[ebiester]

Consider - why did Discord or Slack win over IRC?

It turns out it's very slow to evolve a protocol. How long did it take for IRCv3 to handle channels having persistent history? How about channel takeovers via network splits? We knew these were problems in the 20th century but it took a very long time to fix.

Oh, and the chathistory Extension is still a draft! So is channel-rename! And account-registration?

And why is it still so painful to use Mastodon?

That's but one of many examples. Consider how the consolidation of HTML and HTTP clients was the only way that we ended up with any innovation in those services. People have to keep up with Chrome who just does their own thing.

I want to want a decentralized world governed by protocols, but good software that iterates quickly remains the exception rather than the rule.

[gorjusborg]

All you've said here is that you (and many others) have shown in the past that they've valued convenience and rapid feature development over freedom and stability.

That is good to understand, but when that trade starts causing issues, it is important to remember that there was a trade made.

We aren't as stuck as we think we are, unless we decide not to reevaluate our past choices.

[Gigachad]

Yes, essentially everyone on the planet was willing to trade some freedom for chats that work on mobile or could send images.

Matrix has shown how incredibly difficult it is to make a modern service in a decentralised way. Requirements like preventing spam become immensely difficult.

[jasonvorhe]

If Matrix has shown anything, it's that not everyone should design protocols:

https://hanez.org/document/why-matrix-sucks/

https://forum.hackliberty.org/t/why-we-abandoned-matrix-the-...

https://xn--gckvb8fzb.com/giving-up-on-element-and-matrixorg...

[Xss3]

Preventing spam may not be possible for much longer without verified IDs considering how advanced ai agents are.

Do any fully trustable ID validation services exist? Ones that verifiably never store your ID but just a validity status for a given ID on a blockchain?

[Sayrus]

Assuming you want ID verification, why would you need a blockchain? Your identity is deeply linked to who you are and we have identity documents and trusted entities to provide them. These entities can absolutely act as a third-party to verify who you are. This can happen with several different parameters: whether your identity is provided to the site you are using, whether the site your are using is known to your identity provider, whether identities across sites are identical or only linkable by the trusted party. But in all those examples (that are currently implemented by some countries), blockchain is not a requirement.

Assuming you don't want actual ID verification, the choices are even larger but with different trade-offs.

[drumdance]

In theory yes, in practice it requires lots of different government services to get on the same page. How do you verify a state ID? Usually the DMV. Have they released an API endpoint for that? Almost certainly no. What if instead you're using a passport? Then the federal government needs to do it. What if your passport is from a country with weak government that doesn't have a lot of capacity?

And of course governments attract hackers because they tend to not be up to date on security best practices.

A single abstraction layer on blockchains allows more developers and security experts to contribute and innovate.

[kajaktum]

Preventing spam is as easy as gatekeeping. We should be bringing it back. Perhaps there should be multiple layers of social media. There’s deeper and deeper level of authenticity as you go deeper into the network

[Gigachad]

Phone numbers + phone number country + account age + behavior can be used to build a trust score. It might not be bulletproof but it cuts down spam enough for now.

Imagine a messaging app for example, a 1 month old account with a Nigerian phone number cold DMs an account in Australia. The likelihood of this being spam/abuse is extremely high. Vs a 5 year old account that mostly messages mutual contacts cold DMing an account in their own country.

In many countries, phone numbers are a proxy for ID and are difficult to get without having a local ID. The countries which have not secured their phone number system will be less trusted by spam filters.

[samantp]

Spam is an issue mainly because there are conspicuous meaty targets to be spammed, not in fragmented environments. And a target is meaty for spammers because that target has gathered, more often unnecessary, critical mass (large scale services, broadcast type news /thought leaders/influencers). Else even a small overhead for sending requests will drive away spammer incentive.

E.g. OS exploits were targeted towards Windows, not so much for so many of those Linux distros.

[zbycz]

We can have spam filters for emails without any identity verifications. It is hard, but it should be possible for IM as well, right?

[jauntywundrkind]

There's also this annoying flash perception that wins. As the big companies abandoned XMPP, less people considered it.

It's pretty good today! Lots of things improved a lot! Some big clean ups!

