More on TechnologyTell: Gadget News | Apple News
Standards rot and the Tower of eBabel
July 7, 2006 | 1:07 am
|
By David Rothman
You’ve heard of rotten apples and link rot and a few other kinds. Now here’s another neologism, standards rot.
I’m not the very first to use the term. But oh how it will fit the world of e-books if publishers, libraries and other major stakeholders fail to do sufficient rot-proofing.
The following is my first stab at a definition of “standards rot”:
A deterioration of standards in software or other areas–to the point where open standards become proprietary again. Common causes are (1) undue influence by individual corporations on standards bodies, (2) lack of diligence by companies designing new products, (3) persistent, mutually agreed on ego protection among practitioners of the Not Invented Here Syndrome, (4) insufficient planning of forward and backwards compatiblity so that standards lose their attractiveness and (5) conspicuous or stealthy circumvention of open standards through the introduction of proprietary features.
The International Digital Publishing Forum has been a textbook example of the first three causes and seems keen on qualifying for a mention of the fourth. The fifth possible occurence, in the form of Flash or other adds-ons, will be inevitable if standards are not moved to a more neutral forum. Otherwise the Tower of eBabel will be eternal and commercial e-books will remain a joke–computer “products” more than beneficiaries of a durable format fit for serious literature.
The more the IPDF partisans defend their failure to do consumer-level standards promised in ’98, the more bizarre the groups seems. It’s urgent to move standards out of the IDPF before too much rot sets in.
|
Previous
SUBSCRIBE TO RSS
Comments:
David,
I agree about the danger of “standards rot” but I think you missed a few of the most causes. I will attempt to extend your list with:
6) Non-adoption in mainstream, widely used applications. OEBPS 1.2 is a case in point (i.e. everyone using OEBPS has seemed stuck on 1.1 or earlier), but there are many others – XForms for example is arguably on the bubble wrt this. It’s the “use it or lose it” principle: a widely-adopted standard won’t rot because there’s too much at stake (it might “freeze” but that’s another disease).
7) Cancerous splitting. Linux is a case in point here. Nothing destroys a would-be standard faster than “dialect-itis”. This can be driven by your cause #3 (Egos/NIH) or for other motivations.
Regarding IDPF, as usual you appear off-base.
#1: Undue influence of individual corporations. IDPF has an extremely broad membership base and a well-defined, transparent governance process. The elected Board reflects this in representing publishers, etailers, integrator/aggregators, vendors, and libraries. And since these individuals are employees of their various entities, their financial ties are aboveboard.
#2: Lack of diligence. You don’t explain why you feel IDPF suffers from this but IDPF’s recent work on OCF for example was apparently good enough to be “borrowed” for your proposed OpenReader format.
#3: Ego/NIH. Again OCF is a good example: the WG adopted and generalized the packaging component of an existing standard, OASIS OpenDocument. The work to make it more general is now under consideration to move into OASIS, as part of a planned broader collabration between IDPF and OASIS. No symptoms of ego or NIH here I can detect.
#4: Insufficient planning for forwards/backwards compatibility. Only time will tell here but the OEBPS WG is taking this seriously.
#5: Submarine proprietary features: No evidence of same.
#6: Adoption: Key members have publicly announced intentions to adopt.
#7: Spitting: IDPF is working to unify and eliminate the Tower of eBabel.
Now how about we look at OpenReader on this list?
#1: Undue influence: With a murky “consortium” that says or does whatever its self-appointed leaders decide, undue influence is clearly a danger. And, for-profit corporation OSoft is clearly right smack in the smoke-filled room. Financial promises from OSoft to Rothman contingent on OpenReader’s success have been disclosed. Without a governance structure or any other transparency it is unclear what else may be going on. It is also not clear what other financial ties OpenReader principals may have. Could some proprietary vendor who doesn’t want a single unified open standard be encouraging them to splinter? I trust not, but the point is that we don’t know.
#2: Diligence. Well there’s no WG structure around OpenReader’s technical development and so I guess we are simply to trust that Mr. Noring (whose main claim to fame is apparently his work on the zero-adoption OEBPS 1.2) is diligent. Hm.
