It occurs to me that a "vehicle" in "systems" like DOM, XML and EDI can be 
rationally extended or is currently "utilizable" for significant added 
utility by some ability to "graft" "ordered tuples" "onto" "elements" and 
other structure/data objects.  Further, that  an ability to utilize such 
tuples should be "included" in any tree of structure at the highest level 
possible, and so augment their being invoked at any (lower only?) level in an 
"attached" manner at not only that level, but to one particular instance at 
such lower level.  Thus, arrays like (element/whatever, alias, alias_index#, 
alias_rank_in_index#, value/content/whatever) and so on could 
allow/facilitate things like multi-lingualism, multi-version, 
multi-skill-level, elements's-access-control-list, etc. be achievable via a 
somewhat extension of core capability (even completely encapsule a systems 
structure whithin itself, and the ability to completely encapsule what can 
extensibly be done with it?).

I must assume such function is already achievable via constructions from the 
existing systems mentioned above, but seems all that could be substantailly 
trivialized by an extensibility capability built in and available for 
utilization from whithin said systems at "core" levels.  (It might lack some 
"obvious" "clarity", like "pointers" in "C" affect MANY).  Does anyone follow 
this, particularly, has anyone implemented such  functionality at maximally 
"low" levels of utilization of some such "system's" current "inner" 
capabilities?  Instead of parallel trees of structure for structure_imposed 
and corresponding content-of-structure elements, just one structure.  I don't 
think it would be efficient to start sending so much often unneeded data 
around, but "smart" implementations surely send only the data needed for that 
recipient at that moment, having "realized" the "parties'(tuples')" mutual 
overall data/traffic minimization needs and methods for achievement of same.

I'm alluding to a really! simple and perhaps quite useful and very! small 
extension here.  XML as a specific example; is this doable now, in XML/?, to 
achieve a major reduction of index overhead of parallel/related 
structure/data "content-(re/dis)assembly", let alone version control etc. 
overhead multiplying same???  Really smartly normalized dB tables formally 
and easily achieve this, but it's actually a LOT of work on a LOT of tables 
and a LOT of indexing.  Can't all that be somehow "condensed" by adding 
"pointer arrays", "arrays of pointers", etc. capabilities?

Naaaah.

Regards,

Jim Cunningham

"If you can't remember it, ... ... ..., forget it."  Firesign Theatre


------   XML/edi Group Discussion List   ------
Homepage =  http://www.XMLedi-Group.org

Unsubscribe =  send email to: [EMAIL PROTECTED]
Leave the subject and body of the message blank

Questions/requests:  [EMAIL PROTECTED]

To receive only one message per day (digest format) 
send the following message to [EMAIL PROTECTED], 
(leave the subject line blank) 

digest xmledi-group your-email-address

To join the XML/edi Group complete the form located at:
http://www.xmledi-group.org/xmledigroup/mail1.htm


Reply via email to