2007-04-13

Toward Informatic Autonomy


Toward Informatic Autonomy
fisl 8.0, Porto Alegre, Brazil, April 2007

Louis Suárez-Potts, PhD
Copyright (c) Louis Suárez-Potts
Creative Commons Attribution 2.5 License





Introduction
I want to talk about where OpenOffice.org is and where we are going; to give you a report on the project and open an invitation. But I also want to speak to the logic of free software collaboration and reconsider the stakes at play. And I want to listen to what you have to say. I don't want to just say what I think you should do—we've all heard lectures like that and once is enough.

For in many ways, the future not only of OpenOffice.org but of a lot of free software lies with you, working as leaders in a partnership with other nations. It lies with meeting the challenges of coordination, education, development, distribution; challenges of poverty and infrastructure and bureaucratic and political resistance; all challenges that you are meeting now. The story of intellectual property and of the world's wealth is a story that we are all now rewriting. No one nation or region can any longer rightly claim dominance. Is it too much to assert that empire is in twilight? I don't think so.

But this twilight period is fragile—countering hegemonic power always is, as that power has deep resources and vast momentum; it can, for instance, buy its way back. But we know all this, and we know, too, that resistance is infinitely creative and resourceful, but only if able to use tactics and techniques that allow for the free exchange of ideas.

Resistance to hegemonic power means doing what benefits your community—your local community most. For us, it means in this case using free software and it means building free software. And that is my cue to start discussing OpenOffice.org more directly.

The state of the project
OpenOffice.org has never been strong or better, and is growing daily, as a project, a product, a community, throughout the world. We owe this growth—this world changing growth—to you: those who, like Carlos Dantas de Menezes, who created the enormously useful grammar checker CoGrOO for Brazilian Portuguese, and the Brazilian OpenOffice.org team, who have localized the application and distributed it, and ensured that millions here in Brazil, in the favelas in the cities, everywhere, can use it in their language, for free.

This achievement, of people working together all over the world, with local efforts collaborating with international ones, has enormous ramifications for how we live today and tomorrow. OpenOffice.org gives users the power to produce not just office documents, and thus the power to enter the 21st century as equals—but just about anything else: it's a productivity suite with no horizons. Because the application uses the OpenDocument format, or the ODF, as it is more popularly known, as it is default file format, and because the ODF is an open standard, maintained by an international consortium and not by any one company, and because it is infinitely adaptable, flexible, and powerful, for all these reasons, OpenOffice.org is a tool that gives users extraordinary power.

It gives you the power to produce a vast range of work that you can share or save or edit using other applications, without the anxiety that your friends must have the same application or that you must renew your own end user license with the company that created the application you used. The ODF, the file format, has been implemented by numerous applications already, meaning that a file crated using OpenOffice.org or StarOffice or IBM Workplace or KOffice can be edited using any other implementation. With the ODF, with OpenOffice.org, there is no vendor lockin. And because the ODF can be implemented by both proprietary and free software, it gives corporations and governments relief. They can continue with purchasing patterns established in the last twenty years, where they buy support and training and the application all from the same vendor, while still benefiting from the virtues of free software—flexibility as well as little to no cost. This is why city, provincial, and federal governments in Latin America, in Europe, Asia, Africa; in Australia and Canada, and even the US, have been mandating the ODF. It gives power to people in a way that is friendly to commerce at home and abroad.
It gives you the power to truly own your work. But this power is diluted if users must create their work using the language of empire, English, and not their own; unnecessary dependencies are imposed. So: OpenOffice.org in your language is rightly seen as a crucial step toward informatic autonomy and economic power. I mean by this term, "informatic autonomy" being able to control the documents, files, or more generally, the intellectual property, you create, as well as the tools that create them, like OpenOffice.org. Its opposition is naturally, "informatic dependency," which means being dependent on the vendor for your intellectual property. With informatic dependency, for all the work you do, you are primarily a consumer.

The disadvantages of being only consumers of the tools for participating fully in the informatic economy, the economy of the 21st century and the key to other development, are enough to give one pause. The most obvious risk is losing control over one's intellectual property. But the risks pile up. With little control over the means of producing intellectual property, one must depend on those who have control—the vendor; and inevitably, Brazil, Latin America—and also India and other developing state—could become informatic backwaters, clients to extraordinarily rich and powerful vendors whose only real concern is their profit. So much for freedom.

