Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Development Trends in Web and Oracle

Re: Development Trends in Web and Oracle

From: Noons <wizofoz2k_at_yahoo.com.au>
Date: Mon, 14 Mar 2005 23:07:20 +1100
Message-ID: <42357e79$0$21025$5a62ac22@per-qv1-newsreader-01.iinet.net.au>


DA Morgan apparently said,on my timestamp of 14/03/2005 10:40 AM:

>
>
> It isn't a trade off ... it is bad.
>
> It would only be a trade off it there weren't a far more efficient
> solution available.

All this XML stuff looks to me as a solution looking for a problem that was defined 35 years ago and solved many times since then...

But let's be realistic for a second.

There is undoubtedly merit in an architecture/design/solution (pick your preferred term) that includes metadata with the data.

If we look in detail for example at EDI and other similar technologies, there is an underlying protocol that is signaled by specific bit settings and packet identifiers(in XML parlance: the tags) and an assumed understanding that certain settings mean some types of implicit processing (in XML parlance: the schema).

What XML brings in is a human-readable (text based) architecture, which PRESUMABLY will avoid proprietary interpretations. Given an adequately prepared program, the technology handles transfer of data in a very neat fashion. And maintaining that program becomes unnecessary: just maintain the schema.

As such, the long standing problem of interfacing different EDI or communications protocols disappears. Everyone wins.

PRESUMABLY. Of course, proprietary XML schemas are just around the corner, and so are encryption mechanisms made necessary because NOT EVERY PIECE of data out there is for public consumption.

And of course: the interpretation of a schema is purely arbitrary and there are no pre-defined processing mechanisms anywhere. Which means custom-coding all the way.

And yes: all you need is a committee to decide what a DTD should contain for every industry situation. Pie in the sky if I ever heard one. And we all know what committees do, don't we? :D

All this rigmarole costs a fair chunk of CPU and memory to handle, orders of magnitude more than what was used before with EDI.

The problem is that like you said, the technology is a nail. And someone is trying to convince me that dovetails can be replaced by nails.

I feel a tug in my pants...

-- 
Cheers
Nuno Souto
in sunny Sydney, Australia
wizofoz2k_at_yahoo.com.au.nospam
Received on Mon Mar 14 2005 - 06:07:20 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US