July 17, 2022
There was more activity than usual in forge federation, on various fronts such as diversity, funding, new projects, reports, UX etc. Code development was comparatively less active. Commitments were made to publish releases for forgefriends, F3, ForgeFlux, Starchart and forgefed within a year. The perspectives of funding are looking good but they are a minor concern because forge federation remains a predominantly volunteer driven effort.
ActivityPub support in Gitea
Support for ActivityPub was merged in the Gitea development branch and will be released as part of the upcoming 1.18 release. It is based on go-ap and implements user keypairs and HTTP signatures.
A draft PR is discussed to implement the federated following, starring, issues, forks, and pull requests backend.
Friendly Forge Format (F3)
The specifications are published from the JSON Schema under its own web site and updated via a continuous deployment pipeline triggered when changes are merged.
A number of discussions and pull requests related to the Gitea frontend helped understand how to approach the user interface changes required to implement federation. The first challenge was to figure out which framework to use for e2e testing. By chance, independently of federation, significant work is done to add support for Playwright.
Since the Gitea frontend is currently not tested (less than 1% coverage), the primary challenge will be to explain why tests should be an integral part of the development process.
The work to define a User eXperience started with a use case based on the 2021 User Research report that recommended to focus on federated issues. As more use cases are expected, a new forum category was created and a template defined (see Creating a federated issue for an example).
Learning how to use Penpot is part of the process and contact was made with the Penpot community. The goal is for UX to be an integral part of the development process. The learning curve for the developers currently involved in forgefriends is significant and it is necessary to find more people with UX expertise to move forward.
The first forgefriends online coworking session happened July 4th (four people). It was recorded and allowed for debates that led to a concrete proposal for action to resolve issues such as the persistence of users in a federated context.
More coworking sessions are scheduled for the weeks to come.
Codeberg & Gitea
A call for participation was sent and relayed by Codeberg to encourage people to work on the Gitea federation task list. Nobody showed up just yet but it was boosted dozens of times.
The MVP for the Gitea hosting provider Hostea was announced and its commitment to deploy forgefriends was made public.
The new forkedFrom property was added. It identifies the Repository which this Repository was created as a fork of…. More changes are expected as part of the Software Heritage work to define an ontology to use for indexed extrinsic origin metadata.
A videoconference was recorded to explain why and how the forgefed specifications are they way they are. Each aspect of the specifications is covered in great detail.
The vervis forge is back online and development is expected to continue so that it can be a reference implementation for forgefed.
Contact was made with NLnet to figure out if there is any funding left. However it turns out the contract expired long ago and was not extended.
Software and source codes college
A meeting with French Higher Education & Research organizations was organized to discuss forge federation. It may lead to User Research focused on how they use software forges.
State of the forge federation
The State of the Forge Federation: 2021 to 2023 report was published. It contains a retrospective of what happened in the past year and explains what is likely to happen in the next twelve months.
The retrospective is a compilation of the forgefriends monthly reports, with a broader scope as it includes projects that are essential to the progress towards forge federation even if they have no relationship with forgefriends.
The prospective part that tries to predict what is likely to happen in the year to come collects commitments from organizations and people. It tries to be conservative but should not be considered a roadmap: the work is mostly done by volunteers. A similar report will be written in June 2023 to compare the predictions with the reality.
In the past year, the forgefriends community engaged in a few experiments to improve gender or linguistic diversity, reached out to organizations and people who specialize in helping, participated in discussions. Overall it was not successful but it clarified that:
- Small organizations working on software or very focused on software are on their own
- Dedicating a small percentage of the time (5% for instance) is effective to keep going
A month ago an initiative was proposed to unite a number of small organizations that share most of the core values of forgefriends. This led to:
- A commitment to work on a best practice guide
- Monthly meetings for organizations to exchange on how the guide is applied in practice
- The creation of dedicated forum categories in French and English
An idea to improve diversity was proposed during the first meeting and will be acted upon in the forgefriends chatroom, if there is a consensus.
Most discussions are currently in French because the initiative was bootstrapped in the context of CHATONS which is a French speaking community for the most part. But it is not meant to be exclusive (that would be a contradiction) and forgefriends discussions will be conducted in English.
The last bits required to complete the renaming of fedeproxy into forgefriends were completed.
Transparency and funding
An inventory of funding organizations was conducted, updating those identified last year and a few new ones. Most of them are not currently a good match for forgefriends related work (European Cultural Foundation, Knight Foundation, Plaintext Group, etc.). Three remained: NLnet NGI Zero Entrust & NGI Assure, OTF and Prototype Fund.
A generic grant application focused on the Friendly Forge Format (abbreviated F3) was created. Three grant applications were derived from it, to be sent by the end of the month.
They are meant to get funding for one person only. However, other grant applications could be created to get funding for more people to work on forge federation. There are many aspects to cover in addition to F3, for instance User eXpercience, and it makes perfect sense for other grants to be sent. They won’t compete with each other if they are coordinated publicly.