RE: RE: [xml-dev] XML is easy, was: Re: SV: [xml-dev] XML=WAP? And DOA?

["Joshua Allen" <joshuaa@microsoft.com>]

> >Ahh... but is .Net services implemented an XML subset, which did not

> First, it's Microsoft's SUPERSETS that irritate people.

Well, you have to admit that subsets irritate people a lot more than
supersets, right?

> fine ... trust us, trust us."  So sorry, but trusting y'all up there
> in Redmond to ensure interoperability has not been a winning
> proposition in the past. Sticking with the simple subset of XML

Is that really fair?  I am thinking of things like ODBC, Windows 95
built-in Netware Server, Microsoft shipping for free clients for many
competitive platforms, MS Services for Unix, Host Integration Services.
Even historically, things like LIM, NDIS, adoption of Wolverine stack
for TCP/IP when Windows and Unix worlds began to collide.  There are
obviously issues of contention that have been argued to death (Samba,
"Halloween", Kerberos).  Besides the obvious political and competitive
reasons that someone would argue these points and wish to create this
perception, I don't see how impartial people can ignore all of the
interoperability that Microsoft has made possible.  Are you saying that
people should also throw out XML 1.0 because Microsoft was a primary
author of the spec?  We are saying "XML 1.0 is fine; trust us".  Why do
you believe Microsoft there, but then claim Microsoft is untrustworthy
other times that we say that?  How about all of the client-server and
n-tier web-based systems that use ODBC to abstract data access code from
the database layer?  Were all of those developers duped?  Are you saying
that these people have not received a "winning proposition"?  I would
argue that millions of developers have benefited and perceived a "net
win" from interoperability promises delivered by Microsoft.  In fact,
you would agree that interoperability *is* in Microsoft's best
self-interest in most cases.  I know it is easy to find anti-Microsoft
advocacy sites that take the position you mentioned, but I think it is
incorrect to say that this perception matches the experience of most
developers.