In this talk, Andreas looks at scaling as an ongoing process. Using examples from the history of the Internet - which has "failed to scale, gracefully, for 25 years" -- he discusses the scaling debate and looks at how Bitcoin will scale over the next several years.
(copied from the video description)
Imo, this is as relevant as ever (this talk is from 2016!) and very related to @benthecarman's post about Liquid not being a scaling solution.
That the internet also "failed to scale, gracefully, for 25 years" gives me the most confidence that bitcoin "failing to scale" is a GOOD thing. We won't build scaling solutions if there is no need.
So that there is need is a good thing.
Now please tell me in details why I am wrong :)
Great find.
Some interesting points around minute 23:
"There is no spam transaction. There is no such thing as illegitimate transaction"
"There are only transactions that did get mined and transactions that didn't have enough fees to get mined"
"We will have the scaling discussion every year for decades"
"Every year we will fail to scale for the next application"
reply
Every year we will fail to scale for the next application
And this is the most interesting point of all these very interesting points to me.
This is how we're going to win in my mind: by failing and failing and failing.
Until everyone realizes: we've been wining and winning all along, lol
(this was just a typo first)
reply
Exactly, people have this idea that bitcoin needs to scale to do this and that but really it's gonna be simpler. Not easy but simple. And most of these "innovations" are scams and will fade. Only the strong survive which means only the best scaling solutions that have sustainable demand will make it.
reply
The problem is that failing to scale on the internet means that we throw a NAT in front of everything and need to upgrade some cables.
Fialing to scale bitcoin means we could end up with the fiat system all over again where everyone needs to use trust based solutions. Since consensus is global, soft forks are a lot harder to coordinate, much different than updating to fiber optic
reply
The problem is that failing to scale on the internet means that we throw a NAT in front of everything and need to upgrade some cables.
I disagree. I don't think NAT was the only thing that was necessary. Afaik, NAT only became a thing when the Internet failed to scale so hard, that we actually did run out of IPv4 addresses because no one realized in the early days that we would connect even our toasters to the Internet and not just 10 machines (need to find that hilariously wrong prediction but I believe it was from Robert Metcalfe). And everyone realized that we fucked up so hard with our predictions about Internet adoption that also the migration to IPv6 is going to take way too long. So we're basically stuck with IPv4/IPv6 dual-stack. But the Internet still works, right?
Fialing to scale bitcoin means we could end up with the fiat system all over again where everyone needs to use trust based solutions. Since consensus is global, soft forks are a lot harder to coordinate, much different than updating to fiber optic
Interesting take. I haven't looked at it this way yet that scaling bitcoin is going to be much harder than the Internet. Need to do more research regarding that. I would say I have more knowledge about Internet history than actually about Bitcoin history, lol
But I still believe that "failing to scale" is a good thing, do you not?
reply
But I still believe that "failing to scale" is a good thing, do you not?
it is definitely champagne problems, but we don't want champagne problems when we only have a few million users
reply
This is of my favorite lecture. At first I saw this video than I read transcription in the Internet of money book
reply