History

[Len Bullard]

Betty L. Harvey wrote:
> 
> Personally, I give credit to Yuri Rubinsky, President of SoftQuad until
> his untimely death, to the birth of XML. 

Agreed.  Yuri was influential in most SGML decisions at that time.  He
worked tirelessly.  In this, Charles was the hard headed standards
editor, 
and Yuri was the diplomat.  I miss him.

> I give credit to Tim Berners-Lee
> for the insight to recognize that SGML was format for the Web - HTML.  He
> could have chosen NROFF, RUNOFF - any of those other system formatting
> languages but he chose SGML - Thank you very much.

Yes.  Good call.

> Before I worked for the CALS program I was
> told that working in CALS was a dead-end to your career.

It paid for educations and it was in the initiative where many of us 
learned the ways of making standards and how to live in motels.   :-)

> When the Web came into existance I was working in CALS within the Navy and
> with our local Technology Transfer czar at David Taylor Model Basin.  I
> was positive that SGML could not work at a reasonable cost and effort
> within a large organization, such as DoD. T

This is where we disagree.  As a contractor to David Taylor, USAMICOM, 
and a lead designer of the CASS system, from 1986 to 1996, I worked 
around or on many of the Tri-Service IETM projects.  With the exception 
of the B-series bombers which were *near-black* security, there are 
few of these systems I did not study directly.  They all worked, Betty.  

I was there.  For every bloody inch of it.   The HTML you showed 
me was fantastic because it handled multi-media and because 
it WAS internet enabled.  Few IETM systems were although IADS ran 
fine on Novell.  That was precisely the struggle being in the 
IETM community.  It was meant to be for non-public systems so 
all of the work we did on projects like Beyond The Book metaphor 
were studied but not accepted.  (OTOH, that one is coming back 
if GEAE has their way.)

The problem was we could not standardize on one system and, we were 
being required to in a Tri-Service NIH frenzy of funding and control. 
Ultimately, we ran out of time... and interest.
   
Expensive?  Yes.   But the cost coming down was not XML or HTML:  it 
was 50 cent memory and two dollar processors.   We already had
stylesheet 
based browsers, and well-formed rules for handling instances.  We had 
server-side examples such as the General Dynamics and Lockheed Ft Worth 
systems.  It was working.  It was not esoteric.  It did require one to 
be an able document systems pro to set up and run. 

Sell them?  Heck no.  Everyone had Interleaf.  Then we got hit with 
the first wave of ATOS output to transfer to the IETMs.  Worked
perfectly.

The idea of using the Internet for IETMs died in the 
discussions of GOSIP.  That was TimBLs contribution:  he got a markup 
based system and a web protocol for hyperlinking out first.  There were 
no innovations in that;  but the timing was marvelous.

> I tried to talk the Navy into modifying
> Mosaic code (it was public domain) to support IETMs.  The climate
> wasn't right in 1994 for heresy. Today most IETM projects are
> Web based.

Right.  I can support that enthusiastically now, particularly now 
that with DHTML/XHTML as the client standard, and XML for the kinds of 
things we needed for 87269, it works. Heck, now we even have the 
X family of standards for stylesheets, linking, and so on.  Big 
fun and mortals get to do now what once only the DoDsters 
had and then, only if they were on the right projects.  Progress.

But Mosaic did not meet the requirements.  It was only 
when a scripting language became available internally and 
server side HTML became widely available that the web browsers 
met the requirements for IETMs particularly, 87279 DB-like.  
US MID came closer than both because it was defined precisely 
to meet those requirements.  The fact is, HTML became the 
only viable economic solution.  That is what Megginson asserts 
for XMLs triumph and I agree.  I just have the further heretical 
opinion that without Microsoft there at the very beginning, the 
odds that XML would succeed were little to none. 
It doesn't matter.  I think it is a viewpoint that is correct 
and casts a realistic light on some of the XML hype.

> It does distress me that now that XML has become popular that the pioneers
> are forgotten, especially Yuri.  He was a man who impacted not only the
> web but people positively.  He was a real visionary.  I also believe he
> would be really pleased about the current XML efforts.

Yes.  But he would be the first I think, to quickly acknowledge the 
heritage that XML has, that while some can call themselves inventors 
and co-inventors, it is reinvention.  Better to be known as the leaders 
who took what was and made it better.  That is what XML is.  The 
reasons for that don't matter.  The History does.  Respect is a duty.
 
> XML is becoming prolific because it is a good technology and people
> (computer scientists and marketeers) can understand it.  

Yes.  That is an achievement and it took leaders to make that happen.

> In conclusion - thanks Charles, Yuri, TimBL and many others for their
> vision!

Yes.

len


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 unsubscribe, mailto:majordomo at ic.ac.uk the following message;
unsubscribe 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)