How to solve our engagement problem, foster more cross-pollination?

[aschrijver]

Creating a new topic that was triggered by this:

@lanodan

I wrote a standardization draft and ended up never posting it because of the bunch of changes and lack of examples in the Fediverse Enhancement Proposal project and the more discussion-oriented thread I started went nowhere.

Why did it go nowhere? Let me guess at providing some answers:

• Because the SocialHub community is tiny.
• Because at the time @lanodan was working on this, they were the only one.
• Because the topic scrolled out of view later on, not noticed by the occasional visitor.
• Because too few of the members and people in general think to consult SocialHub and see it as an essential hub for collaboration.

And because of this the community also doesn’t get stronger, as…

@lanodan

I’ll likely end up posting it properly into litepub, here is the draft: https://hacktivis.me/tmp/uploadMedia.html

So now the knowledge lives fragmented across the web, and may be implemented without others knowing about it.

Another example where a dev is more or less left to their own devices is @grishka with FEP-400e: Publicly-appendable ActivityPub collections (got some good feedback, but mostly from one other dev @macgirvin). And same goes for Querying ActivityPub collections

There are many more examples of this - a forum FULL of open issues like this - and usually the devs involved either go it alone or abandon an idea.

Consequences

This may not be bad for the individual project that raises issues in the community. They just go their own way. But it may be frustrating to - having spent the effort to steer up discussion - see an initiative come to naught. Next time people won’t waste their time, or go elsewhere with their issue.

It is very detrimental to having a well-oiled SocialHub. And it is hampering the evolution of the Fediverse technology foundation and open standards in general.

This whole problem is well-known and e.g. @Sebastian has raised it countless times, as have I (more recently).

On the whole I feel a “Tragedy of the Commons” for the Fediverse looms, where - as we become more fragmented - it becomes harder and harder to stay relevant and keep up with technological developments that occur elsewhere and beyond our control.

Improvements?

As I’ve said in Organizing for SocialHub Community Empowerment I think that people who are actively building this community are needed. People involved in different activities than development, and also willing to do the boring chores that need to be done. There’s a win-win for project devs in it, as their project’s future is also on the line.

Also this community needs to grow, and broaden in diversity. I created Fediverse Futures - SocialHub for this, and wrote an analogy for how we might collaborate: SocialHub and Spiral Island: A Fediverse Foundation Analogy

And I created a website that might be a companion to this forum, for documentation and tracking all the open issues we find here: Presenting Fedi Foundation: Empowerment for SocialHub community

Unfortunately, even for this, there’s little to no interest  

We need to think how we can make community work more interesting for people to get involved with.

[weex]

I’ve been toying with creating a repo main for issues (no code), just to state and argue the validity of problems in the Fediverse per my recent but I believe lasting obsession with C4. I have a hard time limiting such a repo even to issues in the Fediverse as there are many problems in the world more broadly, but ActivityPub and fedi devs have very similar interests so that’s a reasonable way to scope such a thing.

This thread would fit, though I think it can be broken up into smaller problems like lack of attention, fragmentation, low signal-to-noise. Fragmentation doesn’t bother me as long as someone can assemble a sensible path through the links.

My feeling is that well-described problems attract people better in this space. We don’t exactly have funds and headhunters to rope people in and that wouldn’t make sense even if we did. I, for one, appreciate all the work you guys are doing and have done to assemble this amazing space.

tl;dr let’s focus on the problem(s) and let solutions come.

[VictorVenema]

So now the knowledge lives fragmented across the web, and may be implemented without others knowing about it.

Isn’t “fragmented” and “cross-pollination” almost the same? Except maybe that in the latter case we would make sure the copies link to each other, or link to the main version. Such hyperlinks immediately makes it easier to find SocialHub.

I maintain an Open Science Feed. It is on Reddit, Mastodon, Lemmy and Twitter. The various versions link to each other so that everyone can use their favourite platform.

We recently launch an initiative Translate Science and announced it on several blogs of the members, on several mailing lists, on Mastodon and Twitter, all linking to the blog of our homepage. (By the way Twitter may be 100 times larger, but Mastodon produced much more interaction. Twitter amplifies controversy; Mastodon quality.)

[aschrijver]

@weex

This thread would fit, though I think it can be broken up into smaller problems like lack of attention, fragmentation, low signal-to-noise. Fragmentation doesn’t bother me as long as someone can assemble a sensible path through the links.

I feel that the main issue here lies with the “someone”. There have been enough initiatives where something was set up, only to fall into decay because there’s no long-time engagement to keep the initiative going. All the community activity is very time-consuming, underappreciated work, but much needed to keep things moving.

