ODF 1.1 is now officially approved as an OASIS Standard in a ballot which ended Wednesday. Accessibility Subcommittee Co-Chair Peter Korn breaks the story.
I played but a bit role in this story, though I watched it unfold with amazement. It was late 2005. A colleague mentioned that there were rumblings of concern in Massachusetts about their recent decision to move to ODF and what impact that would have on persons with disabilities. Although I am not an accessibility expert, I know the basics. (Every programmer should know the basics of accessibility, as well as the basics of internationalization, typography, human factors, performance, security, law, technical writing, project management and how to present and receive business cards in Asia).
Initially, I suggested that a file format has no relevance at all for accessibility. After all, the hard part of accessibility, the integration with screen readers was all at the application and operating system level. What difference could a file format make? The file format is not even involved except when loading or saving the document, right? But since knew ODF, I offered to do a quick spot check and report back. It wasn’t long before I was able to demonstrate a handful of places where data necessary to enable accessibility was not described in the existing specification. For example, although an imported image allowed an annotation of alternate text for use by screen readers, an OLE embedding did not.
To err is human, but what happened next was extraordinary. There is a natural tendency to shrink away from criticism, to retreat inward and retrench, and at all costs avoid admitting errors. But I personally believe that every time we are corrected or criticized, it gives us another opportunity to show our character by how we handle it. The unchallenged person may be a gentleman or a scoundrel. You do not know until he is under pressure. So it is notable that the OASIS ODF TC overcame its accessibility problems not with defiance and not with acquiescence, but by enthusiastically embracing the challenge, engaging the critics, including the aggrieved community, bringing in the experts, both from OASIS member companies as well as outside invited experts, and working within an open and transparent standards development process, rolled up its sleeves and got to work.
The OASIS ODF Accessibility Subcommittee first met on January 27th, 2006. They delivered their evaluation report on ODF accessibility in June of 2006, followed by contributions to the ODF 1.1 specification which was approved as an OASIS Committee Specification in October, 2006, and just this week was approved by the OASIS membership as an OASIS Standard. This took a few days over a year, start to finish.
This is what open standards are all about and why they are so damn important. It isn’t just about patents and lawyers, though that is certainly part of it. It isn’t just about getting your specification approved by ISO, though that is certainly part of it. It isn’t just about how little you can do to earn the label of “open standard”. It is about how much we can do together to improve some parts of the technological landscape that are broken today for some users, and have been for some time.
Congratulations are due the members of the Accessibility Subcommittee for their diligent efforts. But we haven’t heard the last from them. Their charter calls for them to continue their good work, to make additional accessibility improvements to ODF, look at new dimensions of accessibility, consider a wider range of disabilities and create a guide for ODF implementors on the best practices for implementing the accessibility feature of the ODF standard.
I look forward to their contributions to ODF 1.2 and beyond!
Jonathan says
I’m glad that you stopped ranting about OpenXML long enough to talk about the stuff we actually come hear to read about.
So now that we are on the topic of ODF and the speed in which it is changing I have to ask:
When is ODF going to include the stuff we need to implement it right now? You know, like all those config values OpenOffice spews out and the formulas for spreadsheets?
I don’t mean to sound impatient, but it has been over 4 years in the making and we still don’t have all of the basics documented. I don’t know about you, but when I talk about “open standards”, digging through someone elses code isn’t what I have in mind.
Rob says
Jonathan,
If you look around you’ll see no shortage of people writing about ODF news and ODF happenings. I feel no reason to duplicate their good work unless I have a perspective on the issue worth talking about. My goal is to have most of this blog consist of original research and observations.
My commentaries on OOXML, the “rants” as you call them, have been very popular and widely quoted. I’m sure you didn’t visit today hoping to read a status report on ODF, did you?
As for the config values in OpenOffice, do they really belong in the spec? For example, has the CleverAge Add-in for Word found any place where the fidelity of the conversion is diminished because it lacks one of these settings? I’d be interested in hearing if this is an issue.
Personally I think it is axiomatic that an application-independent file format should not specify application-dependent settings. It even sounds obvious when you state it that way. Sure, there will always be application settings, and I think ODF handles this in a sensible way: segregate them to their own XML file and store them as application-dependent name/value pairs.
What else would you suggest? I guess we could declare them to be application-defined rather than application-dependent. That would suggest that application vendors should publish the details of the extensions they add. I guess I’d be open to that.
(Extra points to whoever guesses where OOXML stores their application-dependent settings. And yes, this is a trick question.)
The OpenFormula draft is almost done. The last I heard the Formula Subcommittee was down to their last 20 functions or so. So I’d expect a draft to be posted “really soon now”, followed by review by the full ODF TC, then the solicitation for public comments, followed by OASIS membership vote, followed by submission to ISO as part of ODF 1.2. But vendors obviously aren’t going to wait for ISO publication before they start implementing it. ODF developers are involved with the specification of OpenFormula. So I’d expect implementations to synch up on OpenFormula before this even is voted on as an OASIS Standard.
orcmid says
Hey, kudos!
I would have thought that there was little contact between a format and accessibility as well, but I am learning otherwise. Your explanation is also very helpful.
I’ve got the business card down, but everything else needs more work.
This is nice work. I love the transmittal report from the TC to the big guns.
Anonymous says
This to me shows the difference between OOXML and ODF. We hear that OOXML has flaws, and suddenly it’s something that impugns Microsoft’s credibility, it’s not really as bad as we think it is (even if the files are brain dumps of MS Office), and “contradictions” no longer mean what anyone used to think they mean, so MS should get a standard, too.
With ODF, on the other hand, we hear some complaints that the spreadsheet functions aren’t defined well enough and now we have this group working hard to produce the relevant documentation to address the flaws. We hear that there might be slight issues with accessibility due to it not having a place to store alt tags for some image types, so they add them. In other words, wherever the shortcomings are found, they’re being swiftly fixed.
I really wish Microsoft would start doing that. Sad thing is, it could really improve some of their products.
Joanie says
“What difference could a file format make?”
A huge one when it comes to translating the document into braille. :-) Granted, that is another instance of “when loading or saving the document,” but it’s not often thought of….