2021-04-23 Social CG meeting

robin Yay

redaktor was incredibly active
@bengo 5 minutes is really punishment :wink:

What we did in the last week, after the NGI0 and DAPSI rejections was:

  • Markdown parsing
  • an Audio player with .vtt / .srt / WebVTT support which will also be the
    foundation for Video

VTT can solve a bunch of problems (e.g. subtitles for translation of content that cannot be understood by the viewer or captions for translations/multilanguage or chapters … In terms of audio they can help hearing impaired etc.)

This is one of the examples which just needs an ActivityPub Object and I thought a lot about different patterns for the fact that most properties can have “multiple” and “multi language” values.
Would quickly like to demo.


Does that mean you continue developing it as free software?

Not fulltime, not in a team with 8 people. Thanks to the eurocrats:
Pretty much alone and the main focus is now a Headless CMS for the largest german publishing house.

This EU thing feels like kindergarden. I have talked to MEPs not only accredited but also very personal.
Everybody was expecting the NGI programme to fund solutions with a benefit for EU citizens.
They are my elected MEPs.
They want ActivityPub projects making sense.

Unfortunately there seems to be a totally incompetent administration and funders not doing what the elected people want. This is neither transparent nor democratic.
And while Anke is describing an EU funded fediverse in one hour in the parliament, they fund hypercapitalistic clones or private passports. I have no doubt that the EU will digitall fail completely.
Another “lost decade” …

We’re being more influential than you think. We’ll see in next funding round, Horizon Europe.
The EC is huge and works in chunks of seven years. But they’re very interested in ActivityPub.
We had great feedback from the first webinar. They’re just getting started, give them time.
Actually I think the ‘private counsels’ are more to blame than the Commission. They are not from the free software community and they’re scared touching stuff they do not know. But the Commission is not stupid and they will eventually turn to us. The cascading funding scheme makes up for the slowness of the administration.

1 Like

I do not want any private person to be influential in the EU.
I want elected MEPs to be influential.
We called it democracy.
Especially we do not need wannabe kings.

And shhh:
It is totally unrealistic to do such a software alone, see the UN (longterm) findings of Michael Downey, you need at least a 5-diverse-people-team funded.

MEPs are pretty pissed about nlnet already.
Now that I took Venture Capital, I doubt, I will apply again at EU, the killing machine, pushing people at Open Sea, rendering hommage to dictators.

My goal isn’t to punish, it’s to make space for many people to present, and also for everyone to listen about more than one thing.

Remember the old aphorism, sometimes attributed to William of Ockham:

It is vain to do with more what can be done with fewer

Here is one way of using the 5min parsimoniously:

  • 60sec: summary of the project
  • 60sec: summary of recent progress
  • 60sec: summary of the intention for the next 1-4 weeks, if any
  • 60sec: Any #good-first-issue or way community can help (if willing)
  • 60sec: “If I could wave a magic wand, I’d…”


Of course. You missed my desperate pleas, requests, petitions, appeals and encouragement that a SocialCG meeting should last 2 hours

Maybe, I should prepare a presentation and play it doublespeed :wink:

We are meeting in here due to me not being able to start a bbb. Sorry all. https://meet.jit.si/SocialCG

The logs can be found here:

Minutes of the meeting

I stand by my point that 1 hour is to less time when we have awesome demos or important topics.

Personally, I was running out of time, so here the major points which @redaktor.me did not showcase in the demo:

  • Any webcomponent/widget only needs an ActivityPub Object to work,
    the containers (mini Preview, Card, Row, full Page) expect an Activity and any Object type has it’s own widget expecting, well, an Object … These are fully responsive (was width: 50%; in demo), have noJS fallbacks and adapt to the Container

  • Any webcomponent/widget is Themed !!! You can easily create your CSS theme to use them in your client. They can look nearly as diverse as nature is.
    The aspects (called Design Tokens) like Typo or Colors which I demoed at the beginning can help you to create themes.
    And the widgets come whith Variants like flat, filled, raised, shaped so that the food is varied.

  • Themes have dark/light. You can always change to the light theme but I only showed the dark one.

1 Like

Hi all,
I just stumbled on this, too late obviously, so I apologized for not being here to represent Castopod.
When is the next meeting?

