Today J. A. from NLnet sent a mail to invite fedeproxy to participate in the upcoming webinar they are organizing (June 21st, 2021 9:30am). It would be a 5 minutes talk on “on data interoperability, possibly related to Linked Data”. The webinar will include representative of projects funded by NLnet DAPSI etc. (under the NGI umbrella), including Forgefed.
Although fedeproxy has no implementation yet, there is enough material to explain how Forgefed fits as well as the strategy to use Linked Data in a pragmatic way. I won’t be available on that day but I’d be happy to work with anyone interested and rehearse the presentation via videoconference.
I suppose there is going to be a short Q&A afterwards and that recording the presentation in advance is not really an option. Or maybe it is?
Recording the presentation in advance is actually possible! We use BigBlueButton, which allows to show prerecorded videos (which will not be viewable in the recording of the webinar). There are some notes on showing prerecorded videos in BigBlueButton, copied here: “Shared external videos will not appear in the recording. YouTube, Vimeo, Instructure Media, Twitch, Dailymotion and media file URLs (e.g. https://example.com/xy.mp4) are supported.” So if no one would be available to present on Monday, recording a 5 min presentation and sharing me a link works fine as well.
This is excellent news! I’ll work on this presentation tomorrow. It resonates with today’s discussion with @arthurlogilab regarding how to approach linked data. In a nutshell I think both ActivityPub with forgefed and REST co-exist. They complement each other and their use case is different, as I tried to explain here. I’m however quite new to this and I fully expect pushback from more experienced developers. This is actually my main incentive to speak on that topic
Thanks for being so flexible
linked-data.odp (247.0 KB)
Well done @dachary! Concise and comprehensive.
Though I still find it a missed opportunity in how you position FedeProxy as the “sub-optimal, temporary band-aid solution” until full native support in forge software is achieved. On the contrary FedeProxy has the opportunity to be forever more powerful than native forge-2-forge support, when it sits as an independent service in the middle between forges. Positioning it so needs not be a lot of extra work, other than ensuring future use cases can be facilitated and don’t need an overhaul, or are impossible altogether.
The “sub-optimal” case then constitutes your MVP, while the project itself is more visionary and exciting to potential contributors. You don’t ever need to go there, but will keep the path open.
(Related to earlier discussion on issue federation development plan)
I still don’t get it I’m afraid. But I’m very intrigued.
I’ll start a new topic to explore this further. It deserves its own discussion.
Ha, it might also be me who misunderstands. It is more that FedeProxy is not just a temporary proxy that sits in between forges, but it is a ‘power slot’ where you can plug in any additional functionality that forges on themselves do not and will never provide.
Update: See Multi forge web service with unique features for more info on this idea.