Live data from Hacker News

Ask HN: Is it time to resurrect a Usenet clone?

news.ycombinator.com

81–90 of 330 posts

Re: Ask HN: Is it time to resurrect a Usenet clone?

#81
post #71
post #50

Earlier quoted context omitted.

The problem is, a lot of the platforms that I've liked over the years have been vacuumed up. I still try to use different ones, but it's a lot harder.

Personally most my moves away from online platforms over the last 3 decades years have been been because community owners have massively fucked up. I think Digg did it worst.

I miss BBSes, Usenet, EFnet, Puzzle Pirates, and a few web forums.

They all kind of exist, but they were mostly supplanted by bigger, more centralized things. All of them had a part in their own demise, too.

Then in turn those bigger, more centralized things have tended to screw things up.

Re: Ask HN: Is it time to resurrect a Usenet clone?

#82
I've been thinking a lot about this. Us engineers periodically think some version of "I'd like to put a text box on the internet, and hmm, maybe, yeah, some way of storing and retrieving what someone else put in it". Sounds like another endeavor in a grand tradition of such endeavors.

Everything else past that has insane ramifications. Do people put threats in that box? Do people put personal information in that box? Do people link to CSAM in that box? Do people link to radicalizing videos or viral misinformation in that box? Do people use this to harass and stalk others? To manipulate stock prices? To manipulate elections? The answer is "yes" to every single one of those things. You used to think it would be nice if people used your service, but now you realize to your horror you run a Nazi bar [0].

This is the question of the web now: how do you deal with Nazis? You might disagree, you might think it's about centralized/decentralized or analytics or JavaScript, but the only reason those things are relevant is Nazis.

Don't forget there are two kinds of Nazis. We probably immediately think of the American History X Nazi, but don't overlook the respectable Nazi, the polite, well-groomed, non-tattooed Nazi. The Nazi with a bachelor's, or even a law degree or an MBA. They'll use your service too, in ways that might seem like speech you should allow [1] but is just the first Nazi walking into your bar.

We gotta stop thinking "I'll put a text box on the internet" and start thinking "I'll figure out a way to keep Nazis out of text boxes on the internet". This sounds like a huge task but, and I want to be clear: this is _table stakes_ for any new service. The thing you're building isn't a text box, it's a Nazi-free text box. This is the bare minimum.

Then, and only then, can we start building tech that starts to make us feel good about ourselves. That starts to address the loneliness epidemic, or the bonkers decline in adolescent mental health brought on by social media (don't--ironically--@ me on this one), tech that inspires us and brings us together instead of tech that isolates and depresses us, tech that respects us and our privacy instead of surveilling and exploiting us. We have to start thinking about what our tech will do to people when people start to use it, from 10 users to 1000 users to 1,000,000,000 users. We need to start taking responsibility for the things we create, it's no longer possible to just throw your hands up and say, "look I just put a text box on the internet, how was I to know blah blah blah". We have to know better.

[0]: https://www.upworthy.com/bartender-explains-why-he-swiftly-k...

[1]: https://kottke.org/16/11/the-14-features-of-eternal-fascism

Re: Ask HN: Is it time to resurrect a Usenet clone?

#83
post #65

https://join-lemmy.org/ Lemmy is a federated Reddit clone. It already works and is available. As far as I know currently the best alternative. Please, join a smaller instance currently. The biggest ones are already overloaded.

The political views of some of the major developers (tankie, pro-china, castro-profile-picture communists) has put me off lemmy, particularly because they're (allegedly) removing anything critical of Russia/China and promoting genocide denial on their instance.

Re: Ask HN: Is it time to resurrect a Usenet clone?

#84
post #68
post #66

Earlier quoted context omitted.

I wish there was something else. For building a coherent community federation is a bug not a feature.

Federation makes it easier for users to join. If you don’t need that, there are plenty self-hosted alternatives. For example, HN itself.

How exactly? Ever federated project has all this faff about "which instance should I use?". How is that possibly easier.44

Re: Ask HN: Is it time to resurrect a Usenet clone?

#85
post #54
post #37

Earlier quoted context omitted.

> where most of us trusted each other I think the "trust" back then was really just because only people with a certain amount of money could afford a computer and internet access. Most people on the internet at that time were middle class folk with proper day jobs, or in my case, a child of the aforementioned.

Data storage would be a big question and a design goal for something new and decentralized. Back then it was mostly text and files were much smaller. Now average image floating around is the size of a cheap USB stick.

A "usenet 2.0" that only does (short) text and links/embeddings for media (to youtube etc) would already be a very useful thing, notwithstanding the risk of take-down of controversial media (freeloading on Big Tech infra for the 99.9% that isn't).

Re: Ask HN: Is it time to resurrect a Usenet clone?

#87
Lots of people keep trying but they have a ton of problems.

I have done a lot of research and thought about this quite a bit.

* For a long time P2P automatically meant crypto. I'd rather Google use my data to get richer than Bitcoin bros waste electricity to do the same.

* Many were laboring under the delusion that self hosting is a sane idea. Anyone who's ever self hosted anything knows it takes actual effort and sometimes money, and probably won't be as reliable as Facebook.

* Protocols that are immutable-first are bad. You should at least be able to try to delete your old embarrassing posts.

* Normal people do everything on mobile. It's just so much better of an experience than any desktop OS. Add a keyboard and Android is the perfect device for basic stuff. P2P devs don't usually support mobile.

* Bandwidth is a thing. IPFS still uses too much of it. Helium used way too much last I checked. Disk wear and CPU is also a thing.

* Centralized servers are really good for performance. I think that an ideal platform probably wouldn't use any kind of DHT or anything, it would use a combination of local discovery, and semitrusted servers that are separate from your actual identity.

