The IDPF Standards ‘Bandwagon’
By Jon Noring
Today, IDPF announced the release of the Open eBook Publication Structure Container Format 1.0 (OCF). This standard is a step forward and I congratulate all who made it a reality. It will be useful for many e-book applications.
But the IDPF continues to paint the OCF as a much grander thing than it really is — making a mountain out of a mole hill. This is disturbing since IDPF refuses to address head-on the most important issues regarding the “Tower of eBabel,” such as an industry-wide standardization of a consumer-acceptable DRM system.
For the details of OCF’s suitable role in the digital publishing ecology, refer to the blog article I wrote in late June about the OCF, OEBPS and OpenReader: “Cotton candy PR vs. genuine next-generation standard from OpenReader”.
I invite publishers, especially those who are members of IDPF, to carefully read the cotton candy article since it details what OCF is, and more importantly, what it is NOT. It also puts into perspective the current standards push at IDPF which is critically flawed in several ways. This is tragic in that IDPF is trying to develop standards for a future multi-billion dollar industry, but doing so pretty much in an ad-hoc manner, virtually isolated from other major digital publication open standards work and organizations (NIH Syndrome?).
Bill McCoy, in a blog article about the OCF announcement, describes the IDPF standards push like an unstoppable “bandwagon” and implicitly urges everyone to climb onboard and enjoy the ride. But I prefer to describe it as a bandwagon that is unstoppable only because it is moving rapidly downhill, totally out of control.
(P.S., besides contributing to OCF, I contributed to all versions of OEBPS, including the ill-fated OEBPS 2.0 and the new incremental version of OEBPS now under development. So, despite the apparent lack of memory exhibited by some, I have significantly contributed my share to the IDPF standards. I have the credentials, as a long-time expert in IDPF standards development including stints as a subgroup chair and acting vice chair, to voice my concerns and to point out whenever the Emperor has no clothes.)
October 31st, 2006 at 10:32 am
Do you have supporting facts to back up your expressed serious concerns?
- “NIH” – the OEBPS has adopted technology approaches from OASIS OpenDocument and DAISY/NIMAS, as well as of course XHTML & CSS. These are obvious facts that suggest the opposite of NIH. If you are concerned about NIH, why?
- “out of control” – the OEBPS just completed OCF in a timely manner, per a clearly defined process. This fact suggests the opposite of “out of control”. Again, if you are concerned, what’s your beef?
- “critically flawed in several ways” – presumably you can enumerate said flaws?
October 31st, 2006 at 12:32 pm
<laugh> Where do I start? </laugh>
Regarding “out of control” — yes, the IDPF standards process is out of control. Just as one shouldn’t build a business without a business plan, IDPF should not build a standard without a comprehensive Roadmap that guides all decision-making. For both OEBPS and OCF, there is NO Roadmap to speak of — decision-making is pretty much ad-hoc. Reminds me of the late 60’s where just do what feels good. The original OEBPS Roadmap has essentially been ignored. (A Roadmap goes beyond just a requirements gathering exercise, which also needs to be done properly — last October’s exercise in IDPF was fatally flawed and pretty much useless.)
Regarding “getting a spec out in a timely manner” — just because a spec was done fast, does not necessarily mean it was done right, something which Bill and others refuse to even comment on (or if they do, they’ll simply say “it was done right.”) A spec, especially a complex spec, not done right is a spec that is “out of control” and likely to create more problems than it solves. For complex specs (like OEBPS) which are updated with new innovations (and not simply maintenance updates), making decisions without a comprehensive Roadmap is not doing things right — it is reckless, and the IDPF membership should not tolerate this.
Regarding OCF, that is such a simple spec, and as Bill noted is heavily based on the ODF Container, Java JAR, etc., that is was mostly done already. The OCF spec has so few innovations in it that it didn’t really need a Roadmap to design. However, future versions of OCF, especially as it merges with the ODF Container (assuming that happens), and thus has to be generic, should not proceed until a Roadmap is completed. (Btw, for the future OCF, probably best renamed “Open Container Framework,” IDPF should not only work with the ODF folk, but also bring in other digital media groups to make it truly a media industry-wide standard.)
I’ve been meaning to cogently summarize in one or a series of blog articles all my thoughts regarding IDPF (including the Board makeup and decision-making which has major problems that led to the OEBPS fiasco), OEBPS, OCF etc. But I cannot begin writing these articles for at least a few days.
Yes, I am fairly emotional about the IDPF standards area, mainly because the leadership, for a few years, was not fully committed to standards (OEBPS 2.0 did not happen, I believe, due to the lack of interest by the leadership.) The renewed interest in standards is, I believe, reactionary, and the result has been to throw out proper standards development procedures based on comprehensive requirements gathering distilled into a long-term Roadmap. IDPF can certainly fix these problems, but I’m not sure the Board will be willing to do the necessary fixes, especially in that 2 of the 7 Board members clearly like the way the standards process is now proceeding since they are the ones who restarted the flawed effort, submitted the Charter wording, now lead it, and the resulting standards directly benefit their own business development.