If you want movies you use Radarr, and if you want TV Shows you use Sonarr. And if you want either of those to use torrent sites to find things rather than Usenet, you setup Prowlarr to convert from those random sites into the format Radarr and Sonarr support.

There are others, but that’s a place to start.

The fediverse has been around for a while, and fediverse people are more likely to want browser extensions than most demographics, so honestly that may already exist!

Yeah, the redirect is easy. The hard part is that I’m just some unauthenticated user on whatever.net or something and I clicked the 👍 button. That could pop up a box that’s like “hey, what server is your server?” and then I type it in. That could be done, but it’s kinda crap, and we don’t have anything better 😅

Maybe I should create a new post for this but has there ever been discussions regarding SSO? So you have one identity across all fediverse services?

This comes up from time to time, but usually not for what you want. Pixelfed and PeerTube and Lemmy and Mastodon are actually pretty different experiences. They don’t really do the same thing, they don’t appear to have the same verbs and nouns. So using one to talk to the other is actually a little weird, not because we’re trying to make walled gardens, but just because the focus of Lemmy is on the threaded topic, and the focus of PeerTube is comments on videos in my playlist.

The fact that they can cross-talk at all is kind of an accident. Each system wants to federate with itself, and so they have a protocol to do it. And because it’s an open protocol, anyone can use it. And it is intentional that any compatible software should be able to use it, so Mastodon and other microblogs can cross-talk, that’s on purpose. But with these different “kinds” of services, they all picked the same protocol because it already existed, it already worked, and it met their needs. I don’t think the people making Lemmy really intended PeerTube users to use it, even though it’s sometimes possible in particular ways. They’re compatible because they both used parts of the same protocol, and so when you put them together they happen to have overlap, but that’s almost a coincidence.

The reason SSO sometimes comes up is actually to solve a UX problem that’s plagued the fediverse since the beginning. If I’m a user of lemmy.ca, and I’m looking at lemmy.ml because I got there from a link it Google out something, and want to comment on what I see there, I can’t. Not directly. I can’t click the join button or follow a user or any of it, because this site is not my site. I have to first go to my site, where I have an account, and then find this content on my site, and then interact with it there. That sucks and has always sucked. So one of the proposals people have pitched to fix it is if I could login to lemmy.ml directly with my lemmy.ca account, then I could drive it remotely, in context, while maintaining my actual account somewhere else.

The downside of this is that a whole bunch of random sites have tokens for me, every instance has way more “users” than users, and if any one of them has a security incident then it doesn’t just affect the users of that instance, because that instance also has keys for a bunch of other random instances. And overall the way I’d login on the remote site is to type my home site’s address, to kick off the SSO login, but if I’m doing that anyway I could also type that in and just have it redirect me there natively. So not great.

If we’re talking about using SSO just to only have one credential, this is actually better handled with normal, existing, SSO. Like OpenID or whatever. If Lemmy and Mastodon and PeerTube and PixelFed all allowed creating an account with an existing SSO solution, of which there are several, then you can already create an account on each of them using the same identity provider and not make any new accounts. This is likely cleaner than requiring each of them to be, themselves, an identity provider just so they can all login to each other so you can start with any one of them natively, but from there only have one identity for all the rest. That would add a bunch of extra requirements to being a valid implementation, and maybe lead to some bad or insecure identity providers, and not give that much benefit in return.

But I love SSO as a concept, so we should definitely support the much simpler thing, which is that all FOSS websites support SSO standards, not for fediverse reasons, but just because it’s nice in general. For me 😛

And so I’d say this also answers your broader questions. Since Activity Pub doesn’t allow me to join other servers, it also doesn’t allow me to join other sites.

So a Lemmy post may be compatible with a Mastodon message sent to a group or something, that’s only because the messages Activity Pub sends are similar. But PeerTube is different software with different buttons, and the existence of those buttons on PeerTube doesn’t change anything about what buttons Lemmy has, and I can’t “login” to a PeerTube server with my Lemmy account, so I don’t gain any special abilities outside of what Lemmy can do.

The only way it would be possible is if Lemmy added a feature for uploading videos that sent the same kinds of messages to other servers that PeerTube sends. Then, if they did that, someone on a PeerTube instance could see these messages coming from Lemmy and interpret them on their server as a PeerTube video or something.

But all that Activity Pub allows is exchanging of information between sites. For them to interoperate in a way that makes sense, they need to exchange the same kind of information.

