As we said in an end-of-year post highlighting our work supporting free software development and infrastructure, the Free Software Foundation (FSF) is planning to launch a public code hosting and collaboration platform ("forge"), to launch in 2020. Members of the FSF tech team are currently reviewing ethical Web-based software that helps teams work on their projects, with features like merge requests, bug tracking, and other common tools.

The new site will complement the current GNU and non-GNU Savannah servers, which we will continue to support and improve, in collaboration with their awesome volunteer team. (By the way, if you want to volunteer, please email savannah-hackers-public@gnu.org with a note about your interest!)

Infrastructure is very important for free software, and it's unfortunate that so much free software development currently relies on sites that don't publish their source code, and require or encourage the use of proprietary software. Our GNU ethical repository criteria aim to set a high standard for free software code hosting, and we hope to meet that with our new forge.

We plan on contributing improvements upstream for the new forge software we choose, to boost its score on those criteria. Our tech team is small for the size of the network we maintain, and we don't have any full-time developers who work for the FSF, so we are limited in the amount of time we can spend on the software we choose. We'll communicate with the upstream developers to request improvements and help clarify any questions related to the ethical repository criteria.

So far, we have been researching a list of candidate programs, and analyzing them in terms of ethical and practical criteria. Some of the software candidates we're looking at were found on the Free Software Directory. We aim to initially reach a B rating on the GNU ethical repository criteria, and then to work towards reaching an A rating after we launch. Reaching a B will require LibreJS support, no third party tracking, proper license information, and more. We also came up with a list of practical criteria, which includes two-factor authentication (2FA), high performance, being well supported upstream, and other common forge features.

We are filtering out systems that are targeted toward single organizations or companies, because we want users to be able to sign up and create their own repos on our site. If you're looking for a system to handle your organization's source code management needs, there are some fully free options out there for you, including Kallithea, Allura, and Phabricator.

We also hope that in the future we'll be able to see decentralized, federated collaboration platforms that meet most needs. We will continue to be interested in that direction, but we think the need for this freedom-respecting forge is time sensitive, so we're going to do it with the free software we have available right now. Allowing issues and other data to be imported and exported is a feature that we want in our new forge, because that will at least ensure that users can move to another instance of the same platform.

We are tracking our ongoing analysis on the LibrePlanet wiki, and will continue updating the page with information pertaining to our research about free software for our upcoming forge.

The project will operate with a high level of transparency: we will publish the source code that runs on the server and document how we run the system, and we welcome volunteers to help guide and improve the project. Reach out to us at the LibrePlanet developers mailing list if you're interested in participating.

Up next for the FSF tech team is to do more research about systems that have met our initial requirements, in order to find the best options available. Once we know what we're interested in, we'll start trying them out and performing more extensive tests.

Stay tuned to hear from us about the software stack we end up choosing, and for our site launch announcement!