368
this post was submitted on 11 May 2024
368 points (91.8% liked)
Technology
59589 readers
3300 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I don't understand how any of these visions fundamentally differ from Mastodon.
Decentralized? Yep. It's got no center. Open source? Yep, you can fork it and make your own if you want. Unmoderated? Sure, if you want that, you can set up an instance and host whatever illegal content you want. You'll have a lot of legal problems and most people don't want it, but the option exists.
Is there any point besides money and crypto bullshit? If you want to post short comments that your friends can subscribe to that isn't controlled by a big corporation that gives your data to the government... well we have that. It exists. It's pretty okay. Go use it.
The biggest individual difference is that bluesky makes identity independent of the hosting server (via cryptographic keys) and makes content location independent of the hosting server (via content addressing).
And these features together also enable more efficient caching and propagation in the network as well as enabling features like custom feeds and 3rd party moderation tooling which works the same independently of which server you're on. So Bluesky can give you a better global view of the network and more efficient communication between users on many different servers in the same thread.
Ironically enough, Jack's other favorite place Nostr (which is built as P2P with repeater nodes) is also adding moderation tooling similar to that in Bluesky (labelers making use of the content addressing and account key ID) to flag stuff
First, thanks for that explanation. That's interesting.
Is there a good place to learn more? I can see why having custom feeds and 3rd party moderation tools are good, but I still have a lot questions.
First, is there a genuine benefit to dissociating a users identity from their server? I think the connection between users and their home instances are a brilliant innovation. They seem to bring village culture back to the internet. They help people associate within networks below just the global level. I think the atomization of people online has been a part of why there is so little trust.
Bluesky is open source and have a site for documentation
Splitting off identity means you can bail and take your friends and post history with you when a server either goes down, gets hacked, or if the admin goes insane, or if it gets freenoded (hostile takeover and impersonation)
On bluesky the closeness comes more from the personal connections plus the choice of feeds
I hate the term web5 but the tech stack described here is much better than anything else I've seen. https://developer.tbd.website/projects/web5/
You don't need to go crypto to get there though, IPFS and similar exist and work. IPFS itself is kinda slow, but Iroh is aiming to be a more efficient alternative that solves similar problems. There are also protocols based on BitTorrent.
The way these work is basically:
Then you build stuff on top to keep everything in sync. No servers, aside from the initial connection, which means minimal risk of anything ever going down. If relays go down, anyone can set up another and people reconnect.
The problem is that step 3 is quite complicated, and there are a ton of technical complexities to synchronizing information at scale w/o a central authority. Mastodon/Lemmy/ActivityPub gets around this by having each node (instance) be a complete copy of everything that node cares about. You get a ton of duplication, and eventually that means costs pile up. With a proper decentralized system, there doesn't need to be nearly as much duplication since you can always hop through some peers to find what you need.
You never do. Its only use case is a payment system for online crime. And even for that many criminals prefer gift cards because it's such a hassle to explain crypto-tokens to your victims.
It's useful for anything online where cash would be useful. So paying for services, money transfers between acquaintances, donations to charity, etc. It turns out cash is useful for crime, and privacy-focused cryptocurrencies work like cash, hence are useful for crime.
Don't buy it as an "investment" or sign up for services to earn it, but it is useful for non-criminal things.
No, it's not useful as a cash substitute because of its hilarious inefficiency.
That's not necessarily a given. Ethereum, for example, transitioned to proof of stake instead of mining and seems to have reduced electricity use by 99.5%. I'm not exactly sure where that number comes from, nor do I know a good way to compare crypto to other systems (e.g. do we count all the energy used by banks?).
But what I do know is that Bitcoin kinda sucks from an energy perspective, partially because they limit the number of blocks (e.g. buckets of transactions) per day, so mining is more valuable than on a currency with no such caps (e.g. more demand to mine each block = more miners = less efficiency per mined block).
What seems to be true is that cryptocurrencies have a large upfront energy cost due to speculation, and that plateaus as it hits a certain carrying capacity. So crypto scales decently well, and if you do proof of stake instead of proof of work, it seems to scale even better.
And then we get into the issue of where your energy is coming from. Since cryptocurrencies are global, they can be done anywhere energy is cheap. For example, daytime purchases can be done using excess energy in an area where it's night. For fiat, that energy use is more local, so you're more likely to process a transaction during peak energy use (afternoons), thus higher energy capacity needed. It's a really complicated topic, and I'd love for someone smarter than me to break it down.
But since it's so hard to calculate, there's a lot of bad information, which leads to unnecessary and unfair criticism from people who don't see value in cryptocurrencies. If you ask a crypto bro, they'll point to the massive amount of power used by financial institutions, and if you ask someone who's against cryptocurrencies, they'll compare POS and minor processing use by credit card companies to an entire Bitcoin block (which has lots of transactions). I'd really like to see an updated, neutral look into it, because all the information I'm able to find has huge holes in methodology.
But all of that is kind of irrelevant to the discussion about whether it's useful. If it's not useful, any amount of energy use is wasteful, but if it provides value, there's certainly an amount of energy we're willing to spend on it, so what exactly is that amount?
Also Ethereum is extremely inefficient compared to conventional tech (like just a database). All you need is to realize that complete trustlessness is impossible to understand that a distributed ledger has no problem to solve. And that's why there is no practical application after all these years.
But energy use isn't as simple as measuring transaction efficiency, there's a lot more to a currency than storing who transacted with whom. There's:
Or in terms you've used, someone needs to maintain that database, that database needs to be in some facility, and someone needs to audit the database. All of that is baked into cryptocurrencies. Yet the comparisons I've seen either account for way too much (e.g. bank tech support), or not enough (e.g. only POS and network costs).