Re: [xml-dev] Article on nytimes.com about Microsoft

[Michael Champion <mc@xegesis.org>]

Antitrust compliance is a legal matter on which I am not competent to comment.

 

But please note that I am simply trying to point out the mis-statement in the article "Microsoft has developed its own compression standards". Microsoft (along with several other vendors) has implemented multiple, different binary XML encodings inside  different products, each tuned to optimize different characteristics. All of the Microsoft XML serialization formats are available for license, to the best of my knowledge. None are plausible candidates for standardization; for interoperability we encourage people to use XML 1.0 and other text standards. 

 

Dennis Sosnoski <dms@sosnoski.com> wrote:

> So I gather Microsoft is using this proprietary WCF internal format
>  to
> improve performance between .Net clients and servers, but is not willing
> to make it public and allow other implementations?
>
> This sounds remarkably similar to the behaviors that led to past
> antitrust actions against Microsoft.
>
> - Dennis
>
>
> Michael Champion wrote:
> > As I understand it, the WCF internal format offers performance
> > improvements over XML text for certain classes of SOAP messages. I
> > have no reason to think that its performance/compression is notably
> > better than FI or EXI in similar scenarios.
> > 
> > I wrote "standard" [note the quotes in the original] only because the
> > article said "Microsoft has developed its own compression standards",
> > and WCF's wire format spec is the only thing that even remotely
> > resembles the description in the article. We definitely don't think
> > of this as some de facto standard that is in opposition to FI or EXI;
> > the last thing the
>  world needs is yet another XML-like quasi-standard
> > to further muddy the interoperability story. IMHO, it would make more
> > sense for MS to support whichever standard emerges for efficient web
> > services than to publish and support the specs for the WCF internal
> > format.
> > 
> > I think Microsoft's "binary XML" story is:
> > 
> > - If interoperability is the primary value to consider, use XML 1.0.
> > 
> > - If performance/scalability/bandwidth conservation is the primary
> > value to consider, use whatever technology your platform offers that
> > is optimized for the application scenario you have. For web services
> > on .NET, that's WCF's format. I'm not sure what exactly the other
> > platform vendors have done, but I'm pretty sure they have their own
> > way of working around XML's inefficiencies in their internal
> > communications channels.
> > 
> > - If you want both better
>  performance than XML *and* interoperability,
> > wait awhile to see what actually works and how the standards shake
> > out. As someone mentioned in another post, there are already FI
> > implementations for .NET (and Java, and I'm sure others) -- use what's
> > available even if it's not baked into the shipped frameworks. But
> > let's not standardize first and then figure out if it actually works
> > for real world scenarios or not. And let's not standardize N variants
> > of XML for different application scenarios and expect that there will
> > provide real interoperability across the N**2 variants anytime soon.
> > XML's main value is simply that N=1, and essentially everyone supports it.
> > 
> > The EXI working group apparently believes that N=2 is a manageble
> > number, and that their format can indeed provide both interop and
> > efficiency across a wide range of scenarios. We shall see if they can
> >
>  convince the W3C as a whole of that. Microsoft are not the only
> > skeptics, and not even close to being the most vocal opponents, so I
> > have no idea what whoever leaked that story is trying to accomplish. 
> > Every vendor in this space has a solution that works better in a
> > single-vendor environment than it does in an interoperability
> > scenario. XML makes interop possible, not optimal. 
> > 
> > I do know that we don't consider the WCF wire format a rival to FI or
> > EXI in interop scenarios. We do want to make .NET the best platform
> > on which to develop and deploy web services, but XML parsing
> > efficiency is a small part of that, and the framework can easily
> > support FI or EXI if one of them emerges as an interoperability standard.
> >
> > */Dennis Sosnoski <DMS@SOSNOSKI.COM>/* wrote:
> >
> > Michael Champion wrote:
> > > It sounds to me like somebody is trying to use Microsoft's
>  public
> > > skepticism about EXI (Efficient XML Interchange) for some purpose or
> > > another that is hard to guess at without knowing who is doing the
> > > off-the-record whispering. I assume the event 4 years ago was
> > the W3C
> > > workshop on binary XML, and the rival compression "standard" is the
> > > internal format that WCF can use to communicate with other WCF
> > > implementations.
> >
> > This is an interesting development. Is there any public information on
> > this format? Certainly many other web services stacks would be
> > interested in implementing it as well, if it does offer significant
> > performance improvements.
> >
> > - Dennis
> >
> > _______________________________________________________________________
> >
> > XML-DEV is a publicly archived, unmoderated list hosted by OASIS
> > to support XML implementation and development. To minimize
> > spam
>  in the archives, you must subscribe before posting.
> >
> > [Un]Subscribe/change address: http://www.oasis-open.org/mlmanage/
> > Or unsubscribe: xml-dev-unsubscribe@lists.xml.org
> > subscribe: xml-dev-subscribe@lists.xml.org
> > List archive: http://lists.xml.org/archives/xml-dev/
> > List Guidelines: http://www.oasis-open.org/maillists/guidelines.php
> >
> >