The Case for ODF--Presentation Given in Seoul, December 2006


The danger of incompatible formats
Everyone here can recall the catastrophe that was Katrina. Last year's hurricane devastated the US city of New Orleans; the government response almost finished the work. Much of the problem did not lie with scarce resources—the usual excuse for poorly done jobs—but with chaotic communication protocols and policies that allowed such a mess to develop.

A by now famous case in point, from an article by David Kirkpatrick of Fortune Magazine, which one can find mentioned in Pamela Jones' Groklaw:
In the immediate aftermath of the hurricane, much of the region’s communication systems failed or didn’t work properly. Water and wind knocked out power, toppled phone lines, and destroyed cellphone towers. What systems remained were quickly overwhelmed. When rescue workers’ did have working equipment, like walkie-talkies, they often couldn’t connect with others on different communication systems.
The point is that different communication standards exacerbated the already terrible damage caused by the hurricane and flooding. Rescuers were unable to communicate and coordinate their activities.

And, as Jones goes on to point out, the ePolicy Group explained that the same thing happened with the Boxing Day tsunami of 2004:
Responding agencies and nongovernmental groups are unable to share information vital to the rescue effort," the report recalls of the government in Thailand in the tsunami's immediate aftermath. "Each uses different data and document formats. Relief is slowed; coordination is complicated. The need for common, open standards for disaster management was never more stark or compelling.
Incompatible formats kill. The issue is not academic and it is not merely a thing of money, affecting the affluent and educated. Nor is the issue of formats merely a business strategy. Rather, it is a question of life and death whose seriousness has been, despite at least two horrendous recent catastrophes, downplayed.

The danger of proprietary formats
Moving to a single format would help, clearly. But it wouldn't necessarily help in the long run. Here's a simple thought experiment. A nation comes to depend on a single company to provide all its software needs, including office needs.

Sound familiar?

(But let's ignore the issue of cost; this is a fantasy, after all.)

And one day for whatever reason—security, markets, technology—there is change and the country suddenly realizes its intellectual property, or at least access to it, is not really under its own control but is rather subject to the interests, whims, desires of the company.

Or another fantasy—that no matter how seemingly convenient the application has been, it just won't work with, say, the new Google Web applications or the collaborative software that is now required for business or academic research and development. Why? Because it is proprietary and uses formats that differ from the standard and the company sees no reason to spend money on the desired functionality, as its larger customers seem happy enough.

I'm sure you get the picture: depending on a single company to produce and safeguard your intellectual property is dangerous, as it can lead to the loss not just of opportunity but of your control over your data. Depending on proprietary formats in this day and age is equally problematic. In both cases, the nation depends on something crucial that is out of its ability to control.

The open standard, OpenDocument Format
All which brings us to the OpenDocument Format (ODF).

I'm sure you all know these things and I apologize if I only go over something that' is familiar. But it helps that we are all in agreement on the meaning of terms like "open standard," "XML," and, of course, "OpenDocument Format" or ODF.

Open standards
Open standards are publicly implementable standards: any compliant application can implement them. The application could be open or proprietary; it does not matter. There are several international standardizing bodies that approve standards, and these range in industry support and importance. As I further point out, though, the approval of any body is but one measure of the openness of the standard. The other is whether there exists more than one implementation.

The ODF has been approved by the two most important international standardizing bodies: the Organization for the Advancement of Structured Information Standards (OASIS) and the International Organization for Standardization (IS0). Both bodies are nonprofit international consortia composed of hundreds or even thousands of companies, government agencies, and NGOs; they do not answer to any company but rather balance the interests of government and industry.

Why are open standards good for industry? Open standards lower costs of production and consumption and allow companies to focus on what the product actually does, or rather, allow the user to get the most out of the product without having to worry about the logistics of working her product.

The ISO approval of ODF is particularly noteworthy, as it means that governments throughout the world may now utilize the format. (Many governments are required by law to use only ISO-approved formats or standards.)

The OpenDocument format
In brief,
The OpenDocument Format (ODF) is an open, XML-based document file format for office applications that create and edit documents containing text, spreadsheets, charts, and graphical elements. The file format makes transformations to other formats simple by leveraging and reusing existing standards wherever possible. (Oasis whitepaper, ODF Advantages, November 2006)
XML (Extensible Markup Language) allows the "sharing of data across different information systems."This is why XML is preferred by ecommerce and by just about any organization interested in doing work that exploits the sharing capabilities of the Internet.

