eZ Systems, the company I have been working since end of 2005, has made two significant decisions in the last 6 months. I consider those interesting and worth to look at now, with an eye that has more perspective, distance and less involvement.
I am refering to, on one hand, eZ Systems transferring the eZ Components project to the Apache Foundation (see http://incubator.apache.org/zetacomponents/) and on the other hand introducing eZ Publish Community project simultaneously with a commercial package based on eZ Publish called eZ Publish Enterprise.
Does that reflect a good move to a succesfull Open Source development model ? I think so. Of course it is far from enough to achieve that goal, and certainly not enough also to shape a succesfull business model; many other steps are on the road to success ! But I still think those two moves are going in the good direction. I’ll try to give some hints on this here.
First I would like to highlight the fact that I speak here of a succesfull “development model” and not of a succesfull “business model”. The two things are significantly different, though extremely closely related. In the last months, many folks from the Open Source community, including execs from Red Hat, analysts from 451 or other emerging vendors like Nuxeo have insisted on this important thought: You should consider Open Source as a development model and not a business model if you want it to deliver value. I truely believe this is the right angle to look at the Open Source phenomenon: a powerful and unequaled development model for software.
The first decision I mentionned is about transfering eZ Components to a new project hosted by the Apache Software Foundation (and including a renaming in Zeta Components). The move was definitelly the one to do in my opinion and was an important one. Since the beginning, the eZ Components project was thought as a low level PHP library that would help in building the next generation of eZ Publish software but that would also spread in the whole PHP community with the goal to turn it in a very succesful high quality PHP library/framework widely used on the market, providing eZ Publish with top quality low level components that are also adopted by a wide range of developers and that contribute in making eZ a very good choice when it comes to extandability ! In my opinion, It definitelly make more sens for that to develop this project as an external, independant and community driven one (that would then not suffer from eZ’s unforeseen events) than in an in-house project where eZ as a company would control all the resources and own all the intellectual property. I even think now it was a mistake to start in-house, and from the beginning this project chould have been an external one, but this is not a major mistake at all and the project can catch up on that one, I am sure.
So yes, with some more distance and perspective, I welcome the Zeta Components considering it is the best choice for eZ Systems to try to leverage its initial investment and collaborate with this project to participate in a shared open source development effort.
Another interesting side topic raised here is about joining an existing foundation such as the Apache Foundation (the other choice would have been for instance to make a more independant project). I strongly believe that in that specific case of the Zeta Components, an existing foundation makes a lot of sens. In the past, the Apache Foundation, the Eclipse Foundation or the Mozilla foundation have all demonstrated they were bringing a lot of value and capable to bridge community driven open source projects with business. I see here a strong potential fit ! Zeta Components, by definition, could benefit to many other business, not competing with eZ, and joining forces can only be done in joining an existing organisation such as the one mentionned
The second decision I touched is the introduction of the eZ Publish community project on the side of the eZ Publish Enterprise package. In short, eZ Publish Community project is trying to engage the whole community, enabling and easing contribution to the develoment of the core of eZ Publish itself (and not anymore only on extensions). In less than a year the eZ Publish community has been restarted. There is now again a very good momentum. The community, while being very professional, is growing and developing significantly with high number of members, stronger activity and very significant number of downloads and installations for a commercial open source software. A governance project is baking. The work here of Nicolas Pastorino and the core community team has clearly demonstrated it is worth engaging the community on product development and activating ways of contribution to eZ Publish ! Well managed, this should provide eZ Systems and all the eZ Ecosystem with a significantly increased engineering capacity, for the best of all the ecosystem members.
Combining those two decisions shows a mixed way to deal with Open Source development, that can be hard to understand. On one side, eZ tries to leverage a strong and large existing independant open source software organisation and on the other, to develop its own community with its own tools and infrastructure instead of rallying to an existing organization. Is this the good approach ? Isn’it a contradiction ? Should eZ think about contributing the whole eZ Publish community project to an existing open source software foundation such as the ASF ? I am still sometime wondering about that last idea, but clearly, eZ Publish being the core CMS sofware, and Zeta Components being only a low level infrastructure component, the 2 different choices are making a lot of sens: on one side we have a development effort that can not be shared with competing projects, on the other, a huge interest to mutualize and share the development effort with other users.
Interested in the topic of Open Source product development (btw on that topic, I can only recommend that blog post of Ian Skerrett on successful open development communitiesy), I really think eZ might have here a really nice combined approach that might deliver all its promises if well implemented ! I wish the best of luck to the eZ ecosystem, the eZ Publish community and the Zeta components projects in pursuing in those directions, I strongly hope they will demonstrate this mixed approach can be valuable.
But of course, this is a very little part of the endeavour, and thinking those two decisions are enough to reach a successful Open Source development model would be a huge mistake. The main and most difficult tasks will be in smartly coordinating a community effort and an internal engineering effort, in implementing the right marketing approach to the community projects, both of them, and developing at the same time a sales and marketing effort on the commercial side of the business that doesn’t contradict but leverage the value of the community effort. I again think it is all here about perception, packaging and delivering value on top of Open Source software.
To be continued !