You can now read the Architecture of the World Wide Web, Volume One! Esp. useful is the part about Identification and URIs. Rumor has it that Architecture of the World Wide Web, Volume Two features Berners-Lee on Ice, A Blogger Funeral, and Gurus in Space. ;-)

(via Booch and Walash.)

My original article on Reading Files in Internet Explorer with JScript no longer seems to work with IE 6 SP 2. It seems that the behavior of node.nodeValue changed between revisions of Internet Explorer. Therefore use the following pattern instead:

function loadFile(url) {

var newFrame = document.createElement("xml");

newFrame.async = false;

return newFrame.load(url) ? getNodeString(newFrame) : "";

};

function getNodeString(node) {

return (node.nodeValue !== null) ? node.nodeValue : getChildrenString(node.childNodes);

};

function getChildrenString(children) {

var nodeString = "", numOfChildren = children.length;

for (var ii = 0; numOfChildren > ii; ii++) nodeString += getNodeString(children[ii]);

return nodeString;

};

That can be trivially turned back into a single function if you prefer. I think the new code is a little more elegant than the previous version. Here's a demo (that I may take down if I run out of online disk space). I haven't tested that with IE 6 SP 1, since I no longer have access to that version!

Oh, and I gladly dedicate this code to the public domain. It isn't as if I could copyright something that small and trivial anyway, IMHO.

P.S. Slash's support for souce code is still lacking....

I stumbled upon this insightful presentation on better, faster, lighter java based on the book of the same name. It is flash based, but also free and works well on dial-up connections (btw, make sure you turn on your speakers). It inspired me to check out the book from the library.

Java.net's Patterns Community recently (12-12-2004) produced this horribly generic quote on their front page! Sounds like something a PHB would say, methinks.

JavaPedia: Patterns
One of the richest pages in the JavaPedia is the page devoted to Patterns. A recent post to the discussion suggests you "refactor toward that pattern or away from it depending on whether the pattern's consequences, strengths, and trade-offs are appropiate for the program."

I wonder who the idiot is who wrote that... Oh ya, it was me! Not the best quote of the world esp. taking out of context like that. Hopefully people will follow the link and read the rest of the paragraph. Since that quote is awefully generic, I'll elaborate:

E.G. Initially one JavaScript library I was developing used a Strategy for determining the method of loading external files (hard to do in JS). I didn't intend to use the pattern. It just came about after some coding, so I formalized the relationship.

Furthermore, the concrete strategy was selected by a Factory and a Little Language; it was a really small language and took about ten minutes to code and test. Again, the patterns weren't inserted on purpose, they just emerged. However, while it seemed a good idea at the time, it ended up complicating things. So I ended up using a simple if/then/else statement! The obvious solutions are just missed sometimes after a long time coding, I guess.

I hope that makes more sense now.

P.S. Just joking with you, Java.net admins! :-)

Addendum - I think I figured out why they used that quote. Java.net is currently pimping an excerpt of Refactoring to Patterns by Joshua Kerievsky. I swear that I never heard of it before today. Will check out after I finish Better, Faster, Lighter Java. (QJ 14-12-2004.)

Nemo has finally found a home! It is currently incubating at Java.net. The project page contains the current source, and you may want to read the plug at RDF-Interest for other info. It requires Saxon 8.11 and a build of the Jena2 module checked out of Jena's CVS.

Here's something I hacked together in the last two days (after a few weeks of learning). The project is called Nemo, since I just watched the excellent movie, Finding Nemo. :-) It is a set (well, just two right now) of Saxon extension functions that use the Jena framework to access RDF/XML datasources in XSLT stylesheets. This would allow much more flexible transformation involving RDF data, like the Dublin Core. You use RDQL to query datasources. The project is inspired by (but not based on) RDFTwig from Norman Walsh.

It isn't that impressive right now, but I have a lot to do (1st priority - get a domain so I can create a good package name). The program copyrighted by me, James Francis Cerra, and this preview should be considered under the GPL, Version 2 (get the license text from the link). I currently intend release Nemo under a less restrictive license later (after I put more thought into it). Here's the code for the two classes:

File: Nemo.java

import java.io.FileInputStream;
import java.io.FileNotFoundException;

import com.hp.hpl.jena.rdf.model.Model;
import com.hp.hpl.jena.rdf.model.ModelFactory;
import com.hp.hpl.jena.rdql.Query;
import com.hp.hpl.jena.rdql.ResultBinding;

public class Nemo {
static public RDFSequence find(String sourceURI, String queryString) throws FileNotFoundException {
Model model = ModelFactory.createDefaultModel();

model.read(new FileInputStream(sourceURI), "", "RDF/XML");

