My fear with pixelfed and loops is the single dev seemingly more interested in money and clout than in building something long lasting for the community. I don’t expect it to last long, but my friends really crave an app to exchange reels in and so we’re hoping loops will be sufficient until something more stable comes around.
New to open source? Good ideas don’t really die in open source. If loops.video or Pixelfed are good ideas, open source will just do its thing because no one owns the idea once it goes open source
Good ideas absolutely do die all the time even in open source. If the original dev doesn’t want to play nice, it’s actually pretty difficult to create a new fork that everyone will agree on. Hopefully these federated apps have enough inertia to prevent the userbase from splintering when the original devs move on.
Isn’t the point of this thread that the code isn’t actually open source - that the released code isn’t anything substantial?
I see the backend and front end released. The back end is AGPL and the front end does need a license (assumed it had one, my bad). Assuming he puts the same AGPL license on the front end, I don’t see anything unsubstantial? The community gets to make squooshloops or whatever name comes up and the backend and front end belongs to them at that point to release on their hardware.
It’s all ActivityPub, so mastodon users wouldn’t miss see a thing go wrong? Start subscribing to squooshloops.social accounts from your instance’s deprecated Pixelfed software like nothing happened because the squoosh devs were smart to make it compatible with other ActivityPub apps. Once adoption hits a masse, and maybe ActivityPub specs changed, and squoosh added new things to their spec that other services expect or depend on, new versions of squoosh and other activitypub software stop trying to be backwards with the unmaintained pixelated and loops.video software. Those instances are forced to use a fork or stop federating with those other instances who no longer support the old stuff. The world keeps moving and remember that time when squoosh was originally that failed loops thing.
Kinda like how Chrome has its roots in WebKit which has its roots in KHTML which has its roots in KDE. I’m not seeing a problem honestly. Yes, KDE’s konqueror is dead (rip, I used KDE and loved that jelly K browser icon ever since I was like 12 in 2005). It lives on in chromium though, and has spread all over the world because of electron. Very weird story for that browser, but it somewhat shows my point: open source doesn’t die. It isn’t perfect — lots of valid complaints about chromium and electron, but none of the valid complaints are “that idea is gonna die if Google dies”
Wtf. That pixelfed asshole was writing this shit as closed source before?
Yeah man what an arsehole giving away free things, spending time making free things for everyone to use for free.
If he wants to keep his code private until he’s ready to release it - that’s somewhat up to him. The repo is a step in the right direction, just give it some time.
Looking through the git repo the ignore is lopping off quite a number of files too (server) however I’m not sure how much of those are things like secrets, cache, other scripts and so forth.
Well, it’s kind of up to him. His NLNet grants require it to be open source and dictate what license he uses, but I don’t know the details to know if that just means it has to be open sourced in order to receive the grant (awarded upon completion not ahead of time) or right away.
Does the grant apply to just pixelfed, or loops as well?
He has gotten at least three grants in the past that I saw on the NLNet website for Pixelfed, but none were listed as active when I looked. However, since you have to complete your project to get paid out I didn’t see a way to know if he actually received any or all of those grants. The current active grant is just for Loops. I believe the old Pixelfed grants also required certain licensing and the project being open source though.
Also, as of this morning, he was posting about how in order to make Loops open source and federated he needs to hit $200k on his Kickstarter. However, he was previously challenging people who were claiming it’s not currently federated or open sourced. So now the story is that it’s not yet and he needs more money to make that happen when previously he was claiming it already was both of those things.
It’s…a mess for sure.
That’s funny because it’s not difficult at all to build an app like this. The value in buying such a business would be the scale of its user base, which is likely only several thousand at this point. And, if it were sold, most people would leave since most of those users are just dipping a toe in at this point to see what the app is like. Frankly, the content sucks. People are using this app out of goodwill and an interest in helping to build an alternative to the corpo apps. If dansup sells or does not open source, then that goodwill evaporates.
That’s funny because it’s not difficult at all to build an app like this.
It is a fuckton of unpaid work, it never feels right to talk about donating their time and effort like that.
deleted by creator
The value of loops right now is the attention it is getting. Buyers are trying to swoop in early and bet that they can capitalize on the growth to get more money from other investors.
VC are gambling money as always
Agreed. I am downloading and occasionally checking out some of the content on all of these apps like Loops or Pixelfed because I’d like to see them succeed. I’m not a content generator, but I hope having accounts on and viewing these services helps somewhat.
E:typo
I’m not a content creator, but I recently went on vacation and put my best stuff up.
Without the algo, everything just gets lost to time. When I put a few things up, they faired well, but once they were out of the front seat, ( a day or so ) they were never seen again.
Where I post a couple nice things on bluesky, I get reverberations for days.
This seems to be the main issues with his apps. Visibility is scarce beyond the original post, wherein even with followers the chances of someone seeing your post beyond that honeymoon phase dwindles.
The only exception are users with lots of followers. From what I’ve seen in my brief time using Pixelfed for example, it’s usually accounts with at least a few hundred followers that see a highet retention of visibility.
Building an app like this for 100 people, sure. Making something handle smooth, affordable video delivery at scale. That’s a spicy meatball.
The front-end is a mess for testing but doable. Then if you do live, you’ve got proxies and stream copies.
I host stuff at great scale, it’s a different beast.
You cant do more than dip a toe in, as there is no more functionality and even that is buggy and incomplete.
Yeah this guy toying with selling is pretty insulting to anyone trying to help him out.
For the record was he? Or was he just making a joke that didn’t land.
He is pointing to these apparent offers, likely to gain attention and as a way to pressure people to donate to his kickstarter.
