Re: [xml-dev] XML basics

[Bjoern Hoehrmann <derhoermi@gmx.net>]

* Joe Fawcett wrote:
>I've been asked to contribute to a book on basic XML. I have a list of
>topics that I think should be covered, mostly discussing the basics,
>showing a few examples and when and where each would be used.

Your list is mostly an overview of technologies which can make a fine
softcover giving people an idea of what's out there or can make a fine
"bible" covering everything you might ever want to know but are afraid
to ask, but it's not exactly what I would think of for "basic XML".

I would expect a book on "basic XML" to be either very short, or to
offer a strong conceptional foundation. What is a tag as opposed to an
element, what are trees, how to you handle, read, process, generate,
access and manipulate them, what's up with the global naming stuff you
get with XML namespaces. Depth-first versus breadth-first traversal,
pre-order versus post-order traversal.

Push versus pull versus in-memory representations, imperative versus
functional versus declarative processing, data models and abstractions.
Putting it the other way around, someone with an exceptional under-
standing of XML basics will be able to talk about these things, and if
they aren't able to tell you much about SVG or WSDL or WCF or FLWOR or
RSS, that wouldn't seem much of a problem.

(In a similar way, someone who figured out how to deal with, say, Java
classpath problems trying to use some tool to put into practise what
they gathered from some book would seem far more likely to cope with
basic XML problems than someone who can name three common bugs in XML
Schema implementations.)

Another big issue would be understanding the problem domain, what are
common issues you face in data processing, formal languages, encoding
problems, security issues arising from your home grown parser failing
to perform the kind of input validation an XML process takes care of,
when XML is the best tool for a job and when you are better off using
some comma-separated value format or YAML or JSON or whatever. That is
important knowledge for "basic XML" compared to the book's author's
thoughts on the ins and outs and outlooks in "HTML5 versus XHTML".

Some of the technologies you mention are useful to discuss such issues,
showing for instance how "HTML" is difficult to process and extend be-
cause it has all sorts of complicated and inconsistent rules you need
to apply to understand the basic structure of a document, while with
XML you don't usually have that problem, and at what price that comes.

Similarily, discussing LINQ to XML can be a good vehicle to explain a
more functional approach to XML processing. But the technology is not
very important, depending on your audience you might also throw in some
Haskell.

But again this very much depends on the goals, if you want the reader
to be able to have a profound understanding of things you'd write a
different book than if you cared about them being able to talk about
things, or explore things, or look things up. With just "basic XML"
there is not much to recommend what to include, in fact, it may be
more important "how" you include them.
-- 
Björn Höhrmann · mailto:bjoern@hoehrmann.de · http://bjoern.hoehrmann.de
Am Badedeich 7 · Telefon: +49(0)160/4415681 · http://www.bjoernsworld.de
25899 Dagebüll · PGP Pub. KeyID: 0xA4357E78 · http://www.websitedev.de/