But think of how much better it would be if people stayed woke, if they didn't just throw up their hands call defeat & say it was never going to work. If there wasn't such a bleak rot in our soul, if we could try to play slightly longer games, I think in the medium & long run it would be much much better for us all.

It feels so easy to spread sedition, to project these fatalisms that only big dumb lumbering central systems win. I'm so tired of this bleakness, this snap to convenience as the only perceived possible win. Let the prophecy self fulfill no more, let us arise from this torpor. A little Ubuntu would be ao good for us all. Ubuntu the old saying (that the distro was inspired by) goes: "If you want to go fast, go alone. If you want to go far, go together"

[ljm]

Nobody said how hyper the HT in HTML and HTTP had to be, so here we are.

Oh, TLS also. Encrypted connections over HTTP are trivial.

Arguably this has created far more freedom by making encrypted network traffic default and free. Convenience is also freedom when it comes to accessibility.

[bigbuppo]

Put another way, the services need us more than we need them.

[schubidubiduba]

Short-term yes, long-term it is often the other way around. In many cases, abandoning an open standard for a closed, centralised solution is surrendering to future enshittification for short-lived instant gratification.

[AceJohnny2]

This, by the way, is why Signal isn't federated. Moxie Marlinspike made the same argument.

https://signal.org/blog/the-ecosystem-is-moving/

[627467]

Is Mastodon really hard to use for most people? I guess there's some very specific scenarios it may be.

Also the article presents a false dichotomy in my view: protocols need services to be useful to virtually 99.9999% of humans (or at least they do in the architecture we have built since... email?).

Who uses email without relying on servers? Where is your selfhosted email box sitting on if not in a hosting service?

Even IRC relies on servers for people to talk to. I love to experiment with protocols that do not rely on servers - secure scuttlebut? - but even ssb relied on some seed peer that provides a service to initialize the peering

[m4rtink]

Discord "won" because it provided free voice chat and then also text channels with image upload, all powered by VC money dumping and hosted for you.

Of course it was also clear that eventually the investors will want to cash out & we are seeing the results of that.

[pjc50]

Under-appreciated factor: the problem with decentralization is that it pushes work on to the end user, who is least equipped to deal with it. People actively want centralization of things like anti-spam because it lightens the load. The fact that this gets paid for in insidious ways rather than directly paying for a service causes all sorts of weird market distortions.

Note that Discord doesn't replace IRC, it also competes with TeamSpeak; there's a whole voice and video sub-feature to it. Not everybody uses it but the fact that it's available in the same software was advantageous to the original market, gamers.

[b00ty4breakfast]

Comparing IRC-the-protocol to Discord-the-platform is silly. Apples-to-oranges etc

[caseyohara]

I can't tell if you are replying to the comment or the post because the topic of TFA is literally comparing protocols and services. Discord and IRC are both mentioned in the post.

[just6979]

Pretty sure they're replying to the post that directly contrasts Discord/Slack and IRC.

TFA mentions both, yes, but as a direct example of service/platform (Discord) vs protocol (IRC, XMPP, etc). The comment asks a question that kinda misses the point of TFA.

Discord could be considered to have "won" in that it's got a lot of (new) users and removes some of the limitations of IRC, but that's _because_ it is a service/platform, and comes with all the trade-offs being discussed in other threads here.

Or one could consider IRC to have "won" because as a protocol it simply can't have some of the restrictions possible with a centralized platform.

It's trade-offs all the way down, but protocols will always have fewer restrictions of the kind currently in the zeitgeist, especially decentralized protocols.

[yellowapple]

That's why I'm pretty optimistic about the AT protocol: you get the advantages of app-driven innovation (need a new feature? just define a lexicon for it) without requiring data reliant on that feature to live in that application's silo; the records all exist in each users' PDS, under each users' own control, no matter which applications use those records. And of course, if those features prove to be good ideas, other applications can adopt those lexicons and they're immediately interoperable.

[shabatar]

Totally understand, I am all for decentralized world too. In reality tho most ppl just choose whatever works fast and ships fast and more production-ready I guess, no drafts. Would be great if the world sees an opposite example, by far centralised approach just worked better

[fulafel]

Between IRC and Discord/Slack we had XMPP which almost made it, but then Google etc killed support for it.