What Clean Specs Achieve, WAS: Colonialism, SAX, Java, and Namespaces

[Mark Birbeck]

Rob Schoening wrote:
> The english language sentence "Grass is green" is an expression of the
> proposition that grass is green.  Similarly, I thought that: 
> 
>     <plant> Grass </plant> is <color> green </color> 
> 
> was simply an XML/english sentence expressing not only the above
> proposition, but also some explicit markup relations, namely 
> that "Grass" is
> in a plant context and "green" is in a color context.

Not sure why this should be the case. Surely it expresses whatever the
application interpreting the data says it expresses - which may or may
not be the proposition "Grass is green". This may for example be a
description of the fact that the atomic power plant, code-named 'Grass'
is no longer in emergency status, and has returned to 'green'.

All XML has done is saved you from devising your own file structure,
tagging format, or whatever. Anyone who has ever worked on streams of
data, say like MarketLine in the Stock Exchange, or wherever, will know
that you spend half your life trying to come up with clever ways of
packaging up data:

    SOH
      0xFF
    EOH
    SOM
      0xFE
        G R A S S
      0xFD
      0x20 I S 0x20
      0xFC
        G R E E N
      0xFB
    EOM

SOH is start-of-header, SOM is start-of-message, 0xFE means here comes a
'plant' and 0xFD means 'that's the end of your plant', and so on. And
then of course some smart-arse would come along and say, I need a 0xFE
byte in my 'plant' field, so you'd have to invent a way of escaping it
so that it didn't mean 'here is a plant' in certain situations. And then
someone would decide that it was a serious error if a colour field
contained a plant field, or there should only be one colour per plant in
any given message, and ... whoa! You spent most of your coding time
writing parsers! The same goes for file formats - every time you started
a new project you devised a new file format.

> So, while XML (the language) may be just a file format, XML (in the
> colloquial sense) is much more!

As I'm saying - 'just' a file format is a miracle, be grateful! Try
coming at it from another direction - if it is such a trivial thing,
then how come it took so long to get here?

> Unfortunatey it seems that that W3C is focusing on the XML 
> language, not XML
> in general.

And THAT is their major contribution. If everyone sat around saying
let's devise the standard for video, and let's devise another standard
for sound, and another for the height of basketball players, and no-one
said, "hang on, let's standardise the standards", we would have nothing
to argue about on this list!

> XML technologies have to potential to allow us to have a common
> representation of data on-disk, in-memory and 
> across-the-wire.  This is
> really powerful.  But until we get unstuck from the 
> linguistic details of
> the first, we will have nothing more than yet another file format.

No - it has the potential to have a standard *interface* between these
things. If I want to store my data in reverse-Polish Hamming-coded
object structures, using the eyelids of lizards to represent base-4
numbers (two lizards per byte), then that's up to me. I can exchange the
data with you though, by hopping out to XML and sending you my document
and a DTD. And you don't need to write a JDBC-to-lizard interface to
send the data back to me, either. You don't care what I use, you just
send me an XML document, knowing that because the language is
well-defined, I can at least parse it.

> Until the relationships between all these XML technologies 
> are laid out, I'm
> afraid that the focus on the XML language as the centerpiece 
> will skew the
> whole effort and limit its potential.

But the technologies are only just beginning to be developed - thanks to
the standardisation of standards - so how can you possibly lay these
relationships out without being prescriptive. RDF, for example, is not
'part' of XML, it is a standard that *uses* XML to provide an easily
understandable interface to the information it can represent. And new
interfaces will be developed at an increasingly rapid rate.

As they say, you ain't seen nothing yet!


Mark Birbeck
Managing Director
Intra Extra Digital Ltd.
39 Whitfield Street
London
W1P 5RE
w: http://www.iedigital.net/
t: 0171 681 4135
e: Mark.Birbeck at iedigital.net


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)