Archive of UserLand's first discussion group, started October 5, 1998.

Re: T. Nelson Critique of Embedded Markup

Author:Paul Snively
Posted:11/9/1999; 7:52:08 AM
Topic:T. Nelson Critique of Embedded Markup
Msg #:12883 (In response to 12880)
Prev/Next:12882 / 12884

Erik Neu wrote:

One quote from the Introduction which might be of especial interest to this audience:

>>What if a system could exist which would satisfy all parties--copyright holders and those who would like to quote and republish? What if materials could be quoted without restriction, or size limit, by anyone, without red tape or negotiation--but all publishers would continue to furnish the downloaded copies, and would be exactly rewarded, being paid for each copy?>>

For what it's worth, and at the risk of oversimplifying huge swaths of Xanadu's history, this is neither more nor less than the goal of Xanadu as a project over the last 30 years or so. Now that there are two partial implementations of Xanadu available as open source, it should prove interesting to see just how far the Xanadu project got, and--perhaps most importantly--the extent to which Nelson remains correct about the need to supercede existing languages and protocols in order to accomplish the above goals.

I can't say I fully understand where he is coming from in his first two points of objection. He offers the usual indictments of HTML, but seems to think SGML no better.

Well, it isn't: SGML still makes a large-ish number of ontological committments that I think Nelson identifies well (e.g. hierarchically-decomposed linearity of structure) that make SGML a sub-optimal language in which to represent what Nelson wants to represent.

The shortcomings he mentions for SGML/HTML (for example, re-formatting when re-purposing [or, in his words, transcluding]), seem to me exactly the kinds of things XML/SGML, along with XSL (or CSS for that matter), and possibly XLL thrown in as well, are intended to address.

"Transclusion" is a strictly stronger concept than "re-purposing," which is why there's a new word for it: all transclusion is done by linking, period. "Re-purposing" frequently takes place by copying of content, which is essentially anathema (to appropriate Nelson's appropriation of religious terminology!) to Nelson's worldview for the perfectly good reason that it makes it effectively impossible to maintain the royalty structure that he envisions.

At its core, Nelson's argument isn't even so much about HTML/SGML; it's more about the storage facilities we use and our access protocols for them. It isn't that CSS, XSL, and/or XLL aren't good tools for providing the presentation of content; it's that they take content in and spit (new, modified) content out the other end. Provide a guarantee that CSS or XSL could only be applied by following a link to the original content and I'll bet Nelson would be pleased.

His third point of objection seems far, far deeper: >>We greatly need a general structure to represent all forms of interconnection and structure, and changes in both content and structure; and to visualize and re-use variants and alternatives, comparing them in context in order to understand and choose.>> While it may be of enormous theoretical interest to certain deep thinkers, I imagine it is miles off the radar for most of us.

I've heard this many times in relation to Nelson's long-standing assertions in this regard, and especially these days, it strikes me funny: Nelson has, since the 1960's, been engaged in attempt to construct an online publishing system where anyone can be a publisher, anyone can quote from any other work, anyone can be quoted in any other work, and the first-approximation guarantee is that anyone whose work gets quoted received royalties every time their quoted work is viewed, whether from the original source or a transcluded source. This is why the notion of transclusion is so important. If the quote of your writing above were transcluded instead of copied, and you received a micropayment every time it were viewed, then every time someone read this writing of mine that transcluded your writing, you would receive a micropayment (as would I for my content).

The thorny issues that Nelson and his group had been tackling are precisely the ones he lists: if all referring to other content is by transclusion, what tools should we/could we construct to support, e.g. dissent? Collaboration? Creation or examination of revisions? Compare-and-contrast? Far from being off the radar, these questions are indeed most pertinent in a forum that, in recent history, has wondered aloud how to protect the copyright rights of web-based content creators/publishers and how to protect the reputation ("brand" is the word that typically gets abused to discuss this concern) of content creators/publishers from middlemen in the chain of communication, e.g. Third Voice. Nelson's assertion since the 1960's has been that a new kind of storage medium, complete with new indexing and retrieval algorithms, lies at the heart of answering these difficult questions. He's held these views consistently for at least thirty years and, now that we have access to code, we can see that he and his colleagues had not merely been engaging in idle philosophical talk. I find this deeply ironic, given that Nelson has been accused of having severe ADD.

I also don't understand why he focuses on SGML instead of XML (though from a theoretical viewpoint it shouldn't matter, since he is not getting into more practical objections, like "SGML is just too complex"). Maybe the piece was written before XML appeared on the horizon. (I looked later and noticed a date: Oct 2 1997; several months before the XML spec was approved, so maybe he hadn't heard yet).

In the context of Nelson's theses, there's no appreciable difference between SGML and XML. You also effectively answered your own question: the only appreciable difference seems to be pragmatic, and even that one ("SGML is just too complex") is mitigated to a considerable extent by the availability of reuable software tools to manage that complexity.

Thanks for the pointer to more great stuff from Nelson!

There are responses to this message:

This page was archived on 6/13/2001; 4:53:26 PM.

© Copyright 1998-2001 UserLand Software, Inc.