ISO Releases OOXML FAQ

I Don't Believe in Imaginary Property writes "The ISO has put out a FAQ concerning OOXML, but it may raise more questions than it answers. For one, it promises to address problems if they arise in the future. PJ of Groklaw said that's akin to 'selling you a car with four different sizes of tires and assuring that that if you see it's a problem, you can always bring it in for maintenance.' It also handwaves the OSP discriminatory patent promise issues, when asked about contradictions states that some 'may still remain', and asserts that duplicate standards are 'something that need[s] to be decided by the market place.' Notably, the FAQ does not answer the question, 'what the hell were you thinking?'"

Comments

This one's good.

[Jason Levine]

About investigating charges of corruption of the voting process:

We reviewed the process before it started, all the while during its course and afterwards as well. While the voting on ISO/IEC 29500 has attracted exceptional publicity, it needs to be put in context. ISO and IEC have collections of more than 17 000 and 7 000 successful standards respectively, these being revised and added to every month. This suggests that the standards development process is credible, works well and is delivering the standards needed, and widely implemented, by the market. Because continual improvement is an underlying aim of standardization, ISO and IEC will certainly be continuing to review and improve its standards development procedures.

So they're basically saying: "Since we've done a lot of successful standards before, there can't possibly be anything wrong with how this one was carried out."

Re:What the hell were they thinking?

[zappepcs]

HUH?

What they were thinking was that someone offered a specification for standard and they saw the necessity of having a standard specification and they went ahead and approved it. I believe they are supposed to evaluate that 'standard' to ensure that it meets the specific and general requirements of such a standard before accepting it as a standard. Paying off the mortgage on the summer house is not one of the requirements, BTW.

Whatever PJ thinks is hardly relevant here. What any individual thinks about the new standard is irrelevant except to the extent that he needs to use it. Did you ever take any of those logic tests? Do well, did you?

What people think of the 'standard' is totally relevant. Simply blindly accepting something as the golden rule is ignorant, and this will (probably) lower the esteem of this standards body for a very long time. That is damaging to the purpose of standards, and part of the reason that there are not 47 international standards bodies.

Since OOXML is not the only specification out there, it behoves anyone with contrary feelings to promote their favorite standard rather than try to bring down OOXML. Okay, back to your logic problems. How do you promote your own favorite standard without verbally bashing this one that is trying to supplant the good value of your favorite standard?

Yes, I know that sounds like being negative, but you must remember that using OOXML as a design example of what standards SHOULD NOT BE is a valid method to promote the standard of your choice.

I wonder...

[jd]

...if ISO was paid in euros or dollars. You'd probably want to fast-track to this degree if your bribe goes down in value the longer you take.

Re:Just a small detail

[clodney]

Lots of standards have revisions. I presume that before we had 802.11 a/b/g/n we simply had 802.11.

Just hours ago I was reading the TWAIN 1.9a specification. 1.9a being a big tip-off that the spec has changed over time.

My TV and DVD player are connected with HDMI 1.3 compliant cables.

So yes, if there are problems with the standard they will change the standard. That is standard behavior if you will.

ISO is not like IETF

[grandpa-geek]

IETF requires demonstrated interoperability using prototype reference implementations before they will adopt a standard.

ISO generally first adopts standards, then waits for people to prototype implementations and discover the bugs in the standard (unless someone walks in with existing technology and asks for it to be standardized). When people start reporting that aspects of the standard can't be implemented, ISO works on fixing it.

After ISO adopted the Open System Interconnection (OSI) standards, they had to set up "implementers' workshops" to figure out how to make their newly adopted standards workable. (The OSI standards are the 7-layer reference model and related protocol suite that were pushed aside by the Internet protocol suite, a.k.a TCP/IP. Many OSI protocols were never fully implemented or never made to work.)

The workshops met (one was sponsored by NIST) and produced a lot of documents on things that needed to be done to make OSI work. When the Clinton/Gore administration came into office, they killed US government support for the OSI protocols and told its agencies to use the Internet protocols.