Response from TWIKI.NET Interim CEO (Score 1)

I wanted to respond to the recent postings regarding TWIKI.NET and the TWiki.org project.

On Monday October 27, we posted a communication regarding a "relaunch" of the TWiki.org community to the TWiki.org website.

Please see: http://www.twiki.org/cgi-bin/view/Codev/RelaunchTWikiOrgProject

The key points in that communication are:

- exposition of a new governance model (an Ubuntu-style model)
- expansion of the charter of the project to encompass open standards around enterprise Web 2.0 collaboration (not just the wiki)
- greater focus on enterprise scaleability and integration standards

I want to clarify a few things to start off:

- We invite participation in the project. We took the actions that we did in order to increase the long run relevance of the project and increase the number of developers and users. We are not naïve; we certainly recognized that it would create some turmoil in the short run and that many key developers would choose to fork.

- Anyone is free to join this project. In this sense, no one has been "locked out". What we have done is ask that anyone who registers and contributes to the site adhere to a new code of conduct which very clearly specifies the new governance model. And it is important to note that the governance model isn't democratic (more on that later).

- Both TWiki.org and TWIKI.NET are fully compliant with the GPL, and furthermore, the .org is committed to an exclusively open source approach. Under the prior governance model, there were examples of closed source object modules on the site, such as various installers. We didn't think that was right.

I also want to provide a bit of background as to how I see the open source wiki and collaboration space today as a precursor to why we went with the Ubuntu-style model and adopted what is admittedly a pretty radical move.

I would characterize all of the open source projects in this space as being relatively small efforts, and all of them appear to be "thinking small" to me. That is to say they are fairly narrowly focused on some aspect of collaboration (e.g., originally TWiki was focused primarily on the wiki), or for some specific purpose (e.g., MediaWiki's primary purpose is to support Wikipedia). There simply isn't any large effort that is focused on setting open standards and providing an end-to-end open source solution for the full range of enterprise collaboration needs.

Is anyone thinking about how to create a framework and set of APIs to make it easy for arbitrary blog engines or social networking engines to attach into an enterprise collaboration framework? Is anyone thinking about how to standardize data about people, so that it can be shared between different collaboration apps? AD/LDAP approaches are all aimed at authentication and access control, not at capturing richer information about people. Is anyone thinking about how to augment the OpenSocial API to make it more relevant for enterprises and allow people to manage their social graphs between their consumer and enterprise lives, and implement clear privacy rules between the two?

The answer to all of these questions is no. I'm pretty sure the only group of people that are thinking about these problems, while using an open source and open standard approach, are the people at TWIKI.NET, some university researchers and a small number corporations that we have had discussions with.

Our vision of what we want to do for open source enterprise collaboration is pretty broad. It's going to require a lot of resource to get there. Some of that resource will be provided by open source developers, but realistically it will most likely require a lot of commercially focused effort as well. Most of the successful open source efforts out there have some closely aligned commercial entity. When I was at Cygnus Solutions from 1996-2000, roughly 80% of the development in gcc/egcs was directly funded by Cygnus Solutions (e.g., was contributed by paid employees of Cygnus Solutions).

Now turning to some background on the governance question. The reality is that governance has been a contentious issue in the TWiki.org community for a very long time. The history of all of that is available for review on the .org website. And the reality is the folks who are upset about our actions are the very same folks that have been planning a fork for quite some time. As it turns out, the fork project was registered on SourceForge on September 2, 2008, and the fork URL was registered on September 29th, 2008. Furthermore, when I showed up to the Berlin Developer Summit on September 4, I was informed that the community was ready to announce a fork within 24 hours unless we acceded to a new governance model which implied a very watered down approach to the role of the project founder as well as to his lawful rights to the project brand and trademark.

For a number of weeks, we considered acceding to these demands to keep the community together (all of that history is also available on the site).

However, two weeks ago, after much consideration, we decided that this approach would not be healthy for the community at large, consultants, and TWIKI.NET. People may accuse us of being disingenuous when we say that we have the community's interests at heart, but we actually do. It will be a different community moving forward though. Basically, under the old model, TWiki.org was dead. They just didn't know it yet. What I'm saying is that all of the open source efforts in the space are underfunded and lack a broader vision, including the old TWiki.org community.

There were many issues with the community that were impeding progress: infighting on the governance question; an excessively rigorous approach to testing that actually inhibited less experienced developers from contributing code; lack of a primary commercial sponsor (by the way, I would like to publicly thank Sun for its contributions of hardware to the project); and a tendency for some of the key contributors to withhold code updates in their private repositories and not contribute them back to the project. To be clear, I am not accusing anyone of violating the GPL, simply of pursuing their own commercial interests in taking their sweet time to post new code to the project.

We looked at the situation and we basically said that a headless monster wasn't going to fly. We could've continued to negotiate with the folks who wanted to turn the project a different direction (e.g., kick out the founder), but we decided that we wouldn't win under that scenario, and we don't think the community will either. What is needed is clarity of purpose and resources. That simply could not be achieved under the Debian-style model that many community developers wanted to move to.

So we chose a different strategy, one that would give the project Founder (Peter Thoeny), an ability to set a broader agenda. And one that would give TWIKI.NET a fighting chance to succeed as a commercial entity. We continue to think that both are critical for the success of the open source project. I invite people to review the Code of Conduct: it's really quite simple and straightforward, thanks to the work of the Ubuntu project from which it is derived. And of course all of the code on the TWiki.org website is GPL'd and freely available.

Sincerely,
Tom Barton
Interim CEO, TWIKI.NET