Must XML be SGML compatible? (Was: An incompatible CD

[dlandeck at thomtech.com]

len bullard wrote:

>I suggest also, if we play this long enough, there 
>won't be significant differences between SGML and XML 
>*in common practice*.  That's a goal that's easy to agree on but 
>takes many hands at the table to achieve.

-------
I agree Len.  And, in effect, what XML (as an "application profile" of 
SGML) sets out to do is standardize a best practice for the (essentially) 
SGML encoding of data for the purposes of Internet transmission and 
interprocess communication.

I think the SGML standard, at its core, is very sensitive to the needs of 
tool developers.  Earlier kvetching about marked section syntax was a bit 
off point IMHO.  Deep down in the parse tables and state transition rules 
of a parser, the same code and data constructs support the parsing of 
marked sections as any other token stream. 

Where SGML fails as a dynamic encoding scheme for purposes such as the 
Internet is in its binary notion of "valid", its rigor in demanding full 
declarations, its inability to discuss a fragment of an instance, and 
MINIMIZE, which can too easily allow communications-protocol corrupted data 
stream to masquerade as complete. (Linking and external entity treatments 
don't seem to be as much standards issues as best practice issues.)

Standard SGML forces tools to be expensive and cumbersome not by the core 
features of required for content tagging (with which I include marked 
sections) but by the "bells and whistles" that the standard acquired in 
order to be attractive to the smaller machines and manual methods of the 
mid-eighties.  By methodically dissallowing these XML will allow simpler 
tools to be built, yet adhere to the solid, standardized, base of SGML.

As far as parse speed, I don't think that the differences in SGML and XML 
will account for an economically meaningful discrepancy between the two 
formats, especially because Moore's law is expected to hold for another 
decade and processing power will be the least of our worries.

----------------------------------------------------------------
David Landeck                              Voice: (301) 548-4039
Practice Manager                           Fax:   (301) 527-4009
Advanced Technologies Consulting Group     dlandeck at thomtech.com 
Thomson Technology Services Group
1375 Piccard Drive, #100
Rockville, MD 20850
----------------------------------------------------------------
     




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/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe 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)