No one wants this future. We all want a Brazil, and a world, that has taken FOSS seriously, as a consumer and producer. But it demands formidable investment, and not just by volunteers, but also by government and businesses. It requires government to think not just a year, two, or even five but decades in advance and to make investments in education and training that may have no clear immediate benefit but which will pay off with a professional cadre able to act on Brazil's promise as a leader in Latin America and the world.


Breaking dependencies
When I was in India last January, I presented to students at several institutions on how to participate in OpenOffice.org and also how to start an open-source project. (I'd be delighted to make my slides on the topic available.) I also presented to government and business executives on OpenOffice.org. Although the actual particulars and emphasis of the messages differed according to audience—I wasn't about to speak on how to develop OpenOffice.org to the President's IT policy people—the basic message stayed the same::

  • That OpenOffice.org can address user needs with greater security, ease of use, and of course enormous cost savings; and
  • That sustaining OpenOffice.org is extraordinarily important.

India, a land of 1.1 billion people and less than half the size of Brazil, has a large technologically sophisticated class. Yet for all the technological training I saw, and for all the emphasis on grasping the 21st century by the horns, the default approach to free software was as a user, not producer. "Free" software was primarily interesting because it cost nothing, not because it gives the user the ability, even the responsibility, to be a producer, as well. India is not alone. For most of the world, free software is just a free commodity. The logic of informatic dependency has not been fully broken, and the power that free software represents has not been fully appreciated.

This situation is changing, but from what I could see in India, many of the efforts are still young and in need of coordination. Important federal bureaucrats remain unpersuaded that free software is the answer, let alone a logical place for scarce resources. Many still believe that it makes more sense to pay Microsoft yet more millions—millions!--than to build a local economy based on knowledge and skill that can not only sustain itself but expand. Yet I am optimistic, for at the heart of the desire to work on free software is the realization that not only is the freedom to think and act creatively and collaboratively vastly enjoyable and rewarding but that what has been produced, the software itself, works, and beautifully.

Look at Firefox or at OpenOffice.org: these are but two desktop applications that work better than their proprietary equivalents, have fewer security holes, are extensible, and are free—to be worked on, to be customized, to be changed, to be used.

I used to use Safari. But Safari is not as standardized as Firefox and it does not read Web files as well as Firefox nor save them in the now standard way, and it also looks boring, compared to Firefox. With Firefox, I can change its look, I can add extensions, I can do things with it that are different. I can even add an extension that allows me to read OpenDocument format files. As a consequence, I no longer use Safari.

Firefox is persuasive because it is better, not because it is free, for Safari is also free, as it is included in Mac OS X. OpenOffice.org presents a similar argument. For the vast majority of users, OpenOffice.org is simply better. They can alter its appearance, they can add an ever-growing number of extensions, they can integrate it with other applications, and they can work more efficiently knowing that if they choose to save their documents in the ODF format, they will stay readable as digital documents far longer than any proprietary format. And with somewhere more than 50 million users throughout the world, the community of users is enormous and possessed of an formidable momentum. I fully expect and would make it a challenge, to have 75 million users by the end of this year. That would represent a very sizable portion of the estimated overall office suite market. And those users who have chosen to use OpenOffice.org would have done so in all likelihood because it is better.

Knowing that there is such a market and knowing that people choose software because it works better than the competitor's and is freely available, should be enough for governments and businesses to make rational decisions. Yet many politicians and bureaucrats persist in choosing Microsoft. Why? Well, for one, when threatened with market loss, Microsoft spends money—a lot of it. They lower the price (and alter the product) to accommodate the poorer countries. They spend money on costly localization efforts; they spend hundreds of millions of dollars on advertising campaigns. And, besides, many are habituated to Office. As with cigarettes, or other addictive drugs, which we all know are really bad for us, rationality does not come into play when habit and money speak.

It's time to break that habit and invest where it matters: on education, on projects that give freedom, as well as further the local IT ecosystem, and refrain from throwing good money after bad, or on those that wrap yet more chains around us. What we all want is informatic autonomy, not dependence.