* Federation as seen today isn't that meaningful in practice. If my identity is tied to a server, there's only one server I'm going to consider choosing. The biggest most popular one that might stay around the longest. Might as well just do centralized FOSS at that point.

* You need a killer app to make normal people who aren't in it for political reasons use your thing. If you don't have that, your app will become KaczynskiNet, and I'll just ignore it, because... nobody I know will want it and I probably won't be able to convince them. Free speech is a good thing, and the "controversial" platforms might have a valid place.... but I'm not going to spend much time on platforms almost entirely full of communities I'm not part of and don't agree with, there's not enough hours in the day for that.

* Endless scrolling is evil and sent to instantly kill any idea of persistent communication. It does not belong anywhere but a news feed. Even pagination isn't ideal, we should probably be using time-based "Next N records after $date" pagination.

* I believe it also may subtly cause low grade anxiety by removing reference points and causing content pages to feel like they could change at any moment

* The current social media model developed around centralized servers. Things might have to change a bit.

* Heirarchal nested threads do not mirror real life conversations, and complicate the UI a lot, and make it easy to miss stuff if you aren't familiar with how stuff works.

The protocols involved are Not That Hard. It's been done several times, just never polished. Retroshare, BitTorrent, SyncThing, Jami, and Dat between them probably have 100% of the technology needed to make it work.

Possible solutions for a semi-decentralized model preserving some of the advantages of decentralized:

* Private Key based identity, no unique human-meaningful names at the protocol level. Just use some other platform to actually find your friends instead of trying to replace the whole internet.

* Your data is on your device, and on any server you are registered with and set up sync with. They can be free or paid, not the.protocols problem.

* Actually finding where to look for your friends stuff is not the protocols problem. You have LAN discovery, you can manually enter a server, and the clients know to try the common servers. Add whatever name servers or DHT extensions you want to that.

* Discussion threads could just be best-effort sync. Maybe you don't see every message if the network was partitioned but you probably will eventually.

* Threads could be "Hosted" at all servers that a user who made them is registered with.

* Messages contain lists of what other servers you are registered at, and you get a certificate from all of them proving it.

* That means the client can always find your other stuff somewhere else even if one server goes down.

* You can always choose to mirror a thread on all your own devices and servers for archival, and make it visible on your profile, it stays in sync, or you can freeze it so it doesn't, or you can copy it to a brand new thread.

* Posts are editable and deletable.

* Tombstone problem solution: Content is truly gone if you know the full list of servers a user is synced with, all those servers are up to date with their device, and none of them have it.

* You can make a thread that requires membership at a specific server to post at, for spam restrictions.

* There are no categories. You just share your thread to other threads.

Re: Ask HN: Is it time to resurrect a Usenet clone?

#88
post #7

Two years ago this is where someone would chime in with a blockchain-based solution, paying moderators with usenetcoin.

Unrelated to the root post, but I’m so glad that time is over. There was a ~1year period where it was nearly impossible to have a conversation about solving something that didn’t involve someone scheming about how to reframe the problem as an economic model, and then drown out everyone with some half-baked tokenomics concept as a way to “solve it all!”. It really felt like a step backwards in critical thinking for pe…

> reframe the problem as an economic model,

Every social problem can be legitimately solved by throwing more money at it[1]. That is, effectively, an economics model.

Corollary: Every economic solution can be improved by removing blockchain from it :-)

[1] Sometimes you have to throw the money at people who will spread your solution, sometimes you have to throw the money at media to spread the solution, sometimes you have to throw the money at campaigning, etc.

Re: Ask HN: Is it time to resurrect a Usenet clone?

#89

Earlier quoted context omitted.

Unrelated to the root post, but I’m so glad that time is over. There was a ~1year period where it was nearly impossible to have a conversation about solving something that didn’t involve someone scheming about how to reframe the problem as an economic model, and then drown out everyone with some half-baked tokenomics concept as a way to “solve it all!”. It really felt like a step backwards in critical thinking for pe…

I still think you can use economics to model pretty much any human interaction. The problem is, you can't just slap a token over everything and call it a day – the incentives that drive people are way more intricate than that. On that note, I think many systems will function better without blockchain. Trust was driving the internet (and any other distributed human networks) for the long time – long before Ethereum ca…

You can also model any problem as a hydraulic circuit if you want.

There are other incentives like fun, passion, and anger that will break any economic model.

If HN had some internal-economy model where you were cycling around karma as if it was currency, paying upvotes into new posts to get a return back for your “early upvoting” if they do well, giving you more capital to spread to other new posters… I know exactly how the dynamics of this site would change for the worse. People just post here (free work!) because it’s interesting. How irrational!

There’s some merit to the idea that you need aligned incentives. I’ll give you that. Just doesn’t always need to be directly boiled down into units of exchange.

Re: Ask HN: Is it time to resurrect a Usenet clone?

#90
post #59

Earlier quoted context omitted.

I think a major component of the difference between contributing then and contributing now is friction/skill/interest. Back then, you had to be specifically interested in learning how to get connected, which was a niche endeavor, while the masses were oblivious or too cool for school. Now, you can simply join almost any online community using baseline skills that an enormous fraction of Earth's population already has…

Yeah, it's not a technical thing. It's that the unwashed didn't have access/weren't even aware. Sure, you can do the same thing today given explicit gatekeeping and moderation.

> weren't even aware

I guess that aspect can't really be replicated anymore. Back when information moved slowly (in person, telephone, print) and broadcasts only included things with mass appeal (TV/radio ads, etc.) knowledge of a small but high quality community would remain tight for a long time, growing slowly. But the cat's out of the bag now: even the most obscure thing only needs to be posted in some place with tons of eyeballs (which simply could not occur at scale back in the day), if it's good it's upvoted to the top, and immediately it's no longer obscure.

Post reply on HN