Skip to main content


The thing I love about this post and all its replies, is the stark difference in replies in answer to Brent's very reasonable question.

Some folks:
* It's the starter packs / easier onboarding!

Other folks:
[I'll let you read those replies]

mastodon.online/@BrentToderian…

If you think it's the starter packs / easier onboarding, then that leads you to want easier onboarding for Mastodon👍🏿

If you think it's a character flaw or moral failing in people that choose BlueSky, that leads to doing nothing🤷🏿‍♂️

in reply to mekka okereke :verified:

Yeah I'm on there as well as here, and I like a lot of aspects of bluesky, like the easier onboarding, moderation lists, labelers, quotes, etc, but I *much* prefer fedi/activitypub's model compared to bluesky's big central service.

Of course i dislike that bluesky isn't a nonprofit, is funded by a crypto group, and hasn't yet handed off atproto to a nonprofit or standards org, but that's nerd stuff.

Most people care about ease-of-use, feeling safe, finding their friends, and literally nothing about the protocol or the decentralization or anything like that. Bluesky is more twitter-y than fedi ever was, and i think that's what most folks are looking for right now.

This entry was edited (1 week ago)
in reply to Eris Renée

Just to be clear: i'm very critical of a lot of aspects of bluesky, and it frustrates me to no end that so many people don't see or don't care about the red flags I see with it, but there are still a lot of things we can learn from them from a UX and moderation standpoint.

I also don't think fedi needs to "win" in a competition here. It's not a company that needs to show constant growth. Bluesky is a VC-funded for-profit corporation, of course they're going to grow faster and have more resources to develop nice UX. We should steal some of that nice UX though.

I much prefer the sense of community here. Bluesky feels a lot more influencer-heavy imo.

This entry was edited (1 week ago)
in reply to Eris Renée

@Eris Mastodon has a lot of room for improvements UX wise (and there's a lot of effort that is going underway even though it's, shall we say, not as considered as it should by those with merge rights), but I'm skeptical about how much BS can be an inspiration for those improvements. How much of the “UX improvements” in BS are a direct consequence of it being a centralized system above anything else? (One example for all: you don't need to choose a server because there's only one.)
in reply to Oblomov

@oblomov I’m on my phone now so i’m sorry for my typing.

the lists are a big one. starter pack lists make it very easy to find people you might want to follow, and are more intuitive than our hashtag introduction system, though bluesky’s implementation means anyone can add anyone to a list so they’re also used for harassment - requiring approval is probably the move here.

the moderation lists to auto-mute/block folks are pretty cool too, and i think it could work alongside our per-instance moderation with some thought.

labels are neat, although i definitely expect they’ll be misused by bad actors there. probably need some moderation around them. kinda gotta see how that one shakes out.

Those are the things that scream out to me right now. All this shit takes development time and a lot of thought, though. They have waaay more funding than any fedi project i know of.

This entry was edited (1 week ago)
in reply to Eris Renée

@Eris oh yes the ideas are nice, but the problem is, nice ideas have the tendency of not scaling that nicely when decentralization is involved. Consider for example the starter lists idea: where does the server fetch them from to present them to the user? Who manages addition and removal of entries from the list? How do changes to it propagate?
We already have something like that for the “reference blocklist”, and even that essentially relies on a centralized service (Oliphant) 1/2
in reply to Oblomov

@oblomov @Eris

Starter lists are user generated. I can make my own starter list right now.

Users use the starter lists recommended by their community, and by people they trust.

in reply to mekka okereke :verified:

@Eris thank you for the clarification. I have not joined BS (and have no intention of doing it), so if you could waste some time to fix my ignorance on the matter: how do they get the recommendation after they join? Do they have to actively search for it or is there some UX where they can ask something like “are the comic artists in here” and they get presented with starter lists others have compiled?
in reply to Oblomov

@oblomov @Eris

People usually recommend them, but you can search for them too.👍🏿

I just searched for "comics starter pack" and two entries came up from people that seem trustworthy to me.

One is by @gregpak (Yes, that Greg Pak. Who wrote Planet Hulk, which eventually became Thor: Ragnarok).
m.youtube.com/watch?v=lWqUkMFi…

Another is by Eisner award winning artist Diana Souza.
portfolio.dianasousa.com

There are more, but these two illustrate the point.

(Feels very web ring-y! Shout-out Sage Weil! 🙂)

in reply to mekka okereke :verified:

thank you very much! From the rest of the conversion I had with @Eris (subthread here <toot.cat/@Eris/113480947022388…>) this feels like something that could be implemented also for the Fediverse via public lists and a “one-click” import. The bottleneck would remain cross-instance reach of the list author, which BS solves by being centralized. I'll make an example to clarify what I mean.

1/n

This entry was edited (1 week ago)
in reply to Oblomov

@Eris Let's pretend Fedi platforms implement these public lists and one-click import, and that @gregpak and others well-known names compiled their list on Fedi too. I'm a newcomer to Fedi and join randominstance.social and search for comic-related lists, but since neither @gregpak nor the others are known accounts to the server I've joined, neither are their list, so the search turns out empty. So even with the feature, federation would end up making it less useful.

2/n

This entry was edited (1 week ago)
in reply to Oblomov

@Eris

So this is actually a wonderful feature and I would love to see it implemented, but at the same time I can bet on the fact that by the federated nature of the network it will *still* be a less smooth experience than a centralized service can make it, and this will *still* be a selling point for the corporate silo.

(Again, this isn't to say we shouldn't have it, I actually like the idea a lot. I'm saying it wouldn't be sufficient even if we had it. Ditto for blocklist etc.)

in reply to Oblomov

@oblomov @Eris
The lists don't necessarily have to be available as soon as you connect, if anyone can create public lists that others can easily follow (e.g. they link to their list, or you can see their public Lists in their profile), it will already be a huge step.
in reply to El Duvelle

@elduvelle @Eris oh it will be a huge step, especially since if you join a high-volume server or one with topical interest, you're very likely to find a relevant list now and that will help build your follows quickly.

But in general, they will suffer from the same propagation issue that individual accounts have (what was the story of the journalist that tried Mastodon and dropped it like a hot potato because a colleague couldn't find him? I can't find the link anymore).

in reply to mekka okereke :verified:

I love BlueSky, mainly because it's now our best bet at Twitter not being as influential in the US midterm elections (that are coming up sooner than we think).

I like to brag and boast when I am right about a call, so it's only fair to admit when I'm wrong: I really thought Mastodon would get it together on the onboarding and trust and safety front.

In fairness, Mastodon is getting it together... but just far too slowly to matter. A slow "yes" is often the same outcome as a "no."🤷🏿‍♂️

in reply to mekka okereke :verified:

I think it's important to note that thirty states have elections next year (2025), so the time for anyone looking to fight Trump and his lickspittles politically is already here.