Re: The Three Myths of XML

[Bruce ONeel <beoneel@mindspring.com>]

(not really directed at either Dave or Len)

Hi,
	
I do think that XML is not quite magic, but is important.

In a previous life I worked with astrophysicists.  The normal way
of doing the data processing for a telescope was to write your
own custom programs in what ever language (well, FORTRAN)
on what ever system you had to run on and writing your
data files as compactly as possible in binary.  

That meant that when I wanted to read your dataI either
had to have the hardware, OS, and compiler you had or
I had a porting job ahead of me.

To solve this a standard was produced called FITS
(http://fits.gsfc.nasa.gov) and while it didn't have the same
hype as XML (no money :-) it did polarize the community a bit.

Why?  Well, FITs files were bigger, and at the beginning the 
libraries which read them made your programs bigger and slower
to run.  Some even claimed factors of 10 slowdowns.

The advantages?  The format worked on different systems and
with different hardware, it was paritally human readable, and
it was possible to write and then use general tools rather than having
to write a special tool for your specific data layout.

In the end FITS became more and more popular as systems sped
up, diskspace and memory got cheaper, libraries got faster, and
more and more folks had to keep data around longer and archives
got bigger.

XML has many of the same disadvantages and many of the same 
advantages.  FITS didn't bring world peace and neither will
XML, but in both cases they were worth the time, slowdowns,
and effort.

cheers

bruce


----
Bruce O'Neel <beoneel@mindspring.com> -- XML and Squeak.  
LWxml, a light weight XML parser in Squeak
XML Journal http://homepage.iprolink.ch/~bioneel/beo/beo.html


Dave Winer <dave@userland.com> wrote:
> Len, I didn't get that it was a satire, and I thought his premise was right
> on.
> 
> Basically you'll get compatibility when that's what people want.
> 
> And (my opinion) XML is a fantastic way to do it.
> 
> In other words I agree that it's computer science, nothing amazing about it,
> the amazing thing is when people open their systems to be compatible with
> others.
> 
> Dave
> 
> ----- Original Message -----
> From: "Bullard, Claude L (Len)" <clbullar@ingr.com>
> To: "XML-Dev Mailing list" <xml-dev@lists.xml.org>
> Sent: Thursday, June 14, 2001 7:36 AM
> Subject: The Three Myths of XML
> 
> 
> > Kendall Grant Clark has published a funny satire
> > of the three myths of XML at www.xml.com this week.
> > It makes good reading.  I wonder how many still
> > subscribe to such myths and why.  It appears to
> > me that the selling of the web, the urge to get
> > the publicity and seize the domain before the
> > results are ready have as much to do with this
> > as any particular quality of XML.  On the other
> > hand, very large government pipelines of data
> > aggregation do exist that would benefit enormously
> > from using XML (cheaper and tends toward winnowing
> > out duplication).  As said in the past, the way of
> > markup is as powerful if not more powerful than
> > the actual technology but you have to do a lot of
> > this to understand the obvious.  It is the quality
> > of the information domain (enough similarity to
> > enable a coherent schema that can be shared) that
> > determines if it succeeds.  I can write a schema
> > for you cat but who will you share that with?
> >
> > XML isn't magic.  It is a computer science.  Getting
> > people to agree to use the agreements, that's magick.
> >
> > Len
> > http://www.mp3.com/LenBullard
> >
> > Ekam sat.h, Vipraah bahudhaa vadanti.
> > Daamyata. Datta. Dayadhvam.h
> >
> > ------------------------------------------------------------------
> > The xml-dev list is sponsored by XML.org, an initiative of OASIS
> > <http://www.oasis-open.org>
> >
> > The list archives are at http://lists.xml.org/archives/xml-dev/
> >
> > To unsubscribe from this elist send a message with the single word
> > "unsubscribe" in the body to: xml-dev-request@lists.xml.org
> 
> 
> ------------------------------------------------------------------
> The xml-dev list is sponsored by XML.org, an initiative of OASIS
> <http://www.oasis-open.org>
> 
> The list archives are at http://lists.xml.org/archives/xml-dev/
> 
> To unsubscribe from this elist send a message with the single word
> "unsubscribe" in the body to: xml-dev-request@lists.xml.org