#3: Ego/NIH. IMO the primary reason OpenReader is still intent on defining “its own format”, against the interests of a unified standard. OpenReader folks talk the talk about cooperation with groups like OASIS (vs. IDPF who’s actually doing it).
#4: Planning for forward/backwards compatibility. OpenReader is intent on reducing compatibility via a “fork” of OEBPS.
#5: Stealthy introduction of proprietary features. With the majority of OpenReader’s proposed format “in preparation” (i.e. not publicly available) it’s hard to be sure, but all their talk about “friendly DRM” combined with OSoft’s admission that they have a patent-pending DRM technology has to make one wonder.
#6: Risk of Non-Adoption. http://www.idpf.org/membership/currentmembers.asp vs. OSoft and ??
#7: Cancerous splitting. IMO the cardinal sin that OpenReader is attempting to impose on the digital publishing community.
David, I don’t care what happened in 1998. It’s 2006 and we need to rally the community and make digital publishing happen. You are a strong voice for open standards and the compelling need to make content broadly available – can’t you put your ego aside and help make your vision happen rather than trying to undermine it?
Bill, where in the blazes do I start with such a comment? Unfortunately, I lack the time that you apparently have, so can’t do a point-by-point reply. So I’ll just make a few independent comments.
First, OpenReader is committed to its format being completely an open standard. Your implication otherwise by trying to weave a “grassy knoll” conspiracy theory with smoke-filled room allusions and the like is completely disingenuous.
Second, OEBPS WG has intentionally discarded the huge amount of work done by PSWG in the 2000 to 2003 time frame to develop the next generation digital publication framework. Thousands of man-hours (by very sharp people some of whom I identified by name in a prior comment) were expended to get a good grasp of where the spec should evolve the next few years. When OEBPS WG was formed, that was, in essence, thrown out the window with nary a comment. None of that great work is being factored in the current OEBPS WG decision-making process. None of it.
The OEBPS-Next spec is now being built “spur-of-the-moment” based on the short-term needs of two companies, not on what is best for the long-term of the entire digital publication industry. It is being built like the Winchester House was built — no long-term planning factors in decision-making at all — just do what feels good for the moment. As a contributor myself to the current OEBPS WG effort, I am constrained by the Charter, but let’s say I’m not happy with the general direction OEBPS is taking based on the Charter which simply codifies the “live for today” philosophy.
Third, your implication that OEBPS 1.2 has not been implemented because it must be flawed (yes, that’s what you’re trying to imply) is also disingenuous. An objective analysis of OEBPS 1.2 versus 1.0.1 shows that they are very similar, and 1.2 fixes a lot of problems in 1.0.1. That 1.2 is not being implemented was nicely covered by Ben Trafford in one of his comments: 1.2 was issued during the “e-Book Dark Ages” when there were few, if any, new OEBPS implementations. Microsoft, for example, based their LIT format on OEBPS 1.0.1, and never updated it for 1.2 since they, too, were reevaluating their e-book future during the “Dark Ages,” as was everybody else.
Thus, to restate what I said before: IDPF should have, instead of the current OEBPS WG effort, issued OEBPS 1.2.1, which I mostly completed late last year since I was the chair of the PSWG Maintenance subgroup — it could have even been “modernized” in a few places and issued as 1.3 — that work would have been completed a few months ago ready for Adobe and others to adopt and run with it quickly. Then, earlier this year, work could have started on the true next-generation OEBPS in the then-existing PSWG, picking up where PSWG left off in 2003. And to do things right using a “Roadmap” approach, to build upon all the great work already done.
Fourth, about your comment on OpenReader adopting a container similar to IDPF Container, we did this since it makes sense. The IDPF Container is simply compatible with the ODF Container, with JAR, with ZIP, etc., so we’re joining good company (and of course ODF is being worked on in OASIS.) I know if I said we’d develop our own container based on some other approach, you’d then say we are trying to fork things again. So we’ll do what makes sense to do and not worry about the Catch-22 traps people may try to throw in our path.
Finally, your several month obsession with that little two-person company called OSoft is troubling, and I finally have to comment in public.
