"There can be at most one successful protocol for a given use case."
-
@evan SMTP is great for sending mail but HTTP is perhaps a more modern approach. Likewise FTP. It surely helps if there’s a winning protocol but that can result in being stuck in a local maximum due to upgrade/switching costs. On the other hand, proliferation brings fragmentation. So option ‘E’ - “it’s complicated”
-
@joe I agree, all messaging should be via HTTP, maybe with JSON packets that describe different kinds of objects and activities
-
Evan Prodromoureplied to Evan Prodromou last edited by
I'm somewhat agree.
I've mentioned before how Metcalfe's Law (value of network is proportional to the square of the number of nodes) means that if you divide a population in even halves, each half has a quarter of the value of a united network, and the sum is only half. The math gets worse if you divide in thirds, fourths, or more.
-
Evan Prodromoureplied to Evan Prodromou last edited by
So, a network divided by incompatible protocols will not reach its full potential.
For some people, that might still feel like success. A half dozen incompatible protocols, sometimes bridged and sometimes not, maybe with some dual-stack nodes to spice things up a bit. The medium makes do.
So, somewhat agree, for varying definitions of success.
I think if there's a chance to get the whole network connected, and you can get full flourishing value, you have to try.
-
Evan Prodromoureplied to Evan Prodromou last edited by
I think of the postal network, phone network, email, and the Web. They all have one form of addressing and delivery; one protocol. It took a long time to get there, and a lot of it is effected with bridges, but the benefit has been unbelievable.
-
Evan Prodromoureplied to Evan Prodromou last edited by
If you believe that the medium matters, you have to give it a shot.
-
Ozzie D, NP-hard :bikepump: :vegan:replied to Evan Prodromou last edited by
@evan This is a strong argument! But my instinctual reaction is to ask, "reach it's max potential for what?" What if the network is optimized for delivering something toxic? Balkanization feels like it has benefits.
-
infinite love ⴳreplied to Evan Prodromou last edited by
@evan i see where you're coming from but i think that social networks are not the same as any other communication network. there's an element of community and community management that isn't present when you're just communicating or just publishing. w/r/t metcalfe's law, as harsh as it may sound, some people bring *negative* value to the network. some networks have negative value. there are some networks i don't want to be part of, and some people i don't want to be reachable or to be reached by
-
infinite love ⴳreplied to infinite love ⴳ last edited by
@evan however: i am ambivalent about having this disconnection happen at the protocol level vs. at some kind of access control or filtering/message-dropping level. sometimes it's "enough" to just mute or block or make myself invisible. other times, i don't want to even exist in certain spaces or on certain protocols.
-
@evan I would argue networks filled with people with incompatible values would not reach its full potential. Then you’re faced with incompatibility and fragmentation. One could move the goalposts and call it a success by still having those same people using the same protocol but not within the same networks but then I’d argue how is that reaching its full potential compared to if they were using incompatible protocols?
-
@damon I don't think that's how global communications networks work.
-
Evan Prodromoureplied to infinite love ⴳ last edited by
@trwnh right. You've managed to learn how to avoid web sites you're not interested in, even though your browser uses HTTP.
-
@neumann347 it's a huge slowdown on adoption of a medium. I absolutely agree.
-
Dr.Implausiblereplied to infinite love ⴳ last edited by
@trwnh @evan But there is no "just communicating" or "just publishing". Just because the checkpoints that happen, the crank file on the submissions desk, the extra bin in the mailroom, all the other selection mechanisms that form part of that "protocol" are hidden from view, doesn't mean that community and its management doesn't happen.
The difference is in one velocity, not one of kind.
-
infinite love ⴳreplied to Evan Prodromou last edited by
@evan Well, no, websites are a publishing medium and I don't go around visiting every website in existence, nor does the existence of some website imply anything about *my* website. But there's a social salience in saying "Oh, I'm not on [network]" even if [network] has billions of users. It's like living off-grid -- you explicitly choose not to participate in the primary network, and build your own. In other words, if there was only one protocol, it would become necessary to invent another one.
-
infinite love ⴳreplied to infinite love ⴳ last edited by
@evan In much the same way I might say "I don't have a phone number" or "I don't have a Facebook account", it's not about the total user count. The fact is that me being on those networks actually provides net negative value in my life.
-
Evan Prodromoureplied to infinite love ⴳ last edited by
@trwnh Why? You can also say, "I'm not on email, I don't have an email address." You don't have to invent a new email protocol or use X.500 or Banyan Vines in order to be off the Internet email network.
-
infinite love ⴳreplied to Evan Prodromou last edited by
@evan New protocols are born when there is a reason for them to exist. In a world where SMTP is everyone's solution for transferring bytes from one place to another, all it takes is for at least two people to say "hm actually no we can do something else". Consider those apps that exist solely for a person and their monogamous partner. Could they be standardized onto a single networking protocol? Well, they all use TCP/UDP (probably), but that's an implementation detail, and the network is above.
-
infinite love ⴳreplied to infinite love ⴳ last edited by
@evan So I'm inclined to say that having a single "network" in application layer or higher is not really a possible or desirable state. There are always reasons why someone will want to establish an alternative network. At some point, some need arises that makes someone actually do it. The most you can hope for is a broad majority on the one protocol of your choice, which handles ~everyone's use case sufficiently well. Or ~everyone you care to talk to, anyway. And most protocols aren't there yet