[
Lists Home |
Date Index |
Thread Index
]
Hi Ken-
Writing you and Bill (I don't think he'll be exceedingly offended).
Bill's comments are close enough,
but not quite the truth, the whole truth, and nothing but the truth.
Bill (Klein) is both knowledgeable and a former member of committee.
And generally accurate, but he does write with an edge.
I tried to present what exists with as little formal detail as possible,
and probably should have provided more detail; Bill is better at detail than
I.
First, your comment - no the standard itself will not have difficulty moving
to become a fast track; it is beyond that. That is simply because 2002
standard was one of the first to try to experiment with the fast track kind
of thing and COBOL2002 is now in fact an ISO standard. Once something
receives status of ISO standard, it is out of the 'fast track', and on to
the standard review and update track. The TRs and the folding of the TRs
are part of that process.
An actual fast track currently in progress would be a language like C#,
which came in via ECMA as an established ECMA standard and had some years
(I can check, but I think it was 2 years) to run through the Fast Track
process.
Now Bill's things:
There are 3 items on the agenda relating to TRs for the upcoming (Oct) ISO
work group meeting for COBOL. One of those is for XML TR. So yes, Bill is
correct in his detail that there is no direction to include it; the reason
it is on the agenda is that investigating XML handling and how to do it was
requested at a prior meeting. And there is also no direction to exclude it.
The meeting has not yet happened.
At the meeting, the various countries will hash things out. That's the way
these things work in this arena. Unanimous is desired, but not required.
It is ironic that IBM is singled out as treating XML in their compiler
somewhat differently than how the TR is fleshing out -- the IBM folks are a
strong part of the effort forming the TR. Perhaps one learns what works and
what doesn't work by actually trying something.
As far a guarantees - again Bill is correct, no guarantees. I'm not sure any
work would get done anywhere if everyone waited for guaranteed success. We
do our best, but no guarantees.
Finally, this one probably does have some importance, the standard is just
like any other language standard - a set of rules for syntax and semantics.
Implementers of a standard may not implement everything, or they may add to
their implementation items not in the standard. Try the C++ compilers which
come with Sun, Windows, IBM, and perhaps HP boxes. Close, but not
identical; particularly around the STL. Or try the JVM across different
operating systems.
The COBOL2002 standard has some items, for instance RESUME, which I am
pretty sure have not been implemented by anyone. Maybe it is a good item,
maybe a mistake. The vendors themselves will respond to the market. The
only thing for certain is if that item is implemented by a vendor, it will
have the rules for implementation defined.
Okay, enough.
Basically, Bill is correct, but not exactly telling the entire story.
And the story will unfold more in October.
Thanks,
Barry
Barry Tauber ( btauber@interaccess.com ) 01(847) 267-8012
International Representative INCITS-J4 (COBOL)
Certified MCP (Microsoft), Java (Sun)
-----Original Message-----
From: Ken North [mailto:kennorth@sbcglobal.net]
Sent: Tuesday, September 13, 2005 4:41 PM
To: xml-dev@lists.xml.org
Subject: Standards controversy? (Updating programming languages to support
XML)
It appears adding XML support to the COBOL standard may be more
controversial
that it was for the SQL 2003 standard.
Barry Tauber recently wrote "XML and the New COBOL"
http://www.webservicessummit.com/Trends/COBOL_XML.htm
The article has raised questions:
1. Whether the COBOL standard will move through INCITS and ANSI to become an
ISO
Fast Track candidate.
2. Whether it will win ISO approval. According to William M. Klein, Germany
has
questioned whether XML support belongs in the COBOL standard.
|