Quick clarification, because I can’t tell from your words if you’re confused about the concept of federation or not 😅

If by “join” other servers you mean use their site as if logged in, or like you have an account there, then that is not federation. That’s single-sign-on (SSO) and is not a feature of the fediverse (Mastodon, Lemmy, Peertube, etc). That would be like the “login with Facebook” or Google buttons around, where by having this account on site A, you can instantly signup on site B without making a new password or anything. That’s not how federation works.

Federation is like email. I can have an email with GMail, you can have one with Proton, and someone else can have Yahoo, and I can send an email to you anyway. It doesn’t mean I can “join” Proton with my GMail account, it doesn’t mean I have a Yahoo account, it means I don’t need a Yahoo account to communicate with Yahoo users.

But, if by “join” you meant “join a community” as in subscribe to updates from a group on another server, then most other people’s answers apply. I wouldn’t call that “joining a server”, though, because servers host many communities and you’re not joining all of them.

Joining a community works like joining a mailing list. Activity Pub allows accounts on different servers to communicate without an account on their own server, so my account would send a message to your account saying “I’d like to subscribe to this community, send me a message whenever something happens on it”, and then the other server says “okay, will do”, and then after that will periodically send my server messages saying “hey, here’s that update you asked for”. And when I comment, like right now, it’s like an email being sent from my server to yours, and then your server puts it into the history.

This allows my server to present the community from your server to me, without me having account on your server. Without me having to “join” your server, I’d say.

A related thing to Aspect 3: Project to Product I forgot to cover was the “selling out” aspect. Ignoring the original creator of a project, the other Tinkerers and Enthusiasts who join a project early are the sorts of people attracted to the idea of the project, and also the value of its freedom. That’s why they joined a scrappy little community in the first place, and supported its organic and natural growth.

But when money starts coming into it, that same sort of person is often going to feel a little betrayed by it. I was just doing this for altruistic reasons, and it seemed like you were too, but now this is a financial project? Now you’re charging money? I’m not getting any of that money for the work I did to get us here. It doesn’t feel like “we” are all equals here. It’s not a community once one of us is making money and the rest of us aren’t.

Wasn’t the point of this to give it away? For free, to anyone and everyone? When did that stop being our goal?

Etc.

Feelings can be hurt, incentives can change, and that’s difficult for a project socially.

These are only my opinions:

I think the issue is “phase changes” are always going to be tricky in any community, and the path an open source project takes always includes one or more phase changes.

Aspect 1: Tinkerer vs User.

By their nature, most FOSS projects start with a person who wants to solve a problem for themselves, rather than as a business idea. This attracts the sort of person who is maybe only vaguely worried about solving that problem, but is very interested in the solution itself. Or maybe the tool does 90% of what it needs to for this new use-case, so a tinkerer is happy to go in and add the extra 10% themselves to get it to work.

But at some point along the line a project becomes popular or useful enough that Users show up. And Users want something fundamentally different than Tinkerers. They don’t want to work on the project, they want to use the project as-is to do some other work. It’s just a tool that allows them to accomplish what they’re really trying to do. And the Tinkerer mindset that got the project to here is fundamentally incompatible with the User mindset that allows it to grow outside a small group.

It’s important to note that almost everyone who is a Tinkerer in something, is also a User of other things. Maybe I’m working on this project, but my editor is just a thing I use. I need that to work without me thinking about it, so I can get the other work done for the project I do care about. And if I’m tinkering with my editor then I need my kernel to just work. Or my hardware. Or my internet. Or my electrical grid. These were all things that somebody once tinkered with, that now I’m just using, but that transition is fraught.

So if I’m Tinkering with a Linux phone, I’m more tolerant of issues, I’m invested in the project improving, and it’s fun to overcome limitations. This is essential for the project to start and progress, but there will only be so many people interested in that. Everyone else is a User who just wants to read their emails at the grocery store, and is pissed when that doesn’t work.

Aspect 2: Hobby to Job.

At some point most FOSS projects are an intrinsically interesting hobby, but the idea of financial support makes them an extrinsically motivated job. Studies have shown in a lot of cases being paid to do something actually makes it less enjoyable or interesting, even when it’s the same actions in both cases. So there’s that paradox, my job takes time away from this hobby, but making this hobby my full-time job makes me want to do it less.

But even more importantly at a project level there’s a phase change around funding. Most of these projects when they’re just a few people in their spare time, have no need for money. What does $7 a week really get me. Sure it’s “support” and “thanks”, but it doesn’t do anything to shift how this project fits into my life. Then that grows to $100 a week. That’s money, for sure. But it’s not “quit my job” money. So I get this money, it helps with groceries, but still doesn’t produce more time to work on the project. At, like, $1000 a week, now I can maybe quit my job if I live in certain parts of the world. Everything up until now is kinda “nothing” and then it’s only when we get here that suddenly “something” changes. Phase change. But I still can’t hire a second person. And until I can, any extra money is more income for me, but doesn’t really help the project either. I’m already full-time on it, and $1500 a week doesn’t buy me more time.