OSoft has committed to fully open source their next-generation reading system codebase, the dotReader, and anyone will be able to use and modify it (their current ThoutReader is now open source.) Will Adobe, in their secret development of an OEBPS-based reading system, do likewise? Why does an 800 lb. gorilla that is Adobe concern itself with a little two person company anyway? What are you afraid of? The free market? Competition?
Importantly note, too, that OSoft’s next reader will also be able to easily render OEBPS, mainly because of the similarities between OpenReader and all flavors of OEBPS. (Contrary to what some have said, it’ll be easy for publishers and conversion houses to produce both OEBPS and OpenReader on the same “assembly line” since they are quite compatible. OpenReader, is, frankly, superior to the current and the planned OEBPS-Next, for reasons I’ve discussed previously. In fact, I believe OpenReader+SVG will be better for Adobe’s plans than would OEBPS-Next.)
Personally, as one who always cheers for the little guy, I find your continued verbal warfare against OSoft the last few months to be patently offensive, and not befitting an IDPF Board Member. You should be cheering for OSoft since they are an IDPF member, and it was a stretch for them financially to join IDPF in the first place (IDPF membership fees are still too high, in my opinion.) Mark Carey, the CEO of OSoft, was even willing to commit valuable time, which he has little of these days, to serve on the IDPF Board when he ran for it in May.
Your continued harassment of a small company and certain individuals, who are trying to do the right thing, and not motivated by money or power or ego, is beyond belief. Or is the “right thing” only what Adobe wants?
You can continue to say that you are speaking for yourself and IDPF, but just as you can’t, or won’t, separate the various hats David Rothman wears, so too, whatever you say regarding e-books, is representing Adobe’s position. So your attack on OSoft, on OpenReader, etc., is official Adobe business regardless of what you say to the contrary. You can’t have it both ways, especially given your position in Adobe at a high executive level. Sorry to have to say this, but in my mind, and I believe for many others reading this, you won’t be able to truly speak independently on e-book matters until you leave Adobe and are no longer on their payroll in any capacity. Live with that reality.
[P.s., on a different topic, I find it ironic that OEBPS WG is considering adopting Relax NG for schema purposes. Why not W3C Schema? Obviously Relax NG, developed through OASIS (where have I heard of that organization before?), and now an ISO standard, was a lean and mean schema standard not developed under W3C auspices and I believe was considered competitive to W3C Schema at the time. Talk about a fork in important industry standards! Horrors. For the gall of some people to independently develop standards that were most appropriately developed in W3C — oh the tragedy to the common good for going it alone.]
Jon, it seems you’ve missed the sardonic thrust of Bill’s comment. He’s basically applying the same “tinfoil hat” attitude that David Rothman has been spreading all over the net about IDPF.
To rebut and comment on a few of your specific points:
“First, OpenReader is committed to its format being completely an open standard.”
Then take it to OASIS. Now. Why wait?
“None of that great work is being factored in the current OEBPS WG decision-making process. None of it.”
Um, hi? Remember me? I’ve read every document that you’re discussing, and in my own work as an invited expert, determining what to use and what not to use. I’m sorry you’re angry that work you did seems to have been for naught, but I think we can agree that the people on the WGs are smart enough to figure good ideas when they see them.
“The OEBPS-Next spec is now being built “spur-of-the-moment” based on the short-term needs of two companies…”
You and Rothman keep making this claim. To date, in the last month that I’ve been in contact with IDPF people, I’ve seen nothing of the sort. Of course they’re basing some of their work on their needs — just as you’re basing your OpenReader work on your perceived needs. That doesn’t make it evil, Jon. That makes it smart.
“Thus, to restate what I said before: IDPF should have, instead of the current OEBPS WG effort, issued OEBPS 1.2.1, which I mostly completed late last year since I was the chair of the PSWG Maintenance subgroup — it could have even been “modernized” in a few places and issued as 1.3…”
Except that it isn’t cooked, Jon. Even the most cursory examination of it shows that it isn’t cooked, and that it doesn’t really take into account changes in technology in the last three years. I’m sorry, but it doesn’t. And I think we can agree that I’m technically savvy enough to be able to say that, yes?
