spaduf @ spaduf @slrpnk.net

Posts

97

Comments

452

Joined

2 yr. ago

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.

Lemmy is the “name brand” now for ActivityPub based federated content aggregation

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.

In fact, you don’t need to be based on Lemmy’s codebase in order to be compatible with Lemmy. See: Kbin/Mbin, https://azorius.net, https://narwhal.city (which seems to be lotide’s flagship?), or heck, Mastodon (although the interoperability UX there isn’t the best)

Of all the misplaced priorities by the dev team I really think this is one of the biggest. If they just fixed authorized fetch Lemmy would almost certainly be the goto host for groups among the broader Fediverse.

Don't forget the whole point of the rebrand was to sell off the actual HBO content and fill Max with shitty reality TV. It's the exact same thing only they don't send you an email about it.

I think the Mastadon theming looks pretty good but I really don't understand why it's so difficult to just rebuild their frontend on top of it? Can somebody smarter than me explain why that's such a challenge?

It is certainly becoming a pattern. Similar upsets around both moderation and database optimization.

Let's not forget, there is a very real sense in which building the communities is harder than building the software. Anything that can be done to preserve existing communities is a win in my book.

I really don't think it has anything to do with Rust. Just poorly written firstpass SQL queries. See here: https://github.com/LemmyNet/lemmy/issues/2877

They talked in a recent podcast about how they're doing a full pivot to the fediverse and that certainly got my attention.

Lemmy needs a fork, if only to kick the devs into gear with regards to actually working with/listening to the community. 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. 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. This isn't the first time this has happened and it won't be the last. Mbin is reviving Kbin as a project and we need something similar for Lemmy.

Does anybody know if there are significant differences in moderation or the federation of moderation actions? One of the few things keeping me from switching to kbin.

Definitely talking about linked notes. Obsidian is far from the first or only player in the space. Logseq is out there for the FOSS diehards. I actually very much prefer the Logseq paradigm but struggled with performance issues on my machines.

Love Obsidian and linked notes in general. The potential utility there is insane but it's such a steep learning curve. I really think that in the not too distant future they'll be teaching it in schools.

Do you have any favorites that you would consider posting?

Correct. Lot's of official youtube playlists and links to the various institution's own pages for them. I've tried to keep it so most of the material in my own collection is book based rather than original coursework. Otherwise, whether or not the assignments and assessments can actually be found is kind of a crapshoot. It seems like a lot of institutions pulled back on their coursework sometime over the pandemic, presumably reassessing their value given what appeared to be a long term shift to virtual learning. I'd like to get organized about what resources are actually available and am currently looking for community feedback on how to do that in the most effective way.

Pixelfed also just integrated this feature.

Where did this title come from? Seems like a strange editorial choice on the poster's part.

Paper

I have no idea. What do the links above look like for you guys over there?

Most YouTubers won't switch as they are generally in the business for monetization purposes. Early experiments in monetization have been done on PeerTube but nothing significant enough to be competitive. It does have first-class support for donation links but that's pretty much the bare minimum at this point.