OpenOffice.org, Year Five
[A longer version of this article was presented at OOoCon 2005 29 September.]
Let's begin with some facts and speculations.
OpenOffice.org probably has over 40 million users. Oh sure we can't count them. We can only estimate. But as of 10 days ago, we had at least so far 49 million binaries downloaded. That does not include binaries distributed via the Linux distributors, alternative servers, CDROMs, or even P2P. It also does not include direct hits to our servers. So the number of users may be far larger or even smaller. But who's counting? Probably only Microsoft, and judging from its market behaviour it sees us as counting a great deal.
But what makes OpenOffice.org count is the development and the community behind that development. And in the last year, as 2.0 has approached, the community have picked up the pace astonishingly. In the last month, we have been adding hundreds of new issues per week. Child work space (CWS) activity has vastly increased, as has general commit work. I expect the pace will even increase, as more groups and individuals work on OpenOffice.org, because of its increased popularity.
That enduser popularity is global, as is its development. Until recently, we were using a dual license scheme, coupling the LGPL with the SISSL. But with 2.0 RC1, we shall use only the LGPL, no longer the SISSL. That simple change is already having the expected results of strongly encouraging the publishing of modifications. Publishing is not the same as collaboration but it comes close; so close that bridging that little gap makes sense. Thus, KaiOffice, for one, has indicated a willingness to work with us. And I expect that there will be many more shortly. I call upon SOT Office, Magyar Office, RedOffice, and many others to join us: You have nothing to lose.
That's because OpenOffice.org satisfies a global hunger. We now have over 60 language projects, with more forming all the time. Governments, as well as corporations, are looking very seriously at OpenOffice.org 2.0. These governments include not only the small but very important US state of Massachusetts, but also Canadian provinces, as well as many other city, regional and national governments. For instance, the city of Vienna, which switched to OpenOffice.org; or the nation of Brazil, which can boast a strong and powerful government interest and user groups, as well as at least 2 million users.
These governments and corporations want 2.0 not just because it is cheap or free (support and service contracts are not free) but because the OpenDocument format is essentially democratic. It is something we have seen in the last six months: That the OpenDocument format for Office Applications (OpenDocument) OASIS Standard, is the compelling argument used by governments and enterprises. Why? Because its openness means that data created using it--your intellectual property--will remain not only yours but also legible as long as that standard is maintained, if not longer. Proprietary standards, on the other hand, are subject to the whims of the market. Not so open standards. If there is anything that is singularly important about OOo 2.0, then, it would have to be the for Office Applications (OpenDocument) OASIS Standard.
So: It is not just a question of whether OpenOffice.org has succeeded or not. It has. It is a question of where we want to be in the next five years and how we want to get there.
But first, a very brief and by no means representative examination of some key points in the last year.
First off, we have a new database component, "Base." It's a relational database and will satisfy the needs of those who want an Access equivalent. For more information, see the DBA project and the product description. Then, in the middle of the year, Kay Ramme announced the URE (Universal Runtime Environment). Both Base and the URE are in the 2.0 codeline. According to Kay's announcement, “The URE allows the usage of UNO independently of the OpenOffice.org productivity suite. UNO is OpenOffice.org's underlying component model, allowing language agnostic and remote transparent development of add-ins, components and applications.” The consequences are many; for more insight, see the UNO Development Kit (UDK) project.
In porting, Éric Bachard and his team have done extraordinary work making OpenOffice.org 2.0 run fast and robustly on X11. I have created this presentation and webpage using the latest build, which you can download from the Mac OS X download pages, under "Developer build". It took me five minutes to install, one minute to copy over my favourite fonts.
As Pavel Janík pointed out, in the last year, it has become easier to localise OpenOffice.org. This is good, as we will surely have at least 100 localisations by this time next year. We are still missing all the North American aboriginal tongues, for example, and if the Canadian government is serious about OpenOffice.org, they, and the First Nations, will surely see the benefit of having OpenOffice.org in Native tongues.
And who will do the localisations? I suppose one way of asking that is what is the community dynamic? How are the interests of the stakeholders accommodated? Simon Brouwer's example bears watching. For some time, now, he wrote me, "the Nederlands project has been "cooperating with Sun new UI and help files translation for OpenOffice.org 2.0/ StarOffice 8.0, sharing the work between our volunteers and a professional translation bureau. The combination of the translator's experience and our familiarity with the product was of great benefit to the quality of the translation. Most of the project is completed and we are entering the review phase."
But besides this process, which depends on StarOffice interest, the community has been steadily localising OpenOffice.org to their language. For many, the point is to provide the language needed: it's easier for users to operate the suite when they do not also have to learn English. Thus, Javier Sola has achieved considerable success in distributing his Khmer localised version of OpenOffice.org to Cambodians; and in so doing, he is helping to create an economy.
Each of these steps makes does the double work of creating new consumers and producers; open source is like that. But what is universally difficult is providing the environment for new developers. The Native-Language projects help greatly in this, but they are only one element in the equation.
Making it better for endusers.
With 2.0 the UI is improved and compatibility with MS Office is even better. But we all know that. What is perhaps less well known here are the vast efforts to address the needs of the tens of millions of endusers throughout the world who are using OpenOffice.org daily. They need documentation, support, and more. Users of OpenOffice.org can learn of free and professional support from our new enhanced support page (thanks to Matthew Waldrop). They can also read the extensive documentation, which is daily growing.
The Documentation Project, led by Scott Carr and Gerry Singleton, has now succeeded in producing volumes on OpenOffice.org 2.0, as well as migration guides. Documentation Project has also initiated a program putting together criteria for certification and professional support that independent companies may use.
Some of these guides and other material, like clipart and templates, are distributed for free in the CDROMs anyone can buy from the many, many CDROM distributors listed by Alex Fisher of the CDROM sub-project. Alex has recently enhanced the ISO and the web interface, making it a lot easier for visitors to the site and users of the product. In addition to the CDROM effort, Deepankar Datta is progressively making the P2P sub-project more complete and important for users. It relies primarily on BitTorrent, but other technologies are also coming into play. (I should add that Deepankar also, along with Erwin Tenhumberg, produces the monthly newsletter, which is read by tens of thousands.) His efforts complement the downloads offered by the large number of download mirrors, an archipelago of servers that daily handle awesome loads. My thanks to them for their unstinting generosity and contribution to OpenOffice.org.
What is left for the future?
3.0. It's time to think about doing it. There should be public discussions and debates about 3.0: It concerns everyone. The community should be involved in shaping the product, starting now.
All projects can involve themselves, and some will play greater roles than others. The Marketing project, led by Jacqueline McNally and JohnMcCreesh (Co-lead), have consistently kept OpenOffice.org in the media and on message. The challenge now is to show to the world the full capabilities of the product and the power of the community.
OOoCon 2005 showed what might come next. It showed that the technology is capacious enough to make for a suite that is radically extensible and agile, and a community that in one year's time will very likely double in size and intensity.