Achieving informatic autonomy
How do we achieve this autonomy? Put another way, how best to invest? I can't answer that, of course, though I do like making suggestions—it comes from being trained as an academic--and later this next month, in Montreal, Moscow and Beijing, I'll be doing a lot of suggesting.

Also, here. Let's look at the two areas: use and development. For large-scale users, such as businesses and governments, the primary obstacle, besides the one having to do with negotiating with a multitude of open-source projects and dealing with free software licenses, like the GPL, has to do with finding support and training. It's a myth that there is no support for free software; there is, and it's often offered by huge, multinational companies, as well as by small, local businesses.

Okay, I often hear, so there is support. But our MS Office macros don't work on OpenOffice.org. And there are other minor but infinitely infuriating problems, like subtle changes in bullet points, which make many of our migrated presentations look silly.

I answer: With OpenOffice.org 2.2, which we released just a couple of weeks ago, there is not only greater compatibility with existing MS Office files but increased support for just those things that you need. We are also working on improving macro support. Still, there will surely be problems—working with the black hole of proprietary software does not make it easy. But it is still cheaper to hire a developer or two to fix those minor problems ad hoc than to renew a multi-year license with Microsoft.

But what about Office 2007? It uses a different file format, OOXML, and is pretending to be an open standard. Certainly, Microsoft would love it if it became one like the ODF, which is the ISO standard, though that seems doubtful. The format's schema—how it is defined—is forbiddingly complex and excruciatingly detailed (and at over 6,000 pages, long), contains many contradictions and has seemingly succeeded in making it nearly impossible for any other application to fully implement it. But the big question is, How well does OpenOffice.org work with Office 2007? Let's assume that some people, somewhere, are actually using it—a big assumption, for if they create files in it, their colleagues will have to have it, too, at least in order to view the full file, and from what I read and hear, not nearly as many are adopting it as had been imagined by Microsoft.

At the moment, there are various translators that allow users to work with some Office 2007 files, and soon there will be a native translator that will give OpenOffice.org users superior capabilities. In short, one can have a hybrid environment, meaning, in fact, that there are even fewer reasons not to migrate to OpenOffice.org. In such a hybrid environment, some could be using MS Office, others OpenOffice.org and its derivatives—or other applications that implement the ODF.

Yet I see this period as transitional. What is important is that files created and distributed be freely available to all, and not just to a few who can afford the extraordinary costs of some proprietary software, and that the tools for content production be open.
Besides the political reasons I sketched above, there is also the pragmatic reason openness is good: better products. Remember the example of Firefox vs. Safari? Well, OpenOffice.org is, as I mentioned, extensible, too. And even more: it can move beyond the limits of its legacy, in a way that is close to impossible for others.

I mean by this that the office suite embodies its legacy of being designed for computers isolated from each other and needing integrated applications utilizing a common file format and interface. It is a continuation of the days before the Internet, before deep connectivity, before Google, before mashups. It is a holdover from the 20th century.

And it's still useful. Most of us do not have huge Internet pipes and infinite bandwidth, and for complex documents, such as certain spreadsheets, where each cell may house a formula, having a complete and feature-rich suite on your desktop is necessary.
But for the other elements, the current exciting state of Internet applications allows us to create what's effectively our own suites ad hoc. What is crucial here, of course, is again, the file format (interfaces have more or less converged). It counts for nothing if you create documents limited by their format. That's why OpenOffice.org is even more relevant. It gives you the power to take the ODF-generating element and move beyond the suite. We're serious about this, and have created the ODF Toolkit project (odftoolkit.openoffice.org), whose express aim is to disentangle ODF generating code from OpenOffice.org. Any logical application could utilize this free code.
An obvious example: Say you are writing an email. This happens to me a lot. At some point, the email becomes an essay, or a blog. Now, if you are using an HTML capable client, you can add HTML formatting. But it's not clear you could then cut and paste it into your application and expect formatting to be accurately preserved. In fact, it would probably look awful. But imagine now that you are using a client that can save the file as an ODF document. Formatting is preserved; what is more it can be further edited, as any ODF document, in any number of applications. I find these possibilities exciting. For by using a common file format, we eliminate the needless obstacles that 20th century technology threw at us. We return, in a way, to the pure simplicity of pencil and paper, where we could focus on creating.
OpenOffice.org gives you an unencumbered future. But getting there requires having the engineers who know how to code and work in a free-source project. It's not a question of "finding" them and then hiring them. I wish it were that simple. It's more a question of educating them.