The key advantage then of the ODF is that it is an open, XML-based format that re-uses existing standards. Those developers of whatever rank wanting to work with it do not need to be concerned about encountering new technologies. The structure is clear and logical and the file contents are more or less human readable and certainly manipulable with any text editor. One just has to unzip the file. Furthermore, because the ODF is economically described (one describes XML documents), it is relatively easy for developers to write new implementations or find other ways of exploiting the format.

Remember, we want that. We want as many organizations as possible to use the ODF, and the ODF has thus been designed from the ground up to be open and easily accessible, and the process for its standardization has been similarly open and public. There is nothing secretive about it or the process. The goal is to build a real ecosystem of applications, companies, organizations, around it, not to extend a monopoly.

But the test of an open standard is not just whether this or that international organization, however powerful and important, has approved it. Rather, it lies in the number of implementations available. And in the case of ODF, there are many, and they are not all open source. They include products from the leading enterprises of the day, such as Google's Writely, Sun's StarOffice (StarSuite), IBM's Workplace, KOffice, OpenOffice.org (which I lead) and many, many others. (A fairly current list of ODF applications can be found at http://opendocumentfellowship.org/applications.) We see here some images of ODF implementations, courtesy of he ODF Adoption Committee and Erwin Tenhumberg:

Note, they are not entirely identical. Each application can have its flavor, its features, its layout, its look while still being able to read and write to the format.

Inter-ODF Interoperability
But, you might ask, how well do the various applications using ODF speak to each other? That is, if I create a document using OpenOffice.org, how will it look in Workplace? Or AbiWord, or any of the others?

The answer is that to date, it depends on the document. Some are more compliant than others, with OpenOffice.org and its direct derivatives being the most. But this uncertain state is purely transitional. As a member of the OpenDocument Adoption Technical Committee, I have bee pressing to establish compliance rules, so that users know what they can expect and so that developers and product managers can generally make it so that users need not have any concerns about the specific application they choose.

The practical and economic benefits of ODF
For most users, however, the differences that are today present in the way implementations handle ODF are insignificant and certainly not barriers to widescale adoption. The practical benefits of deciding now to utilize the format (or more accurately, implementations that use it) far outweigh any shortcomings. And, in fact, the very proliferation of vendors and implementations suggest how Korea can directly benefit locally while participating globally.

And this is the fun part.

Using the ODF is good for the nation, both socially, culturally, and economically. In fact, the three are intimately linked, which is why choosing a format that transcends the narrow boundaries of any one company, however large, and which allows for multiple vendors is of paramount importance.

Besides the reasons I pointed out earlier—that an open standard widely deployed and implemented can save lives—the ODF has the more humble but, arguably, no less important capability of preserving a nation's (or any body's) intellectual property. It's also, as I will briefly sketch out, implicitly democratic.

We don't think of nations as really having a wealth of intellectual property, but they do. Much of it is what we see in museums, or appreciate in the daily cuisine or folklore of the nation: it is what people have done and do. And this wealth can be anything but glamorous. All the documents related to taxes—yes, taxes—to business dealings, to law, to academia, to homework, to newspapers; to all those things that in their daily pulse, the measure of your doings, make up the nation that we know: this is your wealth. As a record of your doings and thoughts it is passed to others, around you and to the future. So that just as we now can learn how those who came before us lived and thought and appreciate what they did, just as we can benefit from their wisdom, from recorded history, so too our descendants will be able to benefit from the record of our thoughts, no matter how humble those thoughts be.

And it's endangered.

It's endangered by the use of proprietary formats, to be sure, and by the use of formats that lack wide implementation and that make it difficult to create implementations. The reasons by now should be obvious: both have the effect of hanging the work you have done, the national community has done, on the thread of a vendor's goodwill or economic health.

What is more, proprietary formats are, almost by definition, self limiting. If user A creates a document in a proprietary format with a proprietary application, then user B will have to use the same to read it, or, conceivably, use a reverse-engineered version of what A used. Let's imagine that scenario on a national scale. The government creates legal memoranda using proprietary tools. As a resident you are entitled to read it and you want to, for it pertains to your taxes. But in order to read it, you need the same proprietary tools as the government used. In other words, you need to spend money, extra money, to read what you are otherwise entitled to read.

