home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ralvm13.VNET.IBM.COM
- From: drmacro@ralvm13.VNET.IBM.COM
- Message-ID: <19930107.113434.441@almaden.ibm.com>
- Date: Thu, 7 Jan 93 14:33:13 EST
- Newsgroups: comp.text.sgml
- Subject: Re: SGML for data querying
- Disclaimer: This posting represents the poster's views, not those of IBM
- News-Software: UReply 3.0
- References: <1993Jan7.021616.11906@nuscc.nus.sg> <1993Jan7.142852.23671@infodev.cam.ac.uk> <19930107.090654.710@almaden.ibm.com>
- <1993Jan7.180136.236@infodev.cam.ac.uk>
- Lines: 44
-
- In <1993Jan7.180136.236@infodev.cam.ac.uk> Alasdair Grant writes:
- >
- >Thanks for your reply. I just want to pick up on something - what do
- >you think about including features like comments, tag minimization,
- >and data tagging in a database? If you don't, you aren't really storing
- >the "real SGML", but I honestly can't see the point of remembering that
- >tag of item X was minimized while tag of item Y wasn't. Seems to me
- >that there is at least one level of SGML which gets in the way of
- >database applications, and ought to be stripped out in any importation
- >of SGML into a database.
- >
-
- It's certainly technically possible to preserve all of that in
- a database if the needs of your system require it, but I wouldn't
- loose sleep over a system that didn't preserve comments. Tag
- minimization can be re-applied algorithmically since the minimization
- rules are defined in the DTD, and features like shorttag and
- datatag are best left alone, since they are primarily concesions
- to the problems of data entry, not information identification or
- structuring, and therefore shouldn't enter into the equation.
-
- Most of the folks I've talked to do not seem to consider the requirement
- that SGML source be normalized for database storage to be a problem
- in practice, and we know we can unnormalize it if we ever want to.
- You can get around the problem of preserving comments by providing
- elements whose defined semantic is to contain descriptive comments
- about the elements that contain them or define explicitly-descriptive
- elements for the sorts of things you normally use comments for,
- such as "file headers" that contain author information, change
- tracking info, copyrights, etc. That is the approach we've taken
- within IBM. I will not accept any arguments for preserving comments
- based on the needs of applications that use the data in comments
- for their own purposes. SGML provides better places for such
- information, such as processing instructions, the APPINFO statement
- in SGML declarations, extra-SGML mechanisms unique to the
- application, and so on. Since comments are, by definition, not
- part of the document content, and are not carried by the ESIS
- information, there is no strict requirement to preserve them.
-
- Eliot Kimber Internet: drmacro@ralvm13.vnet.ibm.com
- Dept E14/B500 IBMMAIL: USIB2DK9@IBMMAIL
- Network Programs Information Development Phone: 1-919-543-7091
- IBM Corporation
- Research Triangle Park, NC 27709
-