Autonomy and education
How do we do this? How do we educate all the developers we need?
The basic logic is to open secondary and college curricula to open-source projects and to involve those projects with schools and colleges and other educational institutions. A couple of years ago, Sophie Gautier, lead of the Francophone Project on OpenOffice.org, and I created an Education project. The point was to coordinate student and educational activity related to OpenOffice.org, and it derived from work she was doing and I had been doing. I had even created a process that allowed professors and students to collaborate on OpenOffice.org and get grades. Then, around that time, Google initiated the Google Summer of Code, which prompted the project to refine its to-dos, so as to make them easier for students to do within a semester or summer.

Coincidentally, Mozilla, with greater resources than we could pull together, was doing something similar, and working with the US university Oregon State University, and the Canadian college, Seneca College, based in Toronto, where I, and Mike Shaver of Mozilla, happen to live. I had not known this, but many of the extensions we use with Firefox have been qualified or even created by the students at these colleges, who have taken to working on the extensions with enthusiasm and joy. With reason: how often did we have the opportunity, as students, to do things that really mattered? If your experience was anything like mine, most of the work I did as a student was duplicating what others had done, so that I could learn a lesson. But sometimes I think I mainly learned boredom. Not so these students, for there is nothing boring about working on open source code that millions, tens of millions will use, enjoy, and which exists not in a environment of commercial distance, but in (and for) a community.

OpenOffice.org is now also working with these same colleges and several others. As our extensions project is more clearly defined, more students and professors will participate, that is certain. But it's not just extensions. Students at both colleges, but especially at Seneca, which is leading in integrating curricula with open-source projects, students are also doing core coding. They are not being exploited; hardly. They are rather learning how to code in C++, how to solve problems in real environments, and to solve them in ways that others can utilize their solutions; they are learning how to architect complex code; and how to work with others. They are learning invaluable things that will get them jobs.

The Indian state of Andhra Pradesh, where the tech-heavy city of Hyderabad is located, has understood this, and has arranged coursework with IBM to further educate students in free software. The point is not just to get the students jobs, but more important, to shift teaching coding so that it includes free software projects like Mozilla, OpenOffice.org and many others. Free software is clearly the way of the future, and it's also personally rewarding. Indeed, the lectures I presented there and elsewhere underscored the pleasures and payoffs of working with others on real subjects with actual importance.

But it's glib and ignorant to speak of including OpenOffice.org and other projects' code without fully understanding the educational situation in Brazil, so my points are to be taken more generally, that a new model of teaching code is possible with the involvement of free-software projects. Even with that caveat, there are considerable resource and logistical issues that need to be addressed.
Naturally, I cannot address them here. I can say that there are ways of articulating the relation among students, professors, and the free software project community so that no one is shocked or intimidated. And we have learned that for those students who wish to engage the community directly, they benefit by having a mentor who can show her or him what to do and how to do it.

OpenOffice.org has many community members who are willing to be mentors, and our project is justly famed for providing all community members with a friendly and constructive environment. Students who have worked in the project on code have described their experience as the best in their life—and I'm not making that up.

There are no obstacles to starting this year. We could start with extensions, which are much easier to create, as OpenOffice.org is written in C++. Or, if the students are sufficiently advanced, and I'm sure that many here are, we have listed a variety of interesting and important tasks. The issue is shaping curricula so that those tasks can be worked on, so that the work done indicates that the student has truly learned what she was supposed to learn. And I have no doubt we can help there.

The payoff for such a collaboration is huge. It is not just making a better productivity suite for all nor transforming what even counts as a suite, so that it goes beyond its legacy limitations—and even beyond Google Docs—but in giving Brazil and Latin America and elsewhere informatic autonomy and shaping them as the leaders of 21st century informatics.

Where to go for information
Below are some useful links that should help you get started. I've also included my contact information and am happy to receive messages in English, Spanish and even Brazilian Portuguese.


Contact information
Louis Suárez-Potts
Location: Toronto
Posted by at