Archive: December 26, 2003
Jon Udell: The Social Life of XML
I get sick to my stomach when Udell gushes over XML like this, as he does often. Yeah, XML is useful, sure. But I have a fundamental disagreement with how he looks at it. Maybe I'll say more after I finish reading it.
Anyway, my view on XML is largely that it's plumbing -- it's nothing important in and of itself. And it certainly isn't a database.
Update (Jan 6): I just wanted to moderate my above statement a bit  I didn't mean to be so harsh on Jon. (Data-oriented, as opposed to document-oriented) XML is useful because it gives a shared, human readable syntax for data, but it's overly cumbersome for that purpose, and its data model is pretty heinous. For documents (things that require mixed-content) XML is just dandy. But for data, XML is far from ideal, overly complex, and usually far too verbose as well.
Howard links to the "federal government's opening brief on the merits in the Pledge of Allegiance case pending in the U.S. Supreme Court" in PDF form.
I find I don't have an understanding of exactly what this represents. Is this a summary of the defendant's case put before the Supreme Court so that they can decide whether or not to hear the case in the first place? Or is this involved in some other step of the process?
One of the recent improvements to Linux I've been most interested in has been the O(1) scheduler. Via Slashdot, Ars Technica has an explanation of the O(1) scheduler. Of course, they don't listen to Nielsen and the rest of us and have no permalinks for this article. Judging by their previous URLs, however, the permanent home should be:
http://www.arstechnica.com/etc/linux/2003/linux.ars-12242003-1.html
Man, they don't even have a "printer-friendly" version.
Update: Man that was lame. The article had almost no meat. After explaining processes, Big-Oh notation, timeslices, etc. it finally said something like "So that means that it takes a constant time to choose a process from the queue". Duh. Thanks for nothing Ars Technica. This article wasn't an explanation of the O(1) scheduler at all, and gave almost no details about it.
Read The Wonderful World of Linux 2.6 instead. Much better.
You can gain a lot of flexibility in your CSS by being able to mix and match classes in your HTML. To include multiple classes in one html class= attribute, just include all the classes you want separated by spaces.
Then, to refer to an element that has multiple classes set, the CSS selector syntax is element.class1.class2 (class1 and class2 can be in any order), and of course, element can be omitted and the first part of the selector be left as just "*." or "."
Opera Proposes XForms Alternative: Web Forms 2.0 (via Erik)
Ian Hickson who works for the Opera browser company
(in Norway) has spent the last few months on an
alternative XForms proposal called Web Forms 2.0 that
extends HTML4 and rejects the XML above-all-else craze
(eg. namespaces, XPath bindings, scripting is evil,
and so on)
Sounds good to me
For the past few months I've been working on a
proposal to extend HTML4 Forms in a backwards
compatible manner, to address the needs that were not
covered by XForms 1.0.
Note: This specification is not just blue-sky work,
there is a good chance that large parts of it will be
implemented in several user agents in the medium-term
future.
Ian also lets us know that Opera has no intension to
add the bloated, hypercomplex XForms machinery to its
browser offering anytime soon.
Here's the proposed specification: Web Forms 2.0
|
Generated in about 0.054s. (Used 7 db queries) |
Friday, December 26, 2003
Permalink
new⇒The Elegant Universe
Well I have finally found the crazyguy that preaches useless nonsencein A...
Joseph Baxter: Jan 7, 11:07pm