November 7, 2003
@ 03:23 PM

I've posted previously on why I think the recent outcry for the W3C to standardize on a binary format for the representation of XML information sets (aka "binary XML") is a bad idea which could cause significant damage to interoperability on the World Wide Web. Specifically I wrote

Binary XML Standard(s): Just Say No

Omri and Joshua have already posted the two main reasons why attempting to create a binary XML standard is folly (a) the various use cases and requirements are contradictory (small message size for low bandwidth situations and minimal parsing/serialization time for sitautions where minimizing processing time is prime) thus a single standard is unlikely to satisfy a large proportion of the requesters and (b) creation of a binary XML standard especially by an organization such as the W3C muddies the water with regards to interop, people already have to worry about the interop pain that will occur whenever XML 1.1 gets out of the door (which is why Elliotte Rusty Harold advises avoiding it like the plague) let alone adding one or more binary XML standards to the mix.

I just read the report from the W3C Workshop on Binary Interchange of XML Information Item Sets and I'm glad to see the W3C did not [completely] bow to pressure from certain parties to start work on a "binary XML" format. The following is the conclusion from the workshop 


The Workshop concluded that the W3C should do further work in this area, but that the work should be of an investigative nature, gathering requirements and use cases, and prepare a cost/benefit analysis; only after such work could there be any consideration of whether it would be productive for W3C to attempt to define a format or method for non-textual interchange of XML.

See also Next Steps below for the conclusions as they were stated at the end of the Workshop.

This is new ground for the W3C. Usually W3C working groups are formed to take competing requirements from umpteen vendors and hash out a spec. Of course, the problems with this approach is that it doesn't scale. It may have worked for HTML when the competing requirements primarily came from two vendors but now that XML is so popular it doesn't work quite as well, as Tim Bray put it "any time there's a new initiative around XML, there are instantly 75 vendors who want to go on the working group".

It's good to see the W3C decide to take an exploratory approach instead of just forging ahead to create a spec that tries to satisfy myriad competing and contradictory requirements. They've done this before with W3C XML Schema (and to a lesser extent with XQuery) and the software industry is still having difficulty digesting the results. Hopefully at the end of their investigation they'll come to the right conclusions.