The AI part of this is a red herring. This is above all a big devops failure.
Three takeaways:
1. TEST YOUR BACKUPS. If you have not confirmed that you can restore, then you don’t have backup. If the backups are in the same place as your prod DB, you also don’t have backup.
2. Don’t use Railway. They are not serious.
3. Don’t rely on this guy. The entire postmortem takes no accountability and instead includes a “confession” from Cursor agent. He is also not serious.
4. See #1.
Running a single bad command will happen sometimes, whether by human or machine. If that’s all it takes to perma delete your service then what you have is a hackathon project, not a business.
> If my account tries to send someone $3 million, I'd prefer that it's intermediated by a confused bank employee staring at a screen
This is a nice lens for looking at when stablecoins make sense.
If you're an American using your Chase account to buy coffee at Starbucks, the permissioned, heuristically fraud-checked, slow-settling tradfi system is well optimized for you.
If you are an importer buying $3m worth of bulk coffee from Kenya, you would much rather have an instant 1:1 USD transfer on beautifully efficient machine consensus.
In many countries in the world, the banking system is extractive and unreliable. The "confused employee" is not there to help you. The two weeks of money in transit is no benefit, just a source of additional counterparty risk, cost, and delay.
An immutable and transparent ledger is not for everything but it is a useful primitive.
Many skeptics assume that stablecoins are just about regulatory arbitrage.
That's part of it, but:
1. Progress often depends on evolving obsolete regulation.
Uber works much better than taxis (once upon a time, people could "call a dispatcher" an hour in advance, wait on hold, etc) and yet in the early years they had to work around taxi regs.
2. Blockchains are a fundamentally more robust way to run a ledger.
If any of you have ever written software touching tradfi custody you'll know about "reconciliation"--start of every business day, you get a dump of files in your FTP server in various proprietary formats. You parse the transactions and they don't add up. The Recon team hand-corrects and recategorizes edge cases so that the balance deltas match transaction totals and everything ties out.
This type of absurd duct tape is ubiquitous, and it's a major reason why trad rails have multi-day settlement times and even longer for international. Inflates team size and cost required to run a product. SWIFT is a messaging system -- bankers use it to essentially text each other about wires to figure out issue resolution. Some lower-level trad payments regulations are written assuming that this level of manual oversight is required to prevent ledgering errors and ensure sound accounting.
Stablecoins run on transparent, precise ledgers with machine consensus. This doesn't solve everything, but there are large categories of issues that can occur in trad payments that do not exist onchain.
3. Control is liability.
Some important regulations actually encourage blockchain-based payments. For example, money transmitter law places significant requirements on custodial money transmitters (you take money from Alice, with a promise to give it to Bob) that do not apply to noncustodial channels (you give Alice a mechanism to send directly to Bob).
> The problem with all that, is the fact it remains possible to create a protocol with N big institutions [...] This maintains many benefits of the blockchain and lacks many issues (fast, simple, near zero cost)
That's more or less exactly what this is. Stripe is launching an EVM L1.
The Ethereum Virtual Machine part gives it a mature tech stack with experienced developers and auditors. Plus, well-tested smart contracts that have already processed billions of dollars on other chains can be deployed on Tempo.
The "Stripe L1" part will ensure that it's fast, simple, near zero cost.
They have a nicely implemented E2E protocol. This is operationally convenient: Meta can accurately say that they don't store WhatsApp messages, so fewer access requests go to them. And I'm sure it's nice for engineer morale, too.
However, the app makes it semi-mandatory to turn on backups. If you say no, it keeps nagging you. If you always say no, you are in the 0.1% and everyone you talk to has backups enabled, so all of your conversations are helpfully backed up anyway, just not for you :)
These backups go to Google Drive or iCloud. You can draw your own conclusions about who has access and who handles the LE/IC requests.
> Just because there’s a serious problem doesn’t mean that every response is a good one
Problem: a forum full of misanthropes dedicated to saying the worst things allowed under the first amendment.
Bad solution: erode 1A at the case law level
Bad solution: censor the internet at the backbone level
Freedom isn't free. We're lucky to live in country with robust speech protections. The tradeoff is that there will always be some people who get a kick out of going right to edge of what they can get away. My view is that our civil liberties are worth it.