Query query = new Query(queryString);

query.setSource(model);

return new RDFSequence(query);
}

static public Object get(ResultBinding result, String val) {
return result.get(val);
}
}

File: RDFSequence.java

import net.sf.saxon.om.SequenceIterator;
import net.sf.saxon.om.Item;
import net.sf.saxon.value.Value;
import net.sf.saxon.xpath.XPathException;

import com.hp.hpl.jena.rdql.Query;
import com.hp.hpl.jena.rdql.QueryEngine;
import com.hp.hpl.jena.rdql.QueryResults;

public class RDFSequence implements SequenceIterator {
RDFSequence(Query q) {
initialQuery = q;
QueryEngine qe = new QueryEngine(q);
results = qe.exec();
currentItem = null;
}

public Item current() {
// Get the current value in the sequence (the one returned by the most recent call on next()).
return currentItem;
}

public RDFSequence getAnother() throws XPathException {
// Get another SequenceIterator that iterates over the same items as the original, but which is repositioned at the start of the sequence.
return new RDFSequence(initialQuery);
}

public Item next() throws XPathException {
// Get the next item in the sequence.
if (results.hasNext()) {
currentItem = Value.asItem(
Value.convertJavaObjectToXPath(
results.next(),
null
),
null
);
return currentItem;
} else {
return null;
}
}

public int position() {
// Get the current position.
return results.getRowNumber();
}

private Query initialQuery;

private QueryResults results;

private Item currentItem;
}

Here's an example on how to use it:

<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:jfc="java:Nemo"
extension-element-prefixes="jfc"
exclude-result-prefixes="jfc xsl"
>

<xsl:variable name="query"><![CDATA[SELECT ?x, ?y
WHERE (?x <http://www.w3.org/2001/vcard-rdf/3.0#FN> ?y)]]></xsl:variable>

<xsl:variable name="datasource" select="'datasource.rdf'" />

<!-- Try to make the output look half decent -->
<xsl:output method="text" />

<xsl:template match="/">
From:
<xsl:value-of select="$datasource" />
Query:
<xsl:value-of select="$query" />
Result:
<xsl:for-each select="jfc:find($datasource, $query)">
Name: <xsl:value-of select="jfc:get(.,'y')" />
URI: <xsl:value-of select="jfc:get(.,'x')" />
<xsl:text>
</xsl:text>
</xsl:for-each>
</xsl:template>
</xsl:stylesheet>

Hope you find that useful, and please excuse slash's lame support for inline code blocks.

Norman Walsh has some intersting thoughts on the next version of XML. Woa: déjà vu! I think it is clear that the next version of XML will use namespaces for defining entities. Now how they are resolved is another matter....

If you want insightful comments based on practical experience and factual evidence, don't come to /. - as evidenced by the recent article on Yet Another RDF Browser. So many people just don't get it!

Take this fellow who thinks that XSL is gathering dust. I guess it isn't used much with document markup languages such as Docbook - Oh wait O'Reilly does. Idiots. :-{ I guess Mozilla and mozilla-derived browsers don't access bookmarks, browsing histories, persistant data through RDF - Oh wait Mozilla does. I guess too much metacrap will overwelm the semantic web - Oh wait RDF is just like a more advanced version of the WWW. I'm not the only one who saw that. I guess Google is doomed, right? :-/

If /.'s collective consciousness is this bad with things that I am well versed in, I wonder about the subjects that I don't know much about...

Danny Ayers recently posted article, Exorcising QNames.

First the background. Recently, the xml:gods have been co-oping XML Namespaces for abbreviating URIs in many RDf serializations and XML Schema languages. Unfortunately, experience is starting to show that using QNames in XML content many be a terrible antipattern. Even the xml:gods are starting to doubt whether or not to overload QNames.

Now Danny Ayers has presented an interesting alternative using an element to map the results from a prefix to a namespace. I must be too critical, as I think his idea is a little clunky, while his comments are generally supportive. For one, his idea doesn't allow the abbreviations on the root element of a particular scope. In that case, it must be placed outside the intended element:

<animals
xmlns="http://example.com/animals/"
xmlns:nicer="http://purl.org/stuff/nicer/"
>
<nicer:map
nicer:prefix="cls"
nicer:namespace="http://purl.org/stuff/colors/"
/>
<dog
xmlns="http://purl.org/stuff/dogs/"
paw="cls:golden"
>
<nose>cls:brown</nose>
</dog>
</animals>

The mechanism also just seems to be a replacement for entity declarations. Well then, why not use entity declarations! That SGML-derived syntax is built-into every conforming XML processor, by definition! Furthermore, one doesn't end up polluting the DOM very much with them. These seem like concrete benifits over his proposal. E.g.

