Talk:XML schema

From Wikipedia, the free encyclopedia
Jump to: navigation, search

I wrote this article in an attempt to help disambiguate "XML Schema" from the more general "XML schema"; there was at least one article (schema) that was using a general reference but linked to the W3C XML Schema article. I moved that article to "W3C XML Schema" for now.

Sadly, when I wrote this article, I was unaware that there is a lengthy section in the article on XML that covers XML schema languages. The concept of a schema and the language used to express that concept are two separate things, so I don't see any harm in having separate articles. However, I don't like the duplication, either, so I'll see what I can do to eliminate overlap ASAP.

Also, I'm really not sure whether it is better to do it this way (a long description of an XML schema, with links to W3C XML Schema), or if it is more Wikipedia-ish to have something like this:

  • a "schema (XML)" article
  • an "XML Schema (W3C)" article
  • an "XML schema" disambiguation page that links to the two articles
  • an "XML Schema" redirect to the disambiguation page

Thoughts? - mjb 06:48, 21 Jul 2004 (UTC)

Yes I prefer this other approach you are suggesting. I think it's too confusing having capitalisation of the letter 'S' as the only difference between the titles (XML Schema & XML schema) -- Nojer2 09:34, 8 August 2006 (UTC)
I agree that the article on the W3C XML Schema needs to have a more substantial name difference than just a capital letter. How do we go about making this change? Korval 20:57, 6 December 2006 (UTC)


The section on criticism relates to XSD in particular, not to XML schema languages in general. It should be moved to the XSD article. However, I'm reluctant to do this while the list is still in its current state. Although the bullet points are all valid criticisms, they are a "point of view" and therefore need to be presented very carefully, and with cited authority, to justify their place here.

Mhkay (talk) 21:37, 23 October 2008 (UTC)

Example of C-to-XSD API[edit]

I have again deleted the example C application which someone has added (and re-added after my first deletion).

This article covers multiple schema languages, so why you think it's relevant to include an example of using one host language API to validate a document using one particular schema language, I really can't imagine. It might *just* be appropriate for the XSD article (where you have also added it), but it's questionable even there, since the article is information about the schema language and not a user manual on how to run particular products.

Mhkay (talk) 18:09, 27 May 2009 (UTC)


The "metamodel" section added by user Ottomachin flagrantly violates WP:NOR and WP:NPOV. It is extremely opinionated and unsourced. The single source he has added appears unreliable.

The leader of the original XML team admits that they did not begin with a data model. "In the interests of time, XML 1.0 did not define its own data model" [1]

The only sourced statement, which does not seem to be relevant to anything except the opinions of the editor, and has a very weak looking source as well.

They developed a specification for data structures without themselves defining their own data model. A number of people "co-operated" by email over a short time period to create the original spec.

Neutral enough, except for the quotation marks around cooperated, but unsourced.

The result is therefore unsurprisingly a very good example of "design-by-committee" but that it is also a master piece of incomprehensibility is not quite so obvious a result, though true.

"Masterpiece of incomprehensibility"? Says who? Your personal opinions do not belong here.

The following diagram

The image includes extremely biased statements, including foul language.

is an (the only existent) Entity-Relationship (though in UML format) metamodel of the information elements of XSD. The so-called "conceptual model" (in the specification) of "components" is incoherent therefore this model addresses only the actual information elements themselves.

"Incoherent"? Says who? It is complex, but not incoherent.

It is perhaps worth mentioning that the ER metamodel (a well known and accepted system of knowledge represention with a theoretical and philosophic background extending back to at least Socrates) reqires only approx. 10 entities whereas this monstrosity of modern day neo-something-or-other obfuscation contains more than 50.

Needs sources, except that the statement is irrelevant and biased.

Sahuagin (talk) 00:12, 31 July 2011 (UTC)

Deletion of criticisms section[edit]

Well, I spent my time 'refactoring' the criticisms, only to realize that not only were they originally heavily biased, but they are also not addressing the topic. The criticisms refer to the WC3 XML Schema, whereas this page is for XML schema in general. The WC3 XML Schema page already has a sufficient criticisms section. Sahuagin (talk) 06:21, 29 April 2012 (UTC)

Merge prop[edit]

I don't know why someone decided to write a separate article on a so-called comparison, when xe could have expanded this one. Tijfo098 (talk) 04:06, 23 November 2012 (UTC)

From my prospective, XML schema is not a big deal to have a separate article. Definition could be found in XML article and XML schema languages (talk) 17:11, 25 February 2013 (UTC) MS

Let's vote (to merge or not XML schema with XML schema languages)? --Krauss (talk) 14:05, 25 September 2014 (UTC)