Placing a repo somewhere would need a continuous effort to point people to the existence of the repo and what to use it for + encouragement to actually use it. We already have several repo’s like that

There are many other examples: A FEP process that’s defined but stalled, @cjs Library that people need to be aware about, numerous unfinished discussions in this forum, etc.

They all need a ton of time by volunteers doing community organizing, rallying people and maintenance stuff. And that’s what too few people are willing to do.

@VictorVenema

Isn’t “fragmented” and “cross-pollination” almost the same? Except maybe that in the latter case we would make sure the copies link to each other, or link to the main version. Such hyperlinks immediately makes it easier to find SocialHub.

Same applies here. It is astounding how hard it is to spread knowledge about the places one should look first when interested in federated dev. I’ve been spreading links to SocialHub and the AP watchlists far and wide. It is a continuous task that should be done.

But there are ways to improve overall community organization a bit further to ease things, like e.g update the Wikipedia pages on “Fediverse”.

An important aspect is also that every SocialHub member can easily help, but often don’t. I see a lot of toots where there’s opportunity to drop a link to SocialHub for instance, or where a simple addition of #activitypub hashtag is missing. Or the @activitypub Gup.pe group is tagged, but no member of the group boosts the toot.

[aschrijver]

@weex

My feeling is that well-described problems attract people better in this space

This was what I intended the Fedi Foundation website for. The main content hierarchy is article-style, but I want to add a Gitbook / Just-the-Docs page template where gradually longer-form docs can be elaborated, and also I’d like to have something like Pattern Libraries that show how particular federation stuff can be tackled.

(A Pattern is a combination of a common problem with its best-practice solution)

[weex]

@aschrijver

FEP process that’s defined but stalled

I read FEP-a4ed: The Fediverse Enhancement Proposal Process and noted a couple errors. Do I submit a PR in that repo or should that be discussed elsewhere? In theory, it’s final but I suppose corrections should be welcome.

What I’m finding tricky still is getting the data to prove that bigger problems exist, the value of those problems, and ultimately that if some threshold were crossed, that it would no longer be a problem. For example, I was thinking about developers but taking your suggestion that the SocialHub community is tiny. How many monthly active users are there here? How does that compare to a year ago? Is there a goal amount of activity/users that would be real progress?

BTW, I did create a repo but I’d be interested to know links to others so that I could contribute or pull issues in. I’ll share it once I have a decently written problem to populate it with. Ubuntu had bug #1 (Microsoft has a majority market share) so maybe the fedi could have a similar rallying point.

[VictorVenema]

@aschrijver

It is astounding how hard it is to spread knowledge about the places one should look first when interested in federated dev.

Good that people on the federated internet understand the attractiveness of centralized services.
But my experience is that it is nearly impossible to get only somewhat large groups to agree on a single platform. Having multiple platforms is less good for discoverability, but has other advantages as I guess I do not have to explain here.

@aschrijver

An important aspect is also that every SocialHub member can easily help, but often don’t. I see a lot of toots where there’s opportunity to drop a link to SocialHub for instance

Will try to think of this. This is an example of my proposed solution: to have multiple platforms, but link to each other.

[aschrijver]

@weex we have the #activitypub:fep here for discussions, and yes you could PR I think if the errors are not substantial. For more substantial revisions I asked the question before: FEP-a4ed: The Fediverse Enhancement Proposal Process - #31 by aschrijver

Discourse keeps a bunch of metrics, but not the means to analyse them in any proper detail. Below you see some metrics from the last year:

But this is not really useful. DAU and MAU are not recorded separately. In general there are some active “usual suspects”, that - when taken off these metrics, will significantly lower the graphs.

@weex

BTW, I did create a repo but I’d be interested to know links to others so that I could contribute or pull issues in.

In this SocialCG meeting post @bengo pointed to a repo that until that time I had no knowledge of. Then there are the issue trackers in both the AS and AP spec repo’s. Follow-up was in: Social CG Issue Trackers

In terms of bringing links togeter I suggested having them all in a menu above this forum, but instead they were moved by @how to a (still incomplete) wiki post. Which makes them not findable once more, I am afraid

@VictorVenema

Good that people on the federated internet understand the attractiveness of centralized services.

I know you are saying this jokingly, but I’ve heard the argument of the SocialHub being some kind of centralized service a number of times before. That’s the wrong way to look about it, unless we are becoming a million-people conglomeration. I see us as an essential node in the decentralized fabric. A single instance with but 528 users. But we are not a platform either.

We are a community of people sharing a common interest.

