Neat! I’ll probably try it out. Like it or not, short form video is really popular now, so this feels like a good thing to grow Pixelfed and the fediverse
As someone who cannot stand stories, TikTok, and basically short form video as a whole, I may check this out for the fediverse!!
I occasionally get pulled into the YouTube shorts and hate howuch time I lose to them. Worse was that although I barely use Instagram beyond keeping in contact with friends who only use it, I happened to watch the reels for a little yesterday and they were really entertaining.
A lot of amateur video creators don’t have the experience to keep their work engaging for long periods of time, half the internet feels like SNL sketches that make their best punchline in the first 20 seconds and then milk the same joke for the next 3 minutes. The way short form content cuts through the crap is actually quite nice. It obviously has a whole bunch of its own issues but that’s mostly due to chasing the algorithms favour, not the short form nature of the content.
If anybody’s interested in signing up for the rolling beta, here’s the link
I don’t care for short form videos, but this is still really cool. Hopefully it provides an embed URL for the videos so the ones that get posted to Lemmy can be embedded (which is how Peertube links work).
Right like this is a mixed bag. I feel like my YouTube shorts addiction is fucking my attention span and I need to get back to watching long form stuff but like also a) fedi isn’t going to have server space for much else and b) it’ll bring the user base fedi will need to grow and maybe have the server space someday.
My other thought is that maybe things like peertube need more communities around streaming where the video isn’t saved, at least by default. This could also be a better direction for the internet as a whole; moving away from this idea that every single moment of everything must be recorded for posterity. The internet really does need more liminal spaces. I really liked the aether model but that guy abandoned the project and lemmy is here now anyway. It would be cool maybe to have a lemmy instance where everything is auto-deleted after 6 months and at least puts out requests to also delete on other servers.
I strongly disagree on that point. Preservation of the internet is extremely important for cultural and historical reasons. Just look at the enormous amount of old websites stored on the internet archive and how helpful as well as culturally significant that preservation effort is. And that is only a drop in the ocean compared to how big the internet actually is.
Yes preserving stuff, especially video, takes enormous amounts of data and is hard but is well worth the effort in the end.
but is well worth the effort in the end.
That’s easily said when you don’t have to supply the effort or pay the hosting bill.
takes enormous amounts of data and is hard
Seems like this was acknowledged, but a good point nonetheless (that’s often overlooked).
I’m currently sitting on 4TB of data (that’s largely movies and TV shows), running on 4-year-old hardware, with 3 local replicants, backed up to cloud.
My power and cloud costs are trivial - about 25 cents a day - that’s less than $100/year (after hardware costs, which come out to about $150/year to continue with similar performance levels). My 4 year old “server” idles at about 20 watts. I can probably bring this down to perhaps 10w with a newer NUC or similar.
I could easily store everything my extended family produces (including cousins, about 50 people) with a similar setup. In fact, I’m working on just such a project - an SFF or NUC type device with sufficient.
Edit: autocorrect changed $100 to $10
Great platform name. Though I like the idea, but it’s currently only one developer. It’s the same dev behind Pixelfed. He’s a good dev, but I feel like he’s stretching himself too thin to make both platforms the amount of development they need to pick up steam.
It will also come with an open source “For You” algorithm. I’ve been wishing the fediverse do this since I joined and it took literally banning TikTok for it to happen in the span of a few months by dansup for Loops, who thankfully has a favorable outlook on the use of algorithms on the fediverse.
It’s easy to forget that algorithms were originally added to social media as a benefit to the user. It’s easy to miss the content you want to see when you follow lots of accounts.
I would love to see an algorithm that allows for manual adjustment. Kagi search allows a user to raise or lower the priority of individual results. I can’t think of another platform or service that does this explicitly. Most do it implicitly through various forms of engagement.
That sounds cool. We’ve been needing some sort of federated video sharing platform.
