From: Robin Berjon [mailto:robin.berjon@e...] 

On Jun 06, 2006, at 20:25, Bullard, Claude L ((Len)) wrote:
> Except microformats are a bit of Hytime:  the arch forms.

>As one of those brats who isn't aware computing existed before XML did,
I am bound to ask: "What isn't?" :)

Nothing of importance.  Well... Not quite true.  The real-time aspects
of Hytime didn't get much play although that was originally why it
attracted my attention:  I was looking for a timing model for real time
systems (think, the equivalent of a musical sequencer) for
project/product planning.  Remember that Hytime starts out in a musical
notation standard.  We did end up with that but from the
orchestration/choreography models.  In the 80s, that was considered
radical.  Now it's just BizTalk.

> Interesting: given all of the various projects over the years, if one 
> started with SGML again looking at the various projects, what would 
> the subset be today?  Or would it be a superset now (SGML almost had a

> binary but the roof blew in just as that was happening)?

>Given how much heat there's been around both subsetting XML and an
equivalent efficient syntax for it, I wonder how far 
>from consensus we'd be now, ten years down the lane. Enough of XML to
capture an infoset comprising Document, Element, 
>Attribute, Characters, Comments, and perhaps PIs and entity references,
but no more? Binary as a separate syntax with
> pretty much the same stuff? Or with a little extra super simple
typing?

That is what I wonder too.  But there is only a short freefall before
things have to be useful or they become complex again.  Strange:
complexity really is advanced entropy.  Hmmm...

At some point, 'a byte must change state' as Charles would say.  So once
there was enough agreement on XML, the coding began and then the
investments have to be recouped before another round of speculative
standardization can get going.  It is like Google and much else: the
thinking occurs in the excess capacity of the system.

>I never quite found out what had happened to SGML-B, the available
online docs indicate an intention, and then some form >of ELE seems to
have happened.

I think it died with the web like much advanced research does:  no
funding.

len