<!DOCTYPE animals [
<!ENTITY cls "http://purl.org/stuff/colors/">
]>
<animals
xmlns="http://example.com/animals/"
>
<dog
xmlns="http://purl.org/stuff/dogs/"
paw="&cls;golden"
>
<nose>&cls;brown</nose>
</dog>
</animals>

But we all know that DOCTYPES are evil. (right? right?) And of course these "entities" are really meant as namespaces for xml content rather than shortcuts for replaced content; even though, they act as normal entities. If XML 2.0 is not going to be backwards compatible with XML 1.0/1.1, then I propose that the xml:gods at the w3c (asgard) combine these "weak entities" with xml namespaces. Use &name; for traditional entities and %name: for namespace entities (again, which act the same in effect if not in intent):

<?xml version="2.0" ?>
<animals
xmlns="http://example.com/animals/"
xmlns:cls="http://purl.org/stuff/colors/"
example="&cls:ex"
>
<dog
xmlns="http://purl.org/stuff/dogs/"
paw="&cls:golden"
>
<nose>&cls:brown</nose>
</dog>
</animals>

Of course, it is too late for that solution. :-(

XML 1.0/1.1 could still be used if another namespace and the % sign was overloaded as an escape marker. e.g.

<!DOCTYPE animals [
<!ENTITY ent "brown entity">
]>
<animals
xmlns="http://example.com/animals/"
xmlns:cns="http://example.com/contentnamespace/"
cns:cls="http://purl.org/stuff/colors/"
example="%cls:ex"
>
<dog
xmlns="http://purl.org/stuff/dogs/"
paw="%cls:golden"
>
<nose>%cls:&ent;</nose>
</dog>
</animals>

Then again, two independent escaping systems in one language is silly! Another downside is that a new entity resolver step has to read any of the DOM before handing it's off to the rest of the application. I think that those disadvantages would lead to any significant adoption. So we're stuck with XML ENTITIES for the immediate future.

P.S. Slashdot's comment system still sucks. Ignore any nonsense "&ampn bsp;" and white spaces in the code. Arrrrrrrr!

This is a response to Lach's comments of my response to Simon Willison's blog on Presidental Candidate Endorsements. I am posting it here out of courteousness, since it goes off-topic. Read his comments for the appropiate context, else this seems like jiberish (well, even more jiberish than usual ;-).

Lach, I was confused by Simon Willison's intent. I agree with him that disappointment is a better description of his response. In fact I probably wouldn't have commented if that was his term. However, I want to respond to your comments and questions - even if just to clarify my thoughts.

1. That's disapointment. If you feel profound sadness - i.e. depression - then it's depression.
2. Dialectical Behavioral Therapy suggests that one's feelings are based on their thoughts. So if Simon was depressed then, from my personal experience, one possible reason would be wanted validation. Since he was not depressed, this is a moot point. (I should be calling him Simon Willison or Mr. Willison since I never met him, but oh well.)
3. I never said that they were perfectly equal in good and bad points; however, I agree that I used the wrong term. I was trying to say that any reasonable analysis must conclude that different opinions - i.e. favorable or unfavorable opinions of a particular candidate - can each be formed from rational arguments. Which leads to...

4. I'd like to state that I'm on the fence right now on choosing either Bush or Kerry. Not because I haven't givin this much thought, but because I agree with Bush on certain issues, Kerry on others, and neither on still more issues. So I am examining how important these issues are and how they interact. That is a hard analysis especially since some of my opinions are probably either wrong or poorly derived.

   However, my experience with Bush's supporters is very different from yours. Most seem to be normal people - neither overtly religious or of exceptionally below-average intelligence. (In fact, on average they seem to have average intelligence - in agreement with the definition of "average intelligence".) However, this has nothing to do with a positive opinion of Bush. To think that it does is a logical fallacy.

Hope that clarifies my thoughts.

Dave Shea of Mezzoblue recently posted a blog about long-term love affairs and exploration. Doesn't that sound dirty? ;-)

Actually, he was ranting on how much he enjoys the pursuit of knowledge. Dave Shea Sites says he could spend hours upon hours reading triva from sites such as Wikipeida and the Internet-Movie Database among others. Give them a ring.

I can relate: I've spent hours of joy researching the bounty of information located not only on web sites but also in journals, magazines, and in the bliss of the library. However....

Warning: The following opinion is exaggerated for effect.

I don't think that kind of exploration is as noble and fun as he makes it out to be. It certainly feels enriching; however, reading such trivia doesn't help in my understanding of it. Not really. Nothing is really discovered - just taught through literature.

