Assignment 2 - History, infrastructure, and demographics

ASSIGNMENT 2 - the history, infrastructure, and demographics of Hypothes.is

The objective of this assignment

The objective of this assignment is to describe the community of Hypothes.is. To address this issue the project's history, infrastructure and demographics will be described. The information here has been aggregated through Hypothes.is main communication channels which were in detail described in the last assignment, they consist of awebsite, Github, mailing list archive, forum and lurking in the IRC chat. It is important to point out that this is by no means an official statement by the community or core-team of the project, but solely based on my subjective experience of working on Hypothes.is. As I gain greater insight into the project this view is likely to change, so this should be seen as a snapshot after having worked with them for four months.

Why is the project open?

There are several reasons why the project is open source, aggregated they should make it most likely for Hypothes.is to succeed as a project. The project is build on 12 principles which worship the very core of open source and collaboration. By making the code freely available and by being transparent in the development decisions Hypothes.is not only attracts developers, but also known advisers who support the project. These two groups have a self-reinforcing power and the core-team has been able to accumulate a notably number of collaborators.

Another aspect is mentioned in the FAQ, where Hypothes.is describes that the project needs a critical mass of users for making the tool really useful. They have developed a three point plan to maximize the user base.

  1. Create a tool that’s useful even before the larger community arrives.
  2. Engage pools of domain experts that are willing to help jump start quality analysis.
  3. Drive sufficient traffic to create critical mass.

The openness of the project is an important factor for all of the three mentioned points. It can create a noticeable footprint in the digital community even before the release and thereby attract more collaborators. Generally speaking is Metcalfe's law still true, more users attract more users. The openness also helps by attracting domain experts which as described in point two.

How big is the community?

There is a fully fleshed out community around the project although it is not in public beta yet. There are several groups in the community which I will present in condensed form.

Right now the community is mainly based around the core-team since the project tool is not released yet. The core-team is spread on several continents, which easily can be observed in the IRC channel at the meetings. I expect the community to change dramatically when all the preregistered users get access to the tool. Then the main group will be normal users of the tool and the forum and the wiki will become more important, as they are more convenient to use for normal users than IRC or Github.

What kind of product is it and how is it licensed?

Hypothes.is is licensed under a 2-Clause BSD License, sometimes referred to as the “Simplified BSD License” or the “FreeBSD License”.

What sort of infrastructure does it use?

I have mostly been lurking the community through the communication channels, while talking with Jake and the core-team in person. I have made a few comments and written documentation, but most of my communication was more informal. However, this does not mean that the channels are not actively used.

On a daily basis the project is mainly run by developers and their communication tools are most active. This means that both the IRC channel and the issue-tracker are the tools of choice to get in contact with the community. Although the project has formal directors the daily business is handled by who-ever feels responsible or who can add to the discussion. Once a week there is an IRC meeting where everyone who is interested can attend. On the meeting the current progress and the plan for the coming weeks is discussed. The vision is written down in the roadmap. The document is served through git, so all have access to it. Changes are discussed through the mailing list before being added, removed or altered.

Reflections

I think the decision of making a project open or not is based on the teams preferences, since both can lead to success. It looks like the choice of making Hypothes.is open has helped the project to generate the needed hype around it. It is also noteworthy that the project has been able to gain remarkable monetary support, both by private persons through Kickstarter and from foundations. The plan is to raise three million dollars in the coming years.

The project was founded in Nov. 2011 so it has been in development for some time. Although the tool is still not in open beta they have been able to gather a stable core-team and many advisers. This itself is quite an achievement. The backside of the active core-team is, that there are relatively high entry barriers. Nearly all commits are made by the core-team, and although participation is desired and promoted through the website, the established community can feel overwhelming. The more there is 'out there' the harder it is to join.

Conclusion

Hypothes.is is a remarkable open source project. It has established a stable community around it's project, which is based on principles around openness and collaboration. It seems like there is also rich interest in funding the project, so it is possible to create an interesting product. Time will tell if it can live up to its expectations, but based on the existing community the odds are not bad.

Retrospectively, I am still amazed by Hypothes.is' community. In the few months I have been contributing to the project it has gained even more media coverage. But not everything is as positive. There has not been a single commit from a person from outside of the core-team, although they actively promote it on their website. To be honest I don't know what they could improve to attract more contributors. I think the main reason is due to the complexity and high rate of involvement of the core-team which makes it hard for others to join. However number of contributors can't easily be translated to overall success. Maybe will a highly involved core-team be the best solution for Hypothes.is. Right now they seem to do pretty well.