So far, it seems popular in that people think it’s a good idea - Glenn Reynolds encouraged blogging on local politics in February - but I haven’t seen a large wave of people actually doing it.

I’d like to think that I just haven’t looked hard enough, and that maybe it’s something difficult to find in Google and similar tools. I’ve found some weblogs already, but I’d like to add more to that list.

My criteria are fairly simple. The weblog should focus primarily on local politics, where local is something smaller, preferably much smaller, than a county, state, or province. I don’t mind pointing to subcategories of weblogs with broader perspectives, so long as what I’m pointing to is mostly local. Local weblogs can be from any country, not just the US, though that’s what I’ve listed so far. They don’t have to be in English, and they don’t have to focus on politics, either.

It’s great to publish material that can reach a wide audiences. Sometimes it’s also great to publish for a smaller audience.

Know of any locally-focused weblogs?

Related link: http://www.w3.org/2004/04/webapps-cdf-ws/

The W3C is organizing a workshop on Web Applications and Compount Documents. Some of you may breathe loudly in relief and think “finally!” and some of you may well just wonder what the hell this is for!

This workshop is about two individual topics that could benefit from a general thought process from interested parties in the community. As you probably know, the W3C has recommended quite a few bitching specifications aiming at solving different problematics for authoring contents for the World Wide Web which quite rapidly also became building blocks for developing Web Applications: HTML then XHTML, SMIL, SVG, XForms, etc. Add to this specs like CSS, XSLT, XML Schemas, XBL, etc.

Ideally, all these specifications, with the help of their XML foundations, were leaving the door open for mutual integration with the concept of namespaces which should let you solve multiple document authoring problematics with the right technology all within a single document. This works pretty well when a mixed-namespace document is consumed by a processor that will probably not render the document and only look up the XML structure and data.

However, in practice, it has been found that the rules for integration were often under-defined — if defined at all — in the context of rendering this type of document, say when a human being opens the said document in a web browser. For instance, how can you reconcile different rendering models, event flows, etc? This all is made more difficult by the hope that all this negociation should be done in a magical, transparent manner to the consumer of the document.

Another issue too is that when authors find a way to hack up the right bits and bolts into one document, they probably take different approaches to the development of a Web Application, and maybe it would be desirable to have a common framework or approach. I won’t list more issues here as there are really quite a few of them and I’m sure that a lot of you here have enough background in authoring both documents and applications for the WWW to know most of these.

So this workshop is a good opportunity to get the ball rolling and defining the real hot topics and use cases and ultimately the requirements and maybe even W3C recommendations for what the XML-on-the-client Web should turn out to be. Better still, you get the chance to share your views as this workshop is open to the public. Time to speak up and make a valuable contribution from your valuable experience.

Does it sound like the right time for such a workshop? Do you think the W3C can/should help in these matters?

Related link: http://www.oasis-open.org/committees/download.php/6037/office-spec-1.0-cd-1.pdf

Yesterday, I was looking at the OpenOffice file format specification which was issued as an OASIS committee specification on March 22. If I understand the wording of the spec correctly, RELAX NG is now the normative schema language for OpenOffice. Fragments of the normative RELAX NG schema appear on nearly every page of the spec.

It seems like everytime I turn around, RELAX NG is creeping up on the XML landscape. And not as a result of heavy handed politics, or a brilliant marketing campaign. Nope. It is just what it is, a solid, technically sound, irresitable little schema language. In a world awash with XML-related specifications, it is nice to find a stand out, an example of brevity, clarity, and usability. I consider myself very lucky to have had even a tiny part in its development, but really we have James Clark and Murata Makoto to thank for RELAX NG.

Three years ago, I remember dialing in to the first what was supposed to be TREX technical committee meeting. James and Makoto were already on the line. After a little chatter, the first order of business was James’ announcement that he and Makoto had decided to merge their earlier efforts, essentially marrying James’ TREX and Makoto’s RELAX, and rewrite the charter of the committee. There was no jockeying for position, just a pair of gentlemen who shook hands. A rare thing in this ego-sick world — and look what it gave us.

Now I hear that James Clark has been working on StAX, the Java-based pull-parsing API for XML (JSR-173). I’ve got my hopes up again.

Does anyone out there wear a RELAX NG T-shirt?

Related link: http://www.adaptivepath.com/publications/essays/archives/000315.php

Jeff Veen has captured a painful bit of experience that too many enthusiastic managers, captivated by technology, often fail to see: “Turns out, after all the budget and time we spent, we really didn’t need a content management system at all. We just needed some editors.”

The urge to minimize human intervention has been a key driver in a lot of technology. In my experience, in a slightly different field, a lot of the worst aspects of XML technology and practice today derive from a studied effort to automate as much as possible, removing humans from the loop wherever it seems it might work. XML is treated as just a data format, something for computers to process, and the notions that humans might actually benefit from touching their data directly or creating their own custom structures are repressed. Enthusiasm for using markup and data structures to solve local (rather than utterly standardized) problems is seen as a problem, not as a solution.

At the content management level, similar problems come into play, whatever the underlying format of the data. Technology forces a choice between clarity (standardized workflows with predictable behavior) and flexibilty (open workflows with lots of options), losing the kinds of changeable interactions that are possible when humans talk directly. It’s hard to build excitement about pouring information into a computer for it to be reprocessed somewhere else in the company. By replacing expensive humans with supposedly cheaper computers, these systems tend to create new costs.

Veen’s core suggestion makes sense:

Content management is not a technology problem. If you’re having trouble managing the content on your Web site, it’s because you have an editorial process problem. Your public-facing Web site is a publication. Treat it like one.

Solve publishing problems with people whose job it is to create great content, not technology that hopes to extract content from people along the way. Get an active group of people who care about the problem, not a filtering system that hopes to extract and maybe polish content from people whose focus is likely elsewhere. Expect technology to serve as a useful adjunct to a successful project, not as the heart of the project itself.

(Thanks to Dorothea Salo for posting on this.)

How much do people matter?

Related link: http://x-cp.org/

It’s been a long time since I’ve seen an XML project that really applies XML in new ways to the heart of an old problem, but I’m happy to report that XCP is using XML in ways that go well beyond the Web Services vision.

Instead of applying XML to data problems where people expect transparency, the ><CP consortium takes XML into areas where people have grown used to opacity and pretty much come to expect that they’ll either have to use tools or read binary as it comes off the wire.

I also have to give them a lot of credit for taking this further than the Web Services folks have. Web Services is about connecting organizations that have the time to publish services or write (or buy, borrow, or steal) code that consumes those services. The ><CP work aims to connect everyone with XML, working at a lower level in the network stack than Web Services, which seems to keep building the stack ever higher.

Admittedly, they haven’t gone as low in the stack as an older IETF proposal for XML at the IP layer, but I don’t think anyone thought that was practical anyway.

They also deserve credit for thinking ahead to merchandise, making it easy for people to look like they’ve studied this deeply at a conference somewhere.

I haven’t given a lot of endorsements over the years - I think the last one was for some version of SVG - but I’m happy to have signed on with this group of distinguished folks.

Have any innovative uses for XCP?