It's not just anti-democratic. It's counter-intuitive. For the government does not receive the money that you must spend on the application. The company that made it does. And it could be very remote, its headquarters thousands of miles away, in a wholly different country. Naturally, that country does exact taxes from the company that made the software. And this leaves us with the absurd sitution: Your national duty, in short, benefits not your country but another.

An open standard such as ODF resolves that problem. It does so by:
• allowing both proprietary and open source implementations
• lowering the barrier for entry for interested companies (no license fees)
• maintaining format integrity over time through an open and transparent process

A typical scenario, then, is that user A creates a document in an open format using a proprietary application, such as StarOffice or Workplace. Users B through Z and beyond—basically, everyone else—can then read that document with both proprietary and open source applications. They are not limited by license. Thus, if user A is the bureaucrat and he's created a legal document all should know about, users B-Z, the people, can use OpenOffice.org to read it, fill in the blanks, even edit it, all as the original author sees fit. And they can do this without spending any extra money: OpenOffice.org is free—gratis--and will stay that way. And if they choose instead to use KOffice, why they are free to do so; it also is free. As is AbiWord; as is Writely. As are others and as there will be others.

Extend the scenario through time and not just place. The users of the future will still be able to read the document because the format will not depend on the existence of any one company—the company that "owns" it (no company will own it). For that very reason—that it is independent of the economic health or desires of any one company, the ODF is something that one can bank on. That is why governments demand and depend on the ISO for their standards, so that they will not find themselves investing resources in something that will disappear with the next economic downturn. Governments must depend on permanence. They like predictability; it almost defines the temporal characteristic of a government.

But isn't Microsoft's OpenXML also such a standard? No. The format will be approved by ECMA International next week. At that time, the format will be more properly called Ecma Office Open XML, as Ecma, not Microsoft, will be maintaining it, though Microsoft will still be a major player and can decide not to use the format in its suite. It will then be proposed as another format to the ISO, which will have to evaluate whether the world needs more than one format for doing pretty much the same thing.

Besides the fact that ODF has been approved by the ISO and is already an established standard, what are the differences between OpenXML and ODF? Well, the obvious answer is not that one is free the other isn't—both will be free—but that implementing one is relatively easy; not so the other. It's relatively easy for developers to implement the ODF. The file description is relatively short, at around 700 pages, for starters, and the XML is laid out in a way that developers can readily understand it. What is more, there is an extensive body of documentation and work already in place to support developers and companies.

In contrast, OpenXML is difficult to implement. The XML description—the schema--runs on for thousands of pages and covers every detail. The result is that is, or at least has been, more difficult for developers to create implementations. This is not just ODF propaganda: As I mentioned earlier, there are many implementations already of ODF but so far only one of OpenXML: Microsoft's Office 2007, which is not quite officially out yet. The other suites will surely support import functionality , as many do now, but full functionality will be far more difficult to implement. (Of course, the situation may change as a result of the Novell deal with Microsoft.) By this time next year I fully expect there to be double the number of implementations of ODF. Of course, they won't all be complete office suites; that is not required.

ODF in Korea
I believe that Korea should use the ODF. It's something I believe that every country should do, but Korea presents a special case. I've already gone over the social and cultural argument. But let's briefly look at the economic one, which I've hinted at.

The ODF promotes the creation of a sustainable ecosystem of companies. These may offer complete implementations, like OpenOffice.org, or just save certain files in ODF. And these are the most boring examples I could think of. More exciting ones, that exploit the connectedness of Korea, easily come to mind. A vigorous ecosystem of small and large companies is easily imaginable. They would be satisfying not only the Korean market, selling Korean office products, but also the world market.

I appeal to Haansoft, for instance, to consider the benefits of ODF. One of the most widely respected companies, Haansoft could lead the world in deploying an implementation of the ODF both on the desktop and on the Web. It would be a Korean product: locally produced, globally significant. And other companies could create extensions, applications, whatever—and if they are proprietary, sell them; if open, leave them for others to use. The future of applications is shown by FIrefox and now OpenOffice.org: a core product enhanced by a multitude of free or proprietary extensions.

I invite Korean business and students to join us.

A final note: No other country is as connected or balanced on the edge of the future as Korea. And thus the intellectual wealth of no other country is as much held in the balance. Like the ocean, the Web has a memory but only for those things that float to the surface. Open standards such as HTM, float with the tide, carry messages to one and all and even to the future; they resist the passage of time. Everything else sinks.


Posted: Wed - December 13, 2006 at 06:56 PM          

©