Hey netizens, please, *please* stop doing unpaid marketing work for YouTube by embedding their videos on your websites. I understand continuing to upload videos there to reach the large audience that already use it, but that doesn't mean you have to link to them. If you also upload your videos on a instance, or some other more independent video host, and embed those on your sites, you can help to grow the audience for ethical video hosting networks.
instances.joinpeertube.org/

@strypey Another option: host the videos yourself.

It doesn't have to be complicated, just put the file on your server and link to it with a <video controls> tag.

@alcinnz
> on your server

I don't have a server. Most net users don't have a server. If publishing text on the web, let alone video, required having a server, the net would still be a niche medium for geeks like it was in the early 90s.

@strypey Well, I was assuming that if you had a website you had *some sort of* server you could upload media to...

And how I wish browsers built in one of those peer-to-peer DHT protocols? That'd partially adress the need for a server, though you'd still need one for reliability.

Follow

@xj9 which bits are are those, and what do you propose replacing them with (setting you up to put in a plug for your decentralized architecture proposal - go for it!)
@alcinnz

@strypey @alcinnz

elm to start with. strongly typed FRP in a component oriented architecture. the language isn't so important, but the typing discipline is important for eventful systems. think "kafka meets elixir with a strong dose of KISS". tracker is supposed to be cross platform, currently targeting web. once we have some of the ui patterns worked out I want to shift gears to develop a native platform in chicken scheme. we'll be borrowing a lot from racket to make a solid native platform for developing languages and make elm compile to efficient machine code. the scheme platform will bind to some cross platform libs for graphics, thinking Cairo and OpenGL ES right now. protocol-wise I want to keep the component architecture so its easy to swap out networking stacks. currently planning on building on top of bit torrent and couchdb, but future versions will use named data networking (NDN). I want to keep the language diversity in the system to a minimum so you can work anywhere with the same basic tools: Elm, Scheme, C. I do want to be friendly to experimenting with language concepts, so the scheme based language platform is going to be important. I want to change where we draw the line between engineers and users. applications as we currently know them should be replaced with easy to use programming tools that users can use to take the replicated datas and transform them into comfortable interfaces that work precisely how the user means them to without interference from developers.

https://xj9.io/posts/2019/programming-for-the-next-billion

https://walkaway.wiki/Tracker/

https://git.sunshinegardens.org/xj9/x/bbnet/

https://git.sunshinegardens.org/xj9/g/tracker/

@xj9 @strypey Hmmm, I don't know if I have my mind entirely wrapped around your ideas here.

But it does sound like something I enjoy thinking about hypothetically, even though I'm busy exploring other ideas I find personally more interesting.

The best of luck!

@alcinnz @strypey

kinda hard to condense them into this format. i'll try to expand some in future blog posts :)
Sign in to participate in the conversation
Mastodon - NZOSS

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!