A Bunch Of XML Recommendations

KjetilK writes "During the past couple of days, the World Wide Web Consortium, have advanced several core XML-specifications to Recommendations. You have the Extensible Markup Language (XML) 1.1 and Namespaces in XML 1.1 as well as XML 1.0 Third Edition. In addition, XML Infoset Second Edition is now a Recommendation and VoiceXML 2.0 is now Proposed Recommendation."

My recommendation

[DrSkwid]

use csv

Re:My recommendation

[eraserewind]

Hey, c'mon you've got to be at least 3 meta levels above that if you want to be taken seriously by the w3!

Having fun with TLAs...

[neglige]

So application A communicates with application B by checking in the data and B checks it out again? Well, at least you could roll back the entire communication between A and B and see all the changes...

Re:Having fun with TLAs...

[neglige]

I just noticed I misread your comment. Maybe on purpose, maybe because I'm low on caffein.

Re:Having fun with TLAs...

[DrSkwid]

hehe

I have no idea *what* you're talking about :)

in any case, I really meant tab delimited not csv as such

well, any plain text will do

we're waiting for the ultimate : java bytecode as XML

Re:Having fun with TLAs...

[neglige]

CSV vs. CVS ;)

File that under "failed attempt at being funny" *sigh* Next time will be better, I promise ;)

go standards!

[Chris_Jefferson]

So, XML is the "standard to end all standards, the way in which every single application can communicate with every other application from now till the end of time"

Until they bring out XML 1.1 of course, and now half the apps will be 1.0 complient, and the other half 1.1, therefore destroying the main purpose of XML.

I'm not saying that this update wasn't probably necessary or useful, but all those people who were using XML because "Its the most standardist standard of all time!" should probably take note.

Re:go standards!

[noselasd]

>So, XML is the "standard to end all standards, the way in which every
>single application can communicate with every other application from
>now till the end of time"
Well, many people thinks so. But XML sucks [xmlsucks.org] in many of its current use.

Re:go standards!

[jilles]

The list of changes is pretty much limited to being more tolerant to newer unicode specifications (which xml depends on and which werent available at the time xml 1.0 was finalized)and to avoid similar adjustments to the spec in the future. I.e. any well formed xml 1.0 document is also xml1.1 compliant. There may be some exotic xml1.1 documents (using some unicode exotic characters) that are not well formed xml 1.0. This change is very defensible, unless you are willing to freeze the unicode spec four years ago. The changes have no consequences whatsoever for the validity of existing xml documents (correct me if I'm wrong) and they should work exactly the same (namely as specified by the w3c) in xml1.1 processors.

XML 1.1 Not necessarily a good idea

[Anonymous Coward]

I'm just going to syndicate Elliote Rusty Harold [cafeconleche.org] [scroll down to the Feb. 5th entry] on this one and pass along his suggestion that you don't use XML 1.1; Xerces 2.6 [apache.org] will process it, but most things won't, and most of the benefits of what's new in XML only apply if you're putting your documents into a few (mostly Asian) languages.

Re:XML 1.1 Not necessarily a good idea

[axlrosen]

most of the benefits of what's new in XML only apply if you're putting your documents into a few (mostly Asian) languages.

And it's not Chinese or Japanese or Korean, it's more esoteric languages like Mongolian, Yi, Cambodian, etc. And the benefit only applies to element and attribute NAMES - you can already use these characters in text VALUES in XML 1.0.

XML 1.1 also helps people who use IBM mainframes, or who interoperate with them.

caveat emptor

[4of12]

standard to end all standards

I love standards. And XML, too.

But isn't there enough room in hidden DTD's and XSL to drive a semi tractor trailer full of NonStandards?

Re:caveat emptor

[The Bungi]

Well... the nice thing about standards is that there's so many of them to choose from =)

Don't stop on XML, go with RDF

[axxackall]

While in many cases, like various config files, XML looks very promising, we should not ignore the fact that "plain" XML is not a solution. "Plain" XML is one that can be definined through DTD, but any more comprehensive schema cannot define it.

I always use at least XML-Schema. But also, I whenever I see that semantic of files has to be defined as well - I go with RDF.

The main reason is the same as with XML instead of old-style config files: you don't want data logic to be hardcoded into your application unless it's in application requirements.

One of the best examples is GUI configuration in Mozilla. The framework "knows" how to display GUI elements, while RDF "knows" what to display. If you would try to do it with XML without RDF, very soon your framework would "know" too much of WHAT to display - but that is was not in the framework requirements, therfore it must "outsourced" to RDF.

XML 1.1 incompatibility

[kherr]

Elliotte Rusty Harold has a persuasive argument [cafeconleche.org] against XML 1.1. He is someone who's opinion should be considered. He writes very thorough, good books on XML and has created the most excellent XOM [cafeconleche.org] (same goal as DOM, but easy to use). He also keeps us current on the XML world at Cafe con Leche [cafeconleche.org].

Re:XML 1.1 incompatibility

[fm6]

I have a lot of respect for ERH. As you say,he's written some really good books, including the absolute best XML-for-beginners book [amazon.com]. And I sort of agree with him that a 1.1 spec is premature. But I think he's kind of gone overboard on this issue. Some points:

• Although this revision was partially driven by IBM (they needed some changes to do XML in EBCDIC [terena.nl], there are other problems that people wanted to address. ERH seems to think this is all a case of IBM throwing its weight around -- which is just not true.

