← Prev in month ← Prev in thread
Next in thread → Next in month →

Essentials (was RE: [xml-dev] heritage (was Re: [xml-dev] SGML on the Web))

From
Mike Champion <>
To
XML DEV <>
Date
2002-10-07T16:08:36Z
ID
<514VP87YA5JEQK659KGFGAEXUUT.3da1b184@MChamp>
Thread
Essentials (was RE: [xml-dev] heritage (was Re: [xml-dev] SGML on the Web))
10/7/2002 10:33:26 AM, "Dare Obasanjo" <> wrote:

><troll>
>XML is a data model, the syntax in the XML 1.0 REC is just an annoying inconvenience. 
></troll>

Troll or not, for good or for bad, and despite the fact that XML 1.0 says virtually
nothing about a "data model", what Dare said is the pretty much the assumption 
upon which SOAP 1.2, WSDL 1.2,  WXS, XPath, and  XSLT (I think) 
are formally based.    The precise form of the "XML data
model" is controversial, but all of the above start with the essential form of a
labeled tree with attributes associated with each node;  we could debate forever
whether PIs, CDATA sections, whitespace between elements, etc. are in the data model
or not, because none are authoritative.

So, is "XML" essentially a markup language, and thing like LMNL just minor variants
even though they don't parse into labeled trees, or is "XML" essentially a data 
model with the angle brackets simply the canonical syntax?  I don't think it's
a fruitful argument.  On the other hand, I think it's important to anchor a particular
application or effort on one or the other, or one gets into conceptual morasses.

It will be interesting to see a few years down the road which of these yields 
more valuable fruit.
← Prev in month ← Prev in thread
Next in thread → Next in month →