You should consider blocking them. Especially if they’re consistently bugging you
You should consider blocking them. Especially if they’re consistently bugging you
Yeah, I think it’s a legitimate and growing problem. I think a lot of folks don’t realize, but since growth has slowed from Reddit more broadly, the people who feel they have been “unfairly silenced” are the fastest growing subpopulation around here. If I’m honest, I think the only real antidote is to reestablish growth from communities with kinder dispositions.
I totally thought that was in 0.19 but I haven’t actually seen that yet.
Worth noting, the number of people who come here “to escape authoritarian moderators”. Nearly all of them were moderated for good reason.
I also don’t think the presence of places like hexbear are doing us any favors.
Anki is a lot more robust than just about anything else out there. Particularly as a supplement to other learning. I’ve been through a lot of different language learning systems and my anki decks have been a constant throughout.
I don’t think anybody’s tried exactly nebula-style, but there is already https://newellijay.tv which seems to be a kind of video-outgrowth of an existing rural makerspace? Pretty cool project from what I’m seeing
PeerTube is not really intended as a platform, even less so than most fediverse projects. As it stands, the best way to think about PeerTube is sans discovery mechanisms because I don’t think any are planned. With this in mind, peertube is best thought of as the video extension of the fediverse and the discovery niche is filled through word-of-mouth here and over on the microblogging side.
There are multiple monetization plugins and absolutely no built-in anti-monetization features. I don’t think it’s unreasonable for them to keep the base software monetization agnostic. They talked about this at length during the AMA a couple of weeks ago. I believe this one is the most popular: https://github.com/samlich/peertube-plugin-web-monetization
No reason a Nebula-type model couldn’t see success on peertube
who overnight know intricacies of engines
To be fair, this is an old, old engine with several generation defining blockbusters making use of it. Not to mention the massive modding communities who’ve probably spent more collective hours fighting with the engine over the past few years than Bethesda has.
Captain Disillusion! More VFX than anything else. He debunks viral videos and explains how the effects are achieved.
They were probably more swayed by the loss of their upcoming payout than any love for Altman. Increasingly sounds like Ilya just bungled the timing
As I see it, there are three major ways a fork could gain significant standing among the community:
I honestly think any one of these is easily manageable by a handful of people in off time. Other parts of the fediverse of similar size are chock full of forks.
I would love to contribute but I don’t have the experience for a fork. This is kind of the essence of the whole problem though. Plenty of unutilized contributors who could be driving this project forward but are having a hard time getting involved.
It’s one thing if it’s just a couple of devs working on the project and trying their best, it’s an entirely different thing when a couple of devs are shutting out large numbers of contributors (frequently subject matter experts which they desperately need at this point) over relatively trivial issues.
To the detriment of the community, the admins, and the concept of the fediverse overall.
deleted by creator
Strange take.
Not for folks who have been following the development. It’s one thing if it’s just a couple of devs working on the project and trying their best, it’s an entirely different thing when a couple of devs are shutting out large numbers of contributors (frequently subject matter experts which they desperately need at this point) over relatively trivial issues. It’s become a pattern and will almost certainly continue. At this point a significant number of users have been lost because the devs have been largely unable to capitalize on previous waves on growth due to slow development. Because of all this Lemmy has an awful reputation even among the rest of the fediverse and particularly among people who have tried to contribute. A fork would probably be a significant improvement as far as brand perception goes.
The biggest issues that have come up so far are moderation and database optimization. The moderation issue is significant enough that large instances have considered shutting down, but the database optimization thing is what really drives me crazy. It is absurdly expensive for hosts considering we only have 35k MAU (just one of our midsized instances should be able to host the whole userbase for the cost they currently pay) and it has been largely deprioritized to the point that contributors who have tried to fix it have been told off.
It’s one thing if it’s just a couple of devs working on the project and trying their best, it’s an entirely different thing when a couple of devs are shutting out large numbers of contributors (frequently subject matter experts which they desperately need at this point) over relatively trivial issues. At this point a significant number of users have been lost because the devs have been largely unable to capitalize on previous waves on growth due to slow development.
Not to mention things like authorized fetch, which if fixed would ensure Lemmy/Mastodon interoperability and would effectively make Lemmy the go to place for groups on the fediverse. This would constitute a huge boost in engagement from the broader fediverse.
Because of all this Lemmy has an awful reputation even among the rest of the fediverse and particularly among people who have tried to contribute. A fork would probably be a significant improvement as far as brand perception goes.
This is a great project and I’m surprised by the tone of the response here. I think most folks are forgetting that most of the people dealing with configuration are not programmers by trade. They just need to setup a tool for their use case. To that end, the gap between the existing configuration paradigm and extending their software is practically insurmountable. This language bridges that gap in a robust and purpose built way and that is going to make a lot of people’s lives and jobs easier.
Think about homeassistant and how much less fidly it’d be to get advanced functionality or interfaces if the gap between programming and configuration were closed? There is an absolute fuckton of enterprise and scientific software that will improve in the same way.