- From: Charles F. Goldfarb <Charles@SGMLsource.com>
- Date: Mon, 30 Sep 1996 13:31:49 GMT
- To: "David G. Durand" <dgd@cs.bu.edu> (David G. Durand)
- Cc: w3c-sgml-wg@w3.org
On Sat, 28 Sep 1996 21:37:35 -0400, "David G. Durand" <dgd@cs.bu.edu> (David G.
Durand) wrote:
>
>The definition of "true content" presupposes that RE after markup should
^^^^
>not be significant (the core of the confusion around the SGML RE rules).
>With that defintion in hand we can quickly come to the conclusion that
>SGML's RE/RS rules are necessary to return the true content.
No, the rule is that an RE *caused by* markup is not significant. By definition,
markup is that which is not data. Data is the true information. Therefore, an RE
that wouldn't be there if some markup weren't there isn't data and isn't true
information.
To put it another way, you should be able to remove all the markup and get
exactly the same data as when the markup is present.
--
Charles F. Goldfarb * Information Management Consulting * +1(408)867-5553
13075 Paramount Drive * Saratoga CA 95070 * USA
International Standards Editor * ISO 8879 SGML * ISO/IEC 10744 HyTime
Prentice-Hall Series Editor * CFG Series on Open Information Management
--
Received on Monday, 30 September 1996 09:33:26 UTC