There is peertube for years. A simple plugin for shorts would be enough. But hey, I don’t judge the way it goes. Good luck for loops. I hope dansup has some time for his other projects, like sup. Never heard about sup again, is it dead?
I vaguely remember it being put on hold for some alignment with European privacy laws or something along those lines.
Open source Tiktok?
Good.
I hope there will be a filter to filter by content format
What are the chances we get a Fediverse YouTube eventually?
You mean Peertube?
I was aware of it, but is it part of the fediverse? If so, I was unaware. I thought it was around before.
As far as I can tell, it’s always used ActivityPub for instances to communicate with each other, with video streaming mainly facilitated using WebTorrent on the backend to keep use of the host low.
Sean is really going the wrong way integrating that with pixelfed as the reels counterpart would have been better as the pixelfed servers aldready support big enough file size for a 30s or a minute videos . Like lemmy definetly has got a chance coming out on top against reddit but pixelfed will never overtake or be on par with insta if it goes this way people need groups and calling features etc. because insta is mainly about being a social platform rather than sharing pics my guess would be even tho pixelfed aspires to be insta which is pretty clear it does based on its ui ( don’t argue its supposed to be a place for nerds to post pics because atleast the dev doesn’t see it that way and is often advertised as a fediverse instagram alt ) but it’ll stay as only a place for nerds to post their pics and eventually be unactive and die out .
The devs on fediverse seem unable to grasp the concept that it is not big enough for a thousand projects to spring up suddenly and will do more harm stretching the aldready thin userbase than good . Maybe try making the aldready good softwares than reinventing the wheel ? Its not like we can’t fork it in the future no need to have a couple of alternatives built for just incase scenario or anything . But still i hope it works out well enough for the devs and gets some of Americans as tiktok is banned there no i think and maybe there is a market for alternative .
It’s dangerous to go alone! Take this:
.....,,,,,,I’m not sure why you’ve been downvoted. You are completely correct. There is a trail of partially finished projects. Pixelfed itself is in beta after years and years (and competitors seizing the opportunity with more polished products) , there’s SUP, Loops and that fedi directory to name three more.
What competitors to pixelfed are there?
…pixelfed aspires to be insta which is pretty clear it does based on its ui…
So this is why the Pixelfed UI is pretty bad?
Try different clients if you think the default one is bad .
Got any recommendations?
Sure try pixeldroid from fdroid . Not really a recommendation as in i like it but a recommendation as this is an alternative UI you may or may not like .
This is very interesting, but makes me wonder if nonprofit initiatives will be able to host video sharing platforms in the fediverse, because of the server damends of it.
It certainly is possible but the more fancy stuff would need to be locked behind some paywall. Hosting 720p30 content is very doable but if users have the option they will of course upload everything and their mother in 4k60 at which point gl with the storage costs alone, nvm the need for hardware acceleration.
It’ll definitely need some kind of quality enforcement to make hosting work. It’d be really useful if the app would automatically transcode to the server’s preferred quality when uploading, using the uploaders device. If the server has to transcode all the video the compute costs could get astronomical.
There should be no reason not to transcode onboard, right? Modern mobile devices could probably process video no problem and then the upload would be smaller and quicker than sending the original. Only issue might be long videos, but I think there’s a case to be made that these types of platforms should have a firm duration cap of only a few minutes tops.
Seems this would be a great way to drive financial support.
I’ve reprocessed my own video down to 480, and am surprised at how much stuff is perfectly viewable at that resolution, even on larger devices.
It’s too easy to save stuff at a higher than needed resolution.
This is only semi-related but do you think it’s possible to use a neural network to optimize compression?
Edit: wait I just realized I’m stupid this is 4/5 of Nvidia’s AI ventures.
everything and their mother in 4k60
Here before someone makes a “yo mama” joke
yo mama’s bitrate is so large that nobody’s video decoder is up to the task
Been following this for a while and can’t wait to see what it brings to the table. :-)
@PlasticHam nice!
Thanks! :-)
yaayyyyy! :-)
