I think the “near zero latency” is still a valid UX concern.

Do you know if it’s possible to setup a webtorrent library in a way that it is forced to try fetching from HTTP the first couple of chunks and only after a short delay fire up the lookup for seeders?

I wonder if it would be a proper solution to fetch the first X chunks of a video from a blossom server, possibly at a lower quality, and then switch to something else if the user continues on that video instead of swapping again.

I mean, that should give you both low latency withoug giving up on lower distribution costs, just need to work out the complexity cost of such a solution that is.

I've seen similar services but I wonder what does it imply?

I mean, selling a managed server is great but it sounds like the biggest selling point is “hey you can pay ICANN to give you a pretty name on the web and we do the rest”.

Ok, I guess my beef is with promoting central points of failure such as DNS, into a “properly decentralized” protocol such as #nostr

You see, the “fediverse” ties users into a single instance because DNS is integral to their protocol so “whatever” but nostr is truly permissionless, provider indepentent accounts, everything great about p2p without the bloat (ssb I’m looking at you).

Many projects do both ways, the “local” flake on the repo serves as a source of bleeding edge package and “blessed dev env” (specially useful when combined with direnv, because you just git clone, direnv allow and then get to business) while nixpkgs hosts the lastest release/stable version.

It welcomes testers without alienating everyone else.

For me most of the value from flakes came from two things:

- Coherent setup across multiple computers (including home-manager, nix-darwin and nixOS of course, be it on servers or desktop)

- Reproducible dev envs without host contamination: this one was a blessing when working with a team on proprietary stuff with a bunch of very specifc deps/setup/etc.

I guess nix “alone” fixed enough shit for nostr:npub1xtscya34g58tk0z605fvr788k263gsu6cy9x0mhnm87echrgufzsevkk5s that the time investment to get into the flake train may not sound as appealing as it does for the rest of us.

Heck, I myself was a little hesitant to jump into the flakes train because of the “experimental” feature flag it’s attached to.

am I doing this right?

nostr:nevent1qqs2w0df6qdflz90kexuvm464gs2q6vl44sqrh8hjnj2wgk4kdx3kmczyqewrqnkx4zsaweutf739s0cu7et29zrntqs5elw70vlm8zudr3y2qcyqqqqqqgpr9mhxue69uhhq7tjv9kkjepwve5kzar2v9nzucm0d5c4rcny

True, but I think you need to open a LN channel for that to work.

Might be worth looking into Alby Hub and ZEUS.

What if instead of tethering your account to one set of very specific servers you could broadcast some delegated work (along with some sats) that any server could do for you? Like a Bitcoin tx.

My cheap ass is using Wallet of Satoshi simply because I had some sats laying around from playing THNDR games in the past.

Actually, that would require prior setup to the downtime event.

I don't think the added benefit is worth the decrease of resilience against censorship/servers going down for whatever reason (i.e. some ISP intern causing some BGP fuckup somewhere).

woah, we have like 3 #nix users on #nostr now

nostr:nevent1qqsxml3urjah7n9jzg70y3j6lnq9jgc9ggh4wwh5hehdkq7h4mepdvczyrhus0cperanp80jezrxhrreynxgkmc9szhamcwku9hzkcg8c2rzcqcyqqqqqqgudh4su

I don't think so, despite being experimental, flakes are widely used by the community.

ChatGPT is coming after C-level jobs now.

Judging by the hashtag list, engagement bait/“SEO”.

I wonder if anybody has written a guide like “how to vibe code for programmers”.

For instance, I gave up trying to keep up with AI news, mostly seem to never get anything real done when using AI, yet, somehow, there are people here on #nostr delivering opensource libs/tools through the magic of “vibe code”.

So I, despite being on my late 20s, feel really ancient, like am I not able to adapt anymore..?

❎ fuck you money

✅ self intercourse currency

Have you tried ZEUS? People say it’s self custodial embedded node within an app, and for some reason it should work despite not being always online.

I don’t know about videos but it would be nice if we converted image files into JPEG XL, because it has inherited a party trick from FLIF where you only need to i.e. download half of the file to have half of the resolution.

Should save on bandwidth and file storage, since you don’t need redundant tumbnails and alternative resolution files.

https://flif.info/animation.html

Why is it that media shared on #nostr relies on “some dude’s HTTP server” instead of, IDK, distributed over IPFS, WebTorrent, MetaLink or something along those lines?

It seems that the older a post is, the less likely it is for the server hosting the linked media to still be up, so unless there’s some kind of The Internet Archive bot crawling Nostr, these are lost for good.

I haven’t looked into the Nostr protocol yet, so I’m assuming media is attatched to posts using plain links, without some metadata including the file hash, which leads to another point of concern: a rogue media server could silently change the linked media.

I would love to be proved wrong though.

Alternative headline: “AOL still exists”

Now I feel called out because of the state joke I posted earlier.

A functional programmer, an anarchist and a nixOS user walk into a bar.

These are all the same person and he won’t stop talking about state.

Oh god please make him stop.

I believe “cumbersome” is the keyword here, unfortunately.