Lists Home |
Date Index |
> -----Original Message-----
> From: Rick Jelliffe [mailto:firstname.lastname@example.org]
> Subject: Re: [xml-dev] terra incognita
> XML is not a data serialization language. It is a markup
> language, where text (i.e. words and sentences) can be annotated.
I have heard this type of statement (that XML is primarily about
annotating text) repeated many times. XML _did_ have data exchange as
one of it's priorities, back when 1.0 was released. Take for example,
the below quote from the press statement about the release of XML 1.0
"XML is primarily intended to meet the requirements of large-scale Web
content providers for industry-specific markup, vendor-neutral data
exchange, media-independent publishing, one-on-one marketing, workflow
management in collaborative authoring environments, and the processing
of Web documents by intelligent clients."
A number of people were vocal about the use of XML for data exchange,
from the get go, and by that I mean long before XML 1.0 came out.
Yes, XML was defined as a markup language, meaning it is defined over a
textual encoding. That provides the base framework, but it does not
rule out its applicability for use as a data encoding any more than it
rules out comma delimited text as a useful format for tabular data. I
have seen excellent uses of XML, where there was hardly any non-markup
content at all. Xml is about having a standard, well-defined base
protocol, on which you can layer. Just as TCP means I don't have to
worry about the details of ensuring packet order and successful
delivery, XML allows me to not worry about the byte format of my data.
Using XML, a new data format can be prototyped in minutes, and any bugs
are in your application logic. No more worries about writing lexers and
This view of doesn't care what the purpose of the markup is, whether the
markup is just 'annotations' to the text, or the text is just
annotations to the markup. To select one view and define that as the
correct view, just hurts XML in the long run.