The kind of exploration that I really enjoy is though experimentation and trial. I don't just read about cooking but create a meal. I don't want to just learn about the circuit but wire the thing together and view the results. I don't just want the physics taught to me; I want to derive the equation myself. I don't just learn XUL - I write apps with it.

I find experimentation much more satisfying. Researching is important and rewarding; however, it is mostly useful to me as a guide, overview, or link for free association. I want to understand a thing and not just learn about it!

I just don't understand /.'s moderators. Why was this post on Google's browser possible based on MSHTML moderated as a troll by someone? I sincerely meant everything that I said. I tried to be polite and not condescending - even proactively admitting a social blunder. I also backed up my analysis with some (admittly, anecdotal) evidence. I simply don't understand why it is being moderating like that. What am I doing wrong?

Several of my other posts in the last couple of weeks were moderated down for suspect reasons. I never try to hurt anyone's feelings. This is making me a litle paranoid. I am beginning to feel like an outsider in these discussions... (ironic from a "nerd's news portal.") Why is that?

I know not many people read my journal, but I'm still curious. What books does everyone have in their personal libraries, and what are their publishers?

Internet Explorer is a strange beast. I happened to be writing a DHTML Behavior to fix up some browser incompatibilities. One thing it had to do was download a CSS file and store it as a string for interpretation later. If ActiveX was enabled, then it is simply a matter of using the (relatively cross browser) XMLHTTP object:

function readFile(url) {
var httpRequest = new ActiveXObject("Microsoft.XMLHTTP");
httpRequest.open("get", url, false);
httpRequest.send();
return httpRequest.responseText;
};

In a perfect world, I'd use this one. On the other hand, in a perfect work everyone would be using Mozilla to browse the web. ;-) My script must work even if ActiveX is disabled, which complicates things. There is also a built-in download behavior, but it still doesn't work if ActiveX is turned off. Custom DHTML Behaviors fortunately still work as long as active scripting is enabled, so not all hope is lost.

Microsoft jumped on the XML bandwagon early while developing Internet Explorer 5.0. Their first attempt used something called xml data islands. Instead of loading an XML document directly into the browser, they used an extension to the common SGML version of HTML - an <xml> element. It's sort of like <script> for XML: it can be wrapped around the XML or just link to an external document.

It also exposes a limited interface to an XmlDocument class. Since the object is represented as a SGML element, we can create one with the document.createElement("xml") factory method. There's also a nice side effect, where its interface still works even before being added to the DOM tree. (Some elements, like <iframe>, must be added before being used with penalty of an Unknown Error from IE.) Thus we could load any XML data - using the parts of the XmlDocument interface that work - with the following function:

function readFile(url) {
var newFrame = document.createElement("xml");
newFrame.async = false;
if (newFrame.load(url)) return newFrame.xml;
else return "";
};

Loading the CSS file now would result in an error since it isn't an XML file. (This strict parsing rule - unlike HTML - is actually the correct behavior according to the spec). Unfortunately, it can't be fixed without the ability to resolve external entities, which isn't available for security reasons. The solution is a compromise by adding a few lines to the files so IE thinks they are valid XML documents. This doesn't mean that the CSS files have to be invalid. Thank the W3C deities for adding comment declarations to the core CSS grammar for compatibility with SGML HTML. Internet Explorer's xml processor returns the comments (even though it doesn't half to), so the style sheet data can be inferred from the following:

Beginnning:

<!--

Ending:

/* --> <style/> <!-- */ -->

The SGML-style comments hide the CSS-style comments and the rest of the code. Retrieving the valid CSS is as simple as reading the value of each XML node:

function loadFile(url) {
var nodeText = "", cssText = "";
var newFrame = document.createElement("xml");
newFrame.async = false;
if (newFrame.load(url)) {
var children = newFrame.childNodes;
var numOfChildren = children.length;
for (var ii = 0; numOfChildren > ii; ii++) {
nodeText = children[ii].nodeValue;
if (nodeText) cssText += nodeText;
}
}
return cssText;
};

That's easy from the developer's point of view; however, it's an ugly ugly ugly hack for the user (CSS author). Depending on your requirements and taste, the following hider may be easier to stomach:

Beginnning:

<!-- /* --> <style> */

Ending:

/* </style> <!-- */ -->

Either case doesn't matter to the function above. Just be careful if you're having lunch. ;-)

P.S. I plan on adding this article to my program's documentation. Help editing it and constructive criticism would be much appreciated.

Note: A bug with /. puts a typo in the first code block. That should be "open" not "o pen". Cowboy Neil must hate me!

**IMPORTANT** THE ABOVE NO LONGER SEEMS TO WORK WITH INTERNET EXPLORER 6 SP 2. USE THIS PATTERN INSTEAD