And speaking of user-hostile, locked-down phones...
a galactic irony that Ben Wiser, the Googler who posted this proposal, has a blog where his most recent post is a rant about how he's being unfairly restricted and can't freely run the software he wants on his own device.
One simple reason - separate from the less fun possibilities involving defense procurement politics - is that the F117 was a transition technology.
It proved that super low radar signatures were possible.
The design used those big low poly triangles not because it’s optimal (aerodynamically, very much not!) but because of the limits of computer simulation at the time.
There’s a whole fascinating story about how the theory behind low observable was developed by a Soviet scientist, published, ignored there, then implemented here.
But computer technology quickly advanced to where low-poly aircraft made airworthy by brute force were no longer necessary. See the B2 Spirit, also a very special simulation derived shape but streamlined.
It's an exceptionally flexible and convenient method of control.
Earlier this year, donors to the truck convey protest in Canada had their bank accounts frozen. This wasn't a targeted list of "these 37 people have broken a law"--rather, it was a broad mandate to freeze accounts assocated with the protests, operationalized a bit differently by each bank.
In a society where most businesses don't take cash anymore, this turnkey coercive capability becomes more airtight.
The frog will boil slowly. A few years ago, all US payment processors blocked donations to Wikileaks, after they reported on war crimes in Iraq. Today, most people still think of digital money in the same way as physical cash; in reality, every transaction is a request for permission, with fraud heuristics and blocklists that might say yes or no.
Soon, a guy gets DUI, loses the ability to buy alcohol for six months--who would oppose that? Over time, the scope and frequency of financial deplatforminig will expand. Twitter does one-week suspensions for violating their terms of service. Why not your credit card?
One of the things I love about Palladium (and closely related, Samo Burja's newsletter) is the depth of research.
Like the detail that one of the most egregious episodes from California HSR involved a Spanish company that performed excellently on rail projects in Spain. Overall, this piece makes a strong case that the problem is specifically NIMBYism and loss of government institutional capacity.
I think the million dollar question is how government organizations can hire and retain better. The current situation looks dire. Obviously a charismatic leader with a broad anti-NIMBY mandate would go a ways at getting competent people to want to work in government. You saw that succeed on a small scale with orgs like US Digital Service.
The elephant, after that, is merit-based pay and promotion. Someone needs to sell this to the public. RN literally random cops and plumbers make mid six figures thru overtime while the directors of $100b mega-project are low-energy lifers making less than that. That's not gonna work.
This post is a fantastic example of the poverty trap. We design laws that, on paper, help the poor; and then we blame them for not taking advantage.
When in reality, they are poorly implemented, poorly advertised, and confusing.
With a $10k cap, this program is clearly intended to help people with low assets. How many people with a net worth under, say, $20k in the US? Tens of millions! How many of those would've really loved the extra $850 to offset this year's record inflation? Basically all of them! How many of them found this obscure website with awful UX and received the money? Approximately none!
I skipped Aramco on purpose. It's a vanity valuation.
The true top five each created valuables businesses based on 0-to-1 products.
Aramco didn't create much, certainly nothing worth close to $2T. The Saudi autocrats just list their country's oil reserves (preexisting value, created by nobody) on this state-owned enterprise's balance sheet in order to flex on most-valuable lists.
Political debate on HN has become unavoidable because tech itself has been politicized.
Of course on some level it always was, "everything is political" etc, but a decade ago most participants saw tech as broadly neutral. What was the political valence of Etherpad or Foursquare?
Today, engineers are asked to implement things like the "inclusivity warnings" that just shipped in Google Docs. The scope of "content moderation" has expanded dramatically. Founders are often explicitly partisan in one direction or another.
And the new engagement goes in both directions. The five most valuable companies on earth are all West Coast tech cos now. Political actors of all types are watching and trying to harness or control tech to a much greater extent than last decade.
FWIW this is the same argument once made against human flight. In the late 19th century, there were a lot of debates in the form
> Clearly flight is possible, birds do it
> Sure but how/why is one of the many mysteries of the universe, one we will likely never solve.
"Man won't fly for a million years – to build a flying machine would require the combined and continuous efforts of mathematicians and mechanics for 1-10 million years." - NYT 1903
