Thread
So FAR BE IT FROM ME TO AVOID ADDING MY HOT TAKE but here is my hot take on the crypto takedown du jour 1/n
I hope regular readers are rn clutching their opera glasses, dying to know whether I agree with their predilections or have switched sides and turncoated with the devilish monsters of Web(2|3). Fools of such limited desires! I am more contrarian than you can possibly imagine!
First, I've been reading DSHR's blog for ages--unlike you oiks--so I know the better versions he's written for all of these arguments. I share his pain that it's his hasty talk to the AI folks at Stanford that got the hits, not his thoughtful ponderings over the last few years.
That doesn't mean I agree with his views, but they're /okay/. It's really hard to get any purchase on the real meat of the current state of affairs, and after a while one tires of getting web3 ground beef under one's fingers, and you just work with the gunk you're given.
Anyway, for today's lecture, I'd like to concentrate on one slide, which I hope will show the *class* of errors that DSHR is making here, in the hope that it'll encourage everyone to think more critically -- both about crypto, and crypto-sceptic takes.
And because I'm contrarian, I'm going to choose to defend #chia -- which if you were to run with the surface impressions, and somewhat within DSHR's thesis, is in fierce competition with my own ecosystem of choice, good old, lovely #filecoinā®
DSHR takes a sideswipe at Chia, because he followed it pretty closely in his hunt for alternatives to proof-of-work. Chia uses a consensus system that's based on, oh god I have to explain this in one tweet, that's based on 100GB-sized lottery cards. Phew! Made it.
What you do, is you create, during a fairly CPU-intensive day or two, a number of unique 100GB chunks of data, and then thereafter the rewards you get are related to how many chunks you created versus how many are in the network as a whole.
meanwhile, as you keep poking in to see if you won today's lottery (or rather, one of the 4608 I think daily lotteries), your machine *also* contributes to doing all that tedious decentralized consensus, smart contract executing, ledger maintenance etc
the check-in and processing is pretty negligible -- I set up a chia node earlier this year (due diligence!) on a raspberry pi, and -- yep, I just checked, it's still running, using far less than the synology NAS i have next to it.
Chia is an example of a modern proof-of-not-drowning-polar-bears-in-fossil-fuels-for-fun blockchain. (As is #filecoin, which I work for and is ALSO great, but not really relevant for this discussion.) DSHR's critique is this: Chia sucks as bad as bitcoin, because externalities>
Mainly, because when chia launched it was blamed for a crash in the supply of hard-drives, because Chia fans went and bought up huge amounts of drives, denying them to other people. Also it trashed SSDs.
This is /sort/ of true, though it was very hard to tell whether what was really going on was just the usual covid supply wobbles. All I can tell you was that it only lasted a few weeks, and was confined to the retail channel. That fits with a short, intense, Chia-related burst.
... because, and this is important, there was only a small timeframe when it made economic sense AT ALL to spend money on a new drive to feed your chia node. As Chia was ramping up, there weren't many other chia nodes, so your lottery ticket had a much better chance of winning.
Around that time, you could bet on winning enough chia to make back the cost of buying a fresh drive. It was a tiny window though. And I should know, because I wrote a huge spreadsheet to make sure I wasn't screwing it up.
Full disclosure: I bought about $2K of hard drive, made about $3K, and got to keep the drives, which are now divided between chia, and a few dozen terabytes that I'm putting to good use with linux isos and such.
Are you still with me? Okay, so, after a short while purchasing drives became madly uneconomic. After about 8 weeks after launch, you could really only make any kind of $$$ by using drives you have around. I.e. using pre-existing equipment.
DSHR extrapolates from the boom moment: he thinks people bought up drives, made their $$$, then threw them away, creating e-waste. Meanwhile the price of Chia "crashed" so the whole thing was a waste of time.
In fact, Chia, like many crypto ecosystems, *stabilised*. Check out www.chiaexplorer.com/ -- the network zoomed up to around 30 exabytes (which is insane, btw), and then broadly stayed there. It decentralized sufficiently to guarantee that there isn't a central point of failure
Some of the lines that should now be straight are dipping a little -- for instance, this one which honestly is a bit of a hand-wavy metric, www.chiaexplorer.com/charts/networkDecentralisation but mostly things are doing just fine.
DSHR lost heart in Chia, I think, because it ceased to gather headlines, and all the real work disappeared back into the Discords and Keybase chats where most of the continuing work of crypto systems go on.
But this is the biggest problem with external analyses of the crypto space right now. Of *course* you're going to just see the grungy, bloody foam of the shillers, the rug-pullers and the ponzi schemes. All of those people, like spammers before them, primarily rely on visibility.
There will be a reckoning, and those people -- well, they're not going to go away, they will be replaced with other, crazier things -- but trying to use them as indicators of the dull progress being made, is futile. They don't point in that direction. They point at themselves.
Much of me reading top-level crypto critiques at the moment is full of that thing where it all makes sense, except for the bits which I know about, where it is weirdly wrong. (Like the references to IPFS and scuttlebutt in that two hour video you keep wanting me to watch.)
For me, what's interesting is that I know so many of the crypto-sceptics, and in real life, we have amusing and fair-minded conversations where they admit there's bits of it they like, and I admit there are bits of it that give me the heeby-jeebies.
And then we go online to explain these subtle points, have to elide them or punch em up to fit into less than 300 tweets, and immediately you get pounced on by fucking weirdoes who think you're kissing satan's backside or proof that the Protocol of the Elders of Zion is real
Oh one more point, which I think is important: that chia node doesn't make me much money any more. If it stops making money, i'll stop running one. And the total number of lottery cards will go down, and eventually someone will realise that Chia is a good use of old hard drive.
This is what stable, decentralized economic systems look like -- not hugely profitable, but on the margin, logical to run. Which is how I foresee "proof-of-stake-ish" things working. More like low interest dependable bonds than the make $$$$ fast of these boomtimes.
As in so many parts of our lives right now, I look forward to more boring times. Keep looking for those working to make the world more boring; listening to them is a better guide than those of us who are constantly looking to make it sound more interesting -- or upsetting.
Talking of boring, I assume nobody is ever going to reach this far downthread, but if you are: follow my exciting work at @ffdweb and @filfoundation and if you are at #ETHDenver do the equivalent of buying my soundcloud by coming to these events: bitly.protocol.ai/filethdenver
