Design organisations not licences

There’s an article in the Register this week about Bruce Perens’ “Post-Open Zero Cost Licence”.

In brief, Perens is aiming to try to fix one problem that some people have with open source. Specifically finding a way for maintainers to get paid to continue to develop software.

I’m being careful not to write “fix open source” as there are a range of different issues around open source and not all of them are about money. Perens doesn’t seem to be trying to fix those. He’s also clear than his “Post Open” approach is specifically not an open source model.

It’s also worth noting that there are other initiatives to help open source developers get paid. E.g. Tidelift and OpenCollective.

The Register article focuses on Perens draft Post-Open Zero Cost Licence (which, for transparency I’ve only skimmed) but the model being proposed is not really about licensing. Other licensing and contracts would be part of the overall system.

The intention is to have a not-for-profit (the “Post Open Administration”) which will be responsible for collecting fees from large commercial organisations (1% of revenue for orgs with over $5m annual revenue). Those fees will then be distributed to developers.

The license — as well as some additional agreements (“Paid Contract”, “Zero Cost Contract”, “Operating Agreement”) which are mentioned in the licence, but which don’t seem to exist yet — are intended to set up the appropriate contractual and licensing arrangements for that scheme to work.

As the Post Open website notes, there’s a lot that is still be decided. Like how this organisation be funded, how will developers be allocated a fair share of the collected fees, and what additional work will be required of developers in order to claim their share.

There seems to be a lot of unanswered questions, so this is fairly early work. But I was left thinking about a lot of other questions around how the such as:

• what will be the governance structure of this new organisation?
• will these contracts and agreements, and the ability for the Administration to enforce them, be designed to work internationally?
• what funding is required in order to complete the work, launch the organisation and ensure it is sustainable over the long term?
• will the organisation be funded separately through other means, or will it be taking a share in the raised revenue in order to cover its costs? If so, then how will that be made transparent?
• how will the organisation build up the necessary reserves in order to be able to enforce its agreements?
• how will developers be able to trust that the organisation is being run for their benefit and, over the long term, that it continues to do so?
• who, exactly, is receiving the money being collected? How will it get paid to maintainers and contributors?
• are there other factors that might mean a maintainer or their project cannot participate? Will people be excluded and if so, how? (Lack of anonymity might be one)
• if, as Perens suggests, the Post Open Administration might reduce the need for large organisations to have their compliance departments then doesn’t that mean it needs to take some of that burden, e.g. confirming that projects and their dependencies are or aren’t “Post Open”?
• how might a move to embrace this new scheme impact the open source landscape? will there be a lot of forking of existing projects just before they decide to go “Post Open”?
• will there be additional burdens on developers and projects that don’t embrace this, e.g. to prove that their dependencies are not “post open”?
• if the Administration is split into three organisations (one for handling money, one for handling licensing and one for compliance) then doesn’t that just make all of the above even more complicated?

Perens notes in the Register article that developers will need to trust the Administration. That seems to be the crucial factor here. And not whether its possible to draft a license that sketches out the intended ecosystem.

I think you have to start by designing the organisation and working through all of the above hard questions rather than the drafting of licences and agreements. Otherwise how do you know what you need to draft?

Start with the organisational model not the licence documents.

I’m guessing that Tidelift and others have already worked through these kinds of issues already. My final thought was, why not just put more weight behind those initiatives, rather than starting something else?