Hi Ben,
a very good question. Usually on Saturday as we do alternating FRI/SAT …
@bengo @rhiaro Is there an announcement already, should I write 1 ?


Note that there’s a big issue with @cwebber’s (lack of) presence that makes him a bottleneck. Chris is way too occupied on his own stuff to even be able to respond to requests from the community, or do it juggling with eggs. Since conf.ap.rocks has been broken for some time, and ap.rocks proposed changes have not been merged so far, the Social CG discussed that we should either take collective ownership of the activitypub.rocks domain or move to another one (@bengo has activitypub.com and I have activitypub.eu, that we’re using here for emails because it was simpler to create a new one than access ap.rocks at the time).


I don’t think it’s important for all kinds of things to be subdomains of the same apex domain that is owned by one community member. If your goal is to host something on the web, I don’t see how @cwebber is a bottleneck. You can host it on a domain that you control.

I see this as one of the benefits of ActivityPub-style federation: we can spread a community out across many separately-governed DNS names.

But that does not solve the issue, it only creates more single-point-of-failures. What we need is a proper way where multiple people have access to community resources. (btw, I have fedi.foundation domain).

1 Like

Indeed the problem is expectations. If people expect to see fresh information on activitypub.rocks and they cannot, or they expect to be able to update this site, and they cannot, there is a problem. Indeed we could decide than anyone could host ‘official resources’, but then, more SPoFs, and a dilution of information.

Ideally all information about the Fediverse should go to the Fediverse. In the meantime we need collective ways to facilitate cooperation, not clog it. It’s not a difficult thing to do and I bet @cwebber would be more than inclined to do it. He’s just super busy as usual. But if the Social CG comes to a reasonable proposal, things could be moving forward fast. I think one of the issues why it didn’t was the apparent lack of response to the How to hack on activitypub.rocks : MRs were merged but issues were not taken care of and no more MRs were proposed, nor did the topic seems to interest anyone. One way to fix this would be to add more members to the project, so that we’d have a volunteer team whose role would be to keep this site up to date. Another would be start over somewhere else. In the meantime we can Publish pages for AP guides on activitypub.rocks collectively, here.

Being a bottleneck is not a judgement, it’s a fact. The discussion should revolve around possible ways to remove any bottleneck in this community while remaining cohesive enough that we don’t go each our own way, which would not be a great outcome.

The best I’ve seen is to use a tool like https://www.loomio.org/ to help with that. i.e. come up with a good-enough way of making proposals, voting on them, and enforcing a time limit to reach approximate consensus, then let a robot do the busywork of tallying votes, sending reminder emails, etc.

Tbh, I am against that. What Loomio offers could as well be done in similar ways on this Discourse forum, and would not require a new separate tool. Note that, when I tried Loomio I found it severely lacking as a discussion forum UX-wise… still years behind Discourse in that regard.

1 Like

How can discourse tally votes? Does it support stacked rank voting?

It does not support everything in the same way that Loomio supports, but additional workarounds can be found, I think. By default Discourse supports polls, and they have some settings to configure them. E.g. a public vote:

Howto tally community votes?
  • Let’s use Loomio for voting
  • Discourse has sufficient functionality

0 voters

In addition you can specify topic templates for a specific subcategory, and it can include Mentions, so that email notifications are sent. And in that way there are many other ways to make things easier. Also on Discourse Meta forum there are many theme components and plugins that we may consider to make this power tool even more powerful.

Update: Multipl-choice voting (in this example restricted for trust_level_2 and above, and not public)

Elect the two members of the XYZ team
  • Alice
  • Bob
  • Alex
  • Zena

0 voters

It is not ranked, however, but when the results are in you can see the number of votes per candidate. If two have equal votes, you can do an additional poll.

The main difference I see between Discourse and Loomio is that the former focuses on conversation, that forces people to explain their thoughts and find a way to compose with each other, while the latter focuses on “voting tools” that cuts shorts options to what the powerful being in charge of casting the voting options can come up with. Personally I do not appreciate technological solutions to social (and political) issues. Loomio hes this shortcoming for me that it delegates proper discussion to expedients.

1 Like