• ERH blames this hassle over an "obsolete" coding on IBM's corporate arrogance. In this, he follows the common wisdom that IBM just ignored the whole EBCDIC/ASCII issue until the PC revolution made a switch to standard character sets unavoidable. I actually believed this stuff myself, until I stumbled onto the real story [bobbemer.com]. Which is that ASCII (which was invented at IBM!) was supposed to be the standard for IBM mainframes, but a series of bureaucratic screwups and bad decisions prevented this from happening.

  Yeah, IBM, should have made the change anyway. But we all how hard it is to change a technical direction once a lot of effort has been put into it. Look at the groddy UI at Slashdot. Look at all the web presence providers (including mine!) who won't upgrade to a taint-safe version of Perl because it would break too much of their code. The examples are endless. You may not sympathize, but you damn well should understand.

• If you read the 1.1 spec carefully (I recommend the version with highlighted revisions [w3.org]) you'll realize that 1.1 is in no way a replacement for 1.0. People with a lot of 1.0-compliant XML and who don't need to read any 1.1-compliant XML can simply ignore the new spec, if they choose. The 1.1 spec very clearly states that nothing has changed in terms of what parsers should and should not do with XML 1.0. Provided, of course, people remembered to head all their XML files with the mandatory processing instruction:

  <?xml version="1.0"?>

  And if you've read ERH's books, you know to do that!

We don't need no stinkin' 1.1

[The Bungi]

I don't see why I need XML 1.1. I've been doing markup in Mongolian for ages:

<?xml version="1.1" encoding="UTF-8"?>
<&#4632;&#4925;&#4624;&#4939;>
<&#4768;&#4653;&#4773;&#4661;&#4725;>&#4840;&#46 35;&#4724;&#4814;&#4656; &#4808;&#4757;&#4876;&#4621;</&#4768;&#4653;&#4773 ;&#4661;&#4725;>
<&#4637;&#4821;&#4651;&#4939;&nb sp; &#4812;&#4901;&#4653;="&#4969;">
<&#4768;&#4653;& #4773;&#4661;&#4725;>&#4840;&#4770;&#4840;&#4657;& #4656; &#4840;&#4725;&#4813; &#4621;&#4853; &#4624;&#4648;&#4877;</&#4768;&#4653;&#4773;&#4661 ;&#4725;>
<&#4708;&#4725; &#4812;&#4901;&#4653;="&#4969;">
&nbsp ; &#4840;&#4851;&#4810;&#4725; &#4621;&#4869;&#4963; &#4840;&#4768;&#4776;&#4653;&#4611;&#4637; &#4621;&#4869; &#4840;&#4609; &#4752;&#4813; &#4840;&#4770;&#4840;&#4657;&#4656 ; &#4776;&#4653;&#4656;&#4726;&#4656 ; &#4840;&#4725;&#4813; &#4621;&#4853; &#4624;&#4648;&#4877; &#4840;&#4634;&#4776;&#4720;&#4616;&#4813;&nbs p; &#4752;&#4813;&#4964;
</&#4708;&#4725;>
<&#4708; &#4725; &#4812;&#4901;&#4653;="&#4970;">
&#4768;&#4776;&# 4653;&#4611;&#4637; &#4845;&#4656;&#4624;&#4677;&#4752 ; &#4808;&#4616;&#4848;&#4964;

&#4845;&#4656;&#46 24;&#4677; &#4843;&#4821;&#4678;&#4709;&#4752 ; &#4808;&#4616;&#4848;&#4964;

&#4843;&#4821;&#46 78;&#4709; &#4845;&#4609;&#4851;&#4752;&#4755 ; &#4808;&#4752;&#4853;&#4638;&#4729;&#4752; &#4808;&#4616;&#4848;&#4964;

</&#4708;&#4725;>
</&#4637;&#4821;&#4651;&#4939;>
</&#4632;&#4925; &#4624;&#4939;>

Sheesh, really.

Well, the only thing I haven't been able to figure out is why they did that thing with the NEL or whatever weird carriage return marker.

What I would like

[Maigus]

Is a book which actually does a *good* job of describing the architecture and construction of proper DTDs and the syntax thereof. Anybody got a suggestion? The MSPress Books I've got don't have much other than a few scant examples ("you can produce more complicated DTDs bu that is beyond the scope of this book...").

Suggestions...?

Re:What I would like

[LauraW]

Here's one: XML Schema [amazon.com]. :-)

If it's at all possible to use schemas instead of DTDs in an application, I'd recommend doing so. Schemas are much more powerful and expressive. DTDs are an abomination.

Actual use.

[bcore]

It's interesting to look at the W3C's description of XML: Its goal is to enable generic SGML to be served, received, and processed on the Web in the way that is now possible with HTML. I wonder what fraction of the XML usage out there actually falls under that definition... God knows I've used it for config files, messaging, etc. far more than I've ever used it for processing stuff on the web. Maybe that's just me..

If you're using OWL...

[tcopeland]

....that is, the Ontological Web Language [w3.org], you might be interested in SemWebCentral [semwebcentral.org], a new project hosting site for Semantic Web projects that just opened up.

Many of the tools from the DAML [daml.org] project have already moved over to SemWebCentral; it's definitely worth a look.