@maiki I think this person was me... I think what I was saying was "it turns out that Slack mostly showed that nerds had the right design with IRC and XMPP MUC (which, while less popular, can do the things Slack can do that IRC can't) right all along, but what we didn't do was make it accessible to the general public, which is what Slack managed to do really well."
current status: exhausted :\
An Introduction to Petname systems http://www.skyhunter.com/marcs/petnames/IntroPetNames.html
whoa all the gmail users returned to my xmpp roster suddenly
I went from saying productive things to absolute shitposts in 10 minutes
time to get off the computer huh
🅱️ 🔒 ⛓
I'm not sorry
@rowan Speaking of Bitcoin: it is impossible for Bitcoin to become the world's e-currency for every transation. The planet would burn in the process. However it may be a *backing* as a kind of "e-gold" for *other* cryptocurrencies (and we probably do need one of those), and indeed this is where I expect it to go. (Even most bitcoin devs seem to think so, hence the work on sidechains.)
@rowan One justification then for Veres One's blockchain is getting a DID for yourself takes the equivalent of running a lightbulb for 15 minutes' power, but then you have that identifier for you for life... and most people will need no more than one or a few. So hearing that I was like "oh okay, that's very different than the concerns about bitcoin's power usage."
@rowan When I started looking into DIDs, I said "why on earth are some of these methods being built on blockchains? Can't we just use a DHT?" And indeed the IPFS based method does use a DHT.
I found out in the case of Veres One, the Digital Bazaar folks indeed started with a DHT, but then they wanted to be able to upgrade keys so they needed to pass around a log, oops they needed an append-only structture, and how to make it resistant against state actors.. oops we invented a blockchain!
For: maybe you need those properties, esp if being able to audit the ledger is critical. Probably not though, but there are some justified usecases.
Against: well you might not want a datastructure that grows forever; preventing sybil attacks can be wasteful of resources (eg proof of work, though not all blockchains use proof of work).
Overall blockchains are over-hyped both for and against because in general people don't understand how they work and what they are/aren't useful for.
@rowan Blockchains are frequently not the answer, but it's important to realize that not all blockchains have the same properties as bitcoin. Blockchains share a lot of properties with a signed git log: an append-only, immutable datastructure. But there's probably "consensus" in there too.