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.
this post was submitted on 02 Feb 2025
792 points (98.5% liked)
Fediverse
29591 readers
721 users here now
A community to talk about the Fediverse and all it's related services using ActivityPub (Mastodon, Lemmy, KBin, etc).
If you wanted to get help with moderating your own community then head over to !moderators@lemmy.world!
Rules
- Posts must be on topic.
- Be respectful of others.
- Cite the sources used for graphs and other statistics.
- Follow the general Lemmy.world rules.
Learn more at these websites: Join The Fediverse Wiki, Fediverse.info, Wikipedia Page, The Federation Info (Stats), FediDB (Stats), Sub Rehab (Reddit Migration), Search Lemmy
founded 2 years ago
MODERATORS
Yeah, this one guy has been trying to jumpstart so many projects. And that's admirable, but it's also too much for one dude (and I know he's tried to get some other people, but there has been drama around that - by him, I think).
But more to your point - something feels a little off even though I can't confirm anything. It does feel like this guy might just take some money and run if given the chance.
And I don't necessarily blame him for that - I probably would too. I don't want to work, and if I had money I wouldn't.
But there's just this feeling I can't shake about all of it. And I have a Pixelfed and a Loops account. They are both fine, so we'll see, I guess. But I wouldn't be surprised if I end back up with Mastodon being my sole social place for sharing photos, for example.
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
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"
Well that's great news if you're right. It doesn't change that the creator seems to be immature and therefore not the most reliable social media Steward, but if the SW keeps getting developed or a better system comes from the open source nature as you suggest then I'll be happy.
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.
What's a good idea in open source that did die (and is still absolutely dead, even)? I'm sure you can find a GitHub project and be like "I like this idea and it's dead" fine. But these projects are so used and popular... they aren't going anywhere. That's the context of this discussion. These apps are fine if the dev leaves because they are open source now. The right people will continue the work of this guy goes I think.
Yeah "open source"... The app doesn't even have the source code...
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.
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.
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.
load more comments
(2 replies)
Where's the code? There's only 9 commits to the backend and its only scaffold code. There's not even activitypub?
Same with the app. Was this just a publicity stunt?
Remember it's a solo dev still. I guess a nice weekend project would be to create a new material3 app for it :)
Of course, my problem is just that he isn't being transparent, and is dishonest. If he actually made the code open source, and did not get himself into controversy all time i wouldn't mind him being a solo dev :)
A few days ago he posted (then I think deleted) this, which among other things makes me extremely inclined to not believe anything he says:
In March 2019 I teased the Pixelfed mobile app. The only problem was, I didn't know how to write one. So just like the original "federated instagram" post, I faked a commercial to build hype. It was coming along well. Then in mid 2020, I came out and put the app aside. It wasn't until late 2022 that I found a Pixelfed react native app, and cloned it and tinkered. In 2023 I released the first Pixelfed app, but it was slow. I rewrote it from scratch in May 2024. From faking, to making!
wtf? Thanks for sharing this, i didn't know.
The Loops app certainly feels like it. Not to diminish a solo dev doing something on his own, but it just seems like a standard video player with like, comment, follow. Search feature is only by user; when you type a comment, it doesn't wrap the text, it just moves horizontally until you post it. I think the feed is literally just a chronological list of every video posted, and for some reason, most of the videos I see on it are just like...atmospheric videos. Still camera, nature background, rain in someone's backyard. People love the concept, but I don't see how it's anywhere near ready to compete for users like other options are right now.
The web UI has been teased for so long, too. I'm losing trust with dan. I mean im glad he found success with pixelfed, and i wouldn't even have a problem with how long it takes loops; but he's just dishonest. If he was more honest, transparent and open sourced the code early on i would've had no problem. But this? what is this?
load more comments
(1 replies)
load more comments
(5 replies)
Dan is constantly talking about buyout requests he's receiving for Loops and Pixelfed, so much so that it makes me a bit suspicious of those claims. Nearly every other post I see from his Mastodon feed is bragging about another alleged buyout offer, and how he's not for sale.
Has he ever shown proof of any of these offers he's received? It's not that I want to doubt him, but he's been very persistent with this claim and I think it's fair to scrutinize it at this point.
Nah. If you've ever run a mildly popular website, you'd know that you get spammed with ads for seo, content marketing, and buyout requests.
Yes, but nobody believes those buyout requests. They're clearly phishing attempts.
It makes me question my Kickstarter donation along with his other controversy. Having said that, maybe we need some of that attitude on our side. The incumbent platforms' charters are to end competitive platforms. Some sass can help us perhaps.
It is a time of predatory billionaires grabbing all forms of media. It would not surprise me if he did get offers. It would also be very unprofessional to broadcast the valuation of Loops and Pixelfed while turning down the offer. It would be like recording and showing someone asking for a date and getting rejected.
He open sourced the "code". If you actually look in detail, there's no actual working code.
He also disabled issues on GitHub, so there's no way to report bugs (and thus discuss his code).
While he could be getting offers, everything about it just feels like a scam.
load more comments
(3 replies)
it might be useless gossip but for some reason a lot of fedi has various problems with this guy. for various reasons… he just has like … off vibes?
i guess similar to how lemmy devs are marxist lenninists—at least making it open source means projects can be forked if the founders go off the deep end.
He stirs a lot of shit on his Mastodon account and gets into spats with other fedi devs. Just the other day he got into a one-sided spat with the GtS dev because the GtS dev implemented a feature to randomise the number of active users, which led to goblin.technology topping pixelfed.social in the FediDB charts. He then accused them of doing this to undermine him specifically, of wanting to 'de-legitimize Pixelfeds growth', despite it being explicitly a privacy feature.
My favourite part about this is that Pixelfed has been misreporting stats for years.
load more comments
(1 replies)
not just "off vibes", he posts a bunch of shit that people call him out on, and then deletes it
load more comments
(1 replies)
Not even a surprise open sourcing...
Nvidia did it better :)
I thought it already was open source
It wasn't. He had previously said that he would open source it after it had become more stable, but there was a lot of skepticism that he would follow through.
load more comments
(2 replies)
load more comments
(1 replies)
view more: next ›