I actually didn't mean to send a couple responses to the entire list
that reflected individual requests for assistance, but now that you bring
this up I feel compelled to respond.
Unfortunately, OFX does not take care of this. I think the
historical purpose of OFX was to unify a couple of binary wire transport
protocols that Quicken and Money used, to simplify the life of banks talking
to those two personal finance applications. As such, OFX only had to be
better than what it was replacing, and only had to deal with that universe
of discourse, not the wider world of finance. While OFX defines itself with
the help of DTDs, the text of the spec disallows things like comments and
whitespace around elements, so even the claim that OFX is SGML, or SGML
compatible is somewhat suspect. (I have asked third parties if they have
ever run OFX through SP. No-one has ever said yes.) It is certainly not XML.
This does not stop people from saying the contrary.
All that can be changed. What cannot be changed is the fact that OFX
is a bad design even for what it claims to do. As an example, it suffers
from "tag inflation" because it encodes information in the tag names, such
as <SIGNONMSGSRQV1>, which should be something like <Request use-case="sign
on" version="1">. The current work Microsoft is fostering under the DNAfs
rubric completely sidesteps OFX, though you can write a gateway from OFX to
the newer model. The death knell for OFX, IMHO, came when Microsoft
abandoned MIFST (Marble) at the beginning of December.
I presented these personal opinions to Microsoft last year, when
KPMG first became involved in DNAfs. They are my opinions, as is the opinion
that the current DNAfs schema isn't perfect, but is a lot better than OFX.
David vun Kannon
> -----Original Message-----
> From: B.K. DeLong [SMTP:[log in to unmask]]
> Sent: Monday, December 14, 1998 2:55 PM
> To: [log in to unmask]
> Subject: Re: Uses of XML
> At 10:12 AM 12/14/98 -0500, you wrote:
> >We are working on Microsoft's DNAfs initiative, a framework for
> >interoperability among financial services software vendors. The
> >of all the method calls are specified in XML.
> Doesn't OFX take care of that? Why create two standards??? Doesn't that
> defeat the purpose?
> B.K. DeLong 360 Huntington Ave.
> Director Suite 140CSC-305
> New England Chapter Boston, MA 02115
> World Organization (617) 247-3753
> of Webmasters
> [log in to unmask]