“I find your continued verbal warfare against OSoft the last few months to be patently offensive, and not befitting an IDPF Board Member.”
But David Rothman using this blog as a forum to attack the IDPF is somehow okay? I mean, it’s not even like he’s commenting on technical issues or suggesting new ways to do things, he’s launched an all-out verbal war on the group, including personal comments such as: advertising Nick Bogaty’s salary (which is nobody’s business), claiming that ETI and Adobe are somehow in cahoots and stepping on the little guy, and most recently, calling me both stupid and for sale.
Sorry, Jon. You can’t call Bill for stepping over the line when you’re buddy David is -trampling- it. We’ve both known several of the people he’s slamming for years. I have great difficulty with assuming nefarious behavior on the part of any of the ETI folks — the same people who argued for more openness at the OEBF, who pushed for accessibility, who put their businesses on the line to support something bigger and better than an HTML subset.
It just doesn’t fly.
“I believe for many others reading this, you won’t be able to truly speak independently on e-book matters until you leave Adobe and are no longer on their payroll in any capacity. Live with that reality.”
Okay. You need to be independent and unemployed by an ebook company to be able to comment? Great. If Bill can’t say it, I will: you guys are dividing an ebook community with disingenuity; you’re creating a splinter standard that you fully admit has been lifted from the work of others; and you’re trashing the work of good people who just want to see ebooks fly into the mainstream.
So far, I haven’t heard -any- justification for this course of action beyond, “I did a bunch of work on something and they’re not using it, so screw them!”
I think you guys are really going about this the wrong way, Jon. And to be entirely blunt, if you’re so against the IDPF, then why the hell are you still showing up at the meetings? Go your own way! Take OpenReader to OASIS, let us do our thing, and may the better standard win.
(For the record — my comments do not represent any company. I am not employed nor contracted to any company that is working on ebooks, or any aspect of digital publishing. I haven’t been told to say this by anybody. I am not, nor have ever been a Sith apprentice, and I did not sleep with that woman.)
Bill,
I thought we have had an understanding for several months about not publicly “attacking” each other. I have kept my promise – you have not. Your comments in the above blog mentions OSoft unfavorably in a number of instances.
Your allegations are totally unfounded and I would like you to stop immediately.
# 1 – OSoft’s new dotReader is format agnostic. Recently, OSoft participated in the IDPF press release indicating that we would also support the new OEBPS standard. We are supporting multiple XML standards and are not wed to OpenReader; we deliver whatever the market wants. I even ran for an open IDPF board position in May.
# 2 – OSoft does not drive the market with our own agenda. We respond to market demand.
# 3 – David Rothman does not speak for OSoft – I DO.
# 4 – OSoft has a DRM model. So does Adobe, MobiPocket, Microsoft, and others. Why single us out in your comments? OSoft’s official position about DRM can be found here.
# 5 – OSoft is an endorser of the OpenReader format. So is Fictionwise and eBooks.com. I didn’t see their names mentioned. Here’s the complete list.
# 6 – In your comment, “Financial promises from OSoft to Rothman contingent on OpenReader’s success have been disclosed.”, I do not know where you are getting that information. Library City is a not-for-profit organization that I personally support as a former and long-time library trustee. There are no financial promises and never were. (Please re-read this thread.)
Our dotReader is open source and like two other non-profits I am working with, I have agreed to customize the dotReader (which reads multiple formats) into a Library City version. I am also doing the same for TeachersWithoutBorders.org and will be rolling out a major initiative in Pakistan designed to give teachers and students better online learning tools. I do not ask which social programs you invest in – don’t question mine.
Quite frankly, I am delighted that David Rothman likes the dotReader and would like to adopt it for Library City. He’s not the only one…
I am very sorry you and David Rothman continue your verbal attacks. I want nothing to do with it. OSoft is totally focused on our own business and we are distancing ourselves from this fray. PLEASE keep us out of it! You are not informed about OSoft’s new business model and have no right to disrespect us in a public forum.
I would greatly appreciate your consideration in this matter.
Respectfully,
Mark Carey
President
OSoft.com