And a very important community for the entirety of the Fediverse. The fragmentation in this regard does not have those upsides imho and in this case.

[VictorVenema]

Sorry for my micro-agression. I was not aware this claim was often used.

There is also nothing wrong with centralized services. If everyone uses one service because it is clearly the best that is fine. The problem starts when the system is build in a way that people are forced to stay at the central service even when they are not the best.

It would be nice if everyone were here and preferable to people being spread over many platforms. But I do not think that is an achievable state without a billionaire backing SocialHub. Being the centre of a linked network is an achievable state.

[weex]

@aschrijver

you could PR

Done.

@aschrijver

DAU and MAU

I’ll just ignore DAU since that’s a monthly view. It’s not large but it’s stable so that’s something. Also noted that meetings have been going on so there’s a baseline of momentum. I guess it would be up to a community manager or community engagement working group to set a goal and do things to get there.

@aschrijver

they were moved by @how to a (still incomplete) wiki post

This is a really good list, I find myself collecting links and this would have been a great one to start with, when I “entered the fediverse” a few months ago. If there’s a welcome email, I guess it’d be good to have that in there. I saw on lemmy someone proposing to make videos about the fediverse and a good one would simply go through that page and talk about the links.

I can see the less-is-more argument to issue repos but it does sort of fail once there’s more than one because competition is good between projects and people, but a community shouldn’t compete with itself. To this aim, I suppose it’s best to join the group and participate in the meetings.

[melvincarvalho]

Hi @aschrijver , long time no see

I’m a little bit surprised by this post

I understand that Fediverse has grown to 3 million+ users with 1 million active

So my expectation was that there would be lots of activity. I’ve run forums before and I tend to find the sweet spot is around 10 posts per day

Seems quite a bit less here. Is that because it’s developer oriented, or are there a bunch of forums for devs. Or simply not many of the fediverse population interested in the tech

I’m getting a bit more time these days to look at fedi (been on my list for ages), so perhaps I can help out with some of the things that need doing …

[aschrijver]

Hi there Melvin! There are areas with good activity, and this forum has relatively good engagement at times as well (it goes in waves).

The issue in my opinion is that most stuff that happens is very project-oriented, and most knowledge and insights gleaned remain siloed in these places. There’s progression in apps and some research projects, but little to no progression in the ecosystem as a whole. It never gets easier for newcomers to enter, only more complex, and they need to hack away, dig in other codebases, meet the right people by chance, etc.

Improvements and iteration on the specs, proper documentation, well-known active community to interact with … very little of that. And indeed broadening the community beyond developers, so its members have more diversity in skillsets and backgrounds is important too, imho.

I sometimes feel that the Fediverse and communities within are ‘grassroots movements of individualists’. And on the whole that ‘time is the most precious commodity in this world’ and people just want to spend it on their own things. This is fine, and maybe just the way it is, but it refers to the problem to tackle: How to make it more interesting to participate for the greater good of the Fediverse?

At least for existing federated app developers there’s a clear win-win that is probably overlooked. Some projects like @write.as and @mastodon have their own communities and would immediately benefit from Scaling Up Cooperation. It is a low-hanging fruit.

The kind of involvement that is need is people doing the community work and the chores to bring the community to a higher level, make it more well-known, and expand its member base. It might be beneficial to have something of a ‘Community Team’, so there are always people active and things don’t fall on an individual’s shoulders. There must be “staying power”, we should be in it for the long haul.

And possibly there should be proper incentives for those people to participate: "What’s in it for me?"

The positioning and organization structure of the SocialHub can be improved. I think for once some “centralization” (in scare quotes) is in order, i.e. offer an intuitive structure to discover the best information and easing + encouraging of active participation.

[melvincarvalho]

@aschrijver thanks for the detailed reply

The kind of involvement that is need is people doing the community work and the chores

Is there a list of stuff that needs doing, of just general engagement, reading threads, supporting users etc.?

[aschrijver]

It is a combination of all of that. For instance this forum has many interesting old threads with discussions that were inconclusive. They could be elevated into attention again, but they probably need to be tracked in a better location and some people need to be active maintainers of that list. Then after an issue is solved, or an agreeable outcome reached, preferably there should be a place where they are ‘parked’ in good documentation that other people can find. Things like that…

Then there’s advocacy work. This is time-consuming too, but it is where every member can help (some culture involved, a mindset of advocacy “between the ears”). This involves dilligently dropping links to relevant topics, creating issues in projects where AP may be interesting. “Community marketing” (to use a dirty word ) of sorts.

And there’s the community-building itself. Organizing meetings and events, social gatherings, taking care of follow-ups and the like.