So because of that phase change, when the project is small it feels like there’s no reason to go through the work of setting up donations or subscriptions or whatever, because it’s a hassle that’s just going to get me, like $7. At first.

Aspect 3: Project to Product.

Like the Tinkerer to User spectrum, but for the community as a whole. When we’re all working in our spare time for free, giving the result away for free is easy. And it feels good, because we’re a community. But at some point in the future of this project, one can imagine a point where there’s a company that sells this product for money or makes money off it in some other way.

But when? We’re here, that potential future is there, and like with Hobby to Job, there isn’t really a smooth line between the two. It doesn’t feel like a Product now, I’d be embarrassed selling this and I don’t even know who’d want to buy it. But if it’s ever going to be a Product, someone’s going to have to buy it. Someone’s going to be the first to buy it, even. Who? When? How? These are answers that will have seemed obvious in retrospect, but are perhaps impossible to tell in the moment.

And what’s worse is that the skills it takes to sell things to a company are different than the skills it takes to Tinker. In some ways even opposite skills, given that the Tinkerer just wants other people to be interested in what they’re interested in, and wants to give it away for free. They want to spread it without restriction, that’s why they started working in a FOSS thing in the first place! At which point does that person decide to charge someone money instead.

And what’s even worse than a User is a Customer. Where a User might just want something to work, a Customer feels entitled to the thing working. What else did they pay for, if not a working thing. But $70 is a lot of money for a Customer to spend, and not a lot of money for a Tinkerer to use. So the amount of entitlement the Customer feels outpaces the amount of value the Tinkerer recieved in trying to bend their project to the Customer’s demands.

And as the company gets bigger to support more and more customers, you start needing lawyers and HR and payroll and support people and graphic designers etc etc etc.

This is partly why so many tech projects are picked up by already established companies and deployed as part of their product. Because they already have all that crap the Tinkerer doesn’t like thinking about, and the code is freely available for the taking.

For comparison: https://rosettacode.org/wiki/Read_a_file_line_by_line

Oh, then just go with “Joanne” then! Easy win!

/s

I think it’s because the early marketing and hype compared NoSQL to rdbms. At the beginning they were all “hey man, don’t schemas suck? Isn’t it a pain having to migrate your data? Sometimes you just wanna cram shit somewhere, go fast, break things, and your DBA is a jackass! MongoDB”

And people, at that time, were either like “what the fuck?” and continue to not trust it to this day, or “hell yeah brother!” and then put everything into Mongo and were surprised when it lost some data or got into a corrupted state, or at least were surprised the first time they thought “huh, I really wish there was some consistency to all this data…”

So yeah, I think MongoDB didn’t come into the scene as “I’m a new kinda thing that has niche uses” it came on as “hey pussy, why are you still using your dad’s DB. Are you afraid?” and people still carry that in their hearts

I’m a seeker. When I’m on mobile and I double tap and instead of going back a few seconds it pauses and unpauses, or changes from full screen or something, or worst of all “likes” the video, I have literally left websites before and been like “I’ll find this somewhere else”. I cannot stand having to wing it with a tiny scrub bar under my thumb.

A button that skips back is fine, but often this comes without a skip forward which is sometimes annoying.

When my phone is in portrait for browsing, and then I hit full-screen on a landscape video, I expect it to rotate for me, rather than showing the video the same size it already was but with huge black bars above and below. Some sites do it, it’s not technically impossible, figure it out.

And then never use TikTok 😛

That’s great! It’s what’s been holding me back from other self hosted solutions, is that they seemed to want a full two-way sync. Maybe it’s time to make the switch!

If you have a lot of photos, does immich require they all be downloaded locally, or does it fetch them from the server as needed?

By “this date” do you mean a notification thing that surfaces photos from this time years ago, etc?

The tailscale clients are, I believe, open source. It’s just the server that’s not, and you can run the unofficial but well supported “headscale” as a server if you want. But this requires you to run this somewhere publicly accessible, like a VPS, for coordination and NAT-punching purposes.

But! I’m pretty sure as the business operates right now, that tailscale doesn’t have access to the actual data connections or anything, it’s all encrypted, they’re basically just there for simplicity and coordination. And their business model is to offer simple things for free, like small numbers of devices, with the hope that you like the service and convince your business to pay for the fancy version for money. So I don’t think it’s quite as bad as the typical “free because I’m harvesting your data” models.

That all having been said, I run headscale 😛