Hacker News: devman0

New comment by devman0 in "If AI writes your code, why use Python?"

devman0 — Tue, 12 May 2026 13:12:51 +0000

If you use the typing system (which I do religiously) Python becomes a lot easier to reason about in larger projects it also makes linters and refactoring tools easier to use.

New comment by devman0 in "Florida judge rules red light camera tickets are unconstitutional"

devman0 — Mon, 09 Mar 2026 20:39:42 +0000

Idk how Florida handles it but several states citations issued by red light cameras and those issued by officers are handled entirely differently for the exact reason you mention. Camera citations are entirely civil, you don't get points against your license. If a cop issues the ticket it does become a misdemeanor moving violation.

New comment by devman0 in "Put the zip code first"

devman0 — Sun, 08 Mar 2026 03:37:15 +0000

Even if a zip code contains multiple cities, each ZIP has one "preferred" locality name and you can default to that. Any of the locality names within a zip code is deliverable for all addresses in that zip code.

New comment by devman0 in "America's pensions can't beat Vanguard but they can close a hospital"

devman0 — Tue, 17 Feb 2026 20:44:05 +0000

Businesses can use deposit management services to spread cash among many banks. Bonus points they also are less impacted by the poor business practices of one bank.

Individuals can do this too with investment brokers or wealth management providers.

Alternatively we could just made FDIC coverage unlimited, but then that creates poor risk taking incentives, which is the whole point of not setting the expectation of the a bailout by making exceptions.

New comment by devman0 in "America's pensions can't beat Vanguard but they can close a hospital"

devman0 — Tue, 17 Feb 2026 16:22:50 +0000

It is supposed to be if the amounts are above $250,000. I have no problem with the first $250k being risk free, that is a policy that is well published and that we all "agree" on. Making arbitrary policy decisions that in some cases depositors should be made whole when risky behavior (such as depositing above the insurance limit) bites them is problematic. Stick to the policy or change the policy don't make one off exceptions because that sets weird expectations.

89% of deposits at SVB were uninsured.

New comment by devman0 in "Discord/Twitch/Snapchat age verification bypass"

devman0 — Thu, 12 Feb 2026 17:14:56 +0000

The privacy issue has two facets, when I show ID to get in to a club or buy alcohol, the entire interaction is transient, the merchant isn't keeping that information and the issuer of the credential doesn't know that happened (i.e. the government).

Just allowing a service provider to receive a third party attestation that you "allowed" still allows the third party to track what you are doing even if the provider can't. That's still unacceptable from a privacy standpoint, I don't want the government, or agents thereof, knowing all the places I've had to show ID.

New comment by devman0 in "'The old order is not coming back,' Carney says in speech at Davos"

devman0 — Wed, 21 Jan 2026 04:26:53 +0000

Right, I mean the Roman Republic declined and gave way to Roman Empire for a long time before Rome finally waned. That doesn't feel like it would be a good time for the rest of the world if the United States gave way to the United Empire for next 500 years.

New comment by devman0 in "What came first: the CNAME or the A record?"

devman0 — Mon, 19 Jan 2026 20:20:05 +0000

Even if 'possibly preface' is interpreted to mean CNAME RRSets should appear first there is still a broken reliance by some resolvers on the order of CNAME RRsets if there is more than one CNAME in the chain. This expectation of ordering is not promised by the relevant RFCs.

New comment by devman0 in "IPv6 just turned 30 and still hasn't taken over the world"

devman0 — Sat, 03 Jan 2026 03:13:05 +0000

> It's called a firewall. You want a firewall. IPv6 also has a firewall. NAT is not a firewall. NAT is usually configured as part of your firewall, but is not a firewall.

Expanding on this. NAT as deployed in most soho/residential settings requires a stateful firewall to track connections + port mapping logic.A stateful firewall is also used for IPv6 edge security and using the same basic posture (out allow, in established/related only) except the only difference is it isn't also doing an address mapping. Nobody is out there saying folks should run a wide open IPv6 edge, and as far as I'm aware no one is shipping IPv6 ready consumer routers that do that (but I'm prepared to be proven wrong in the responses).

New comment by devman0 in "Disks Lie: Building a WAL that actually survives"

devman0 — Sun, 14 Dec 2025 23:21:55 +0000

Wouldn't that just be because the construction is common in the training materials, which means it's a common construction in human writing?

New comment by devman0 in "Offline card payments should be possible no later than 1 July 2026"

devman0 — Sat, 04 Oct 2025 04:59:17 +0000

Just because an imprinter was used doesn't mean the transaction was necessarily "offline". Depending on merchant's policy, the cashier would call their processor, give them some transaction details and receive an auth code for the transaction which would be written on the imprinted ticket, thus authorizing the transaction at the POS just the same way it's done today (except with humans and phones, instead of an electronic handshake).

New comment by devman0 in "Open Banking and Payments Competition"

devman0 — Thu, 14 Aug 2025 19:41:46 +0000

I would argue, the issue is that they had exclusive control over the customers bank accounts. I have no problem with fintech that integrates with existing accounts over the top.

The issue is with fintechs that are not really banks, so they keep your money in a "program bank" where you are not the direct customer. You can't go to the program bank and say Fintech XYZ went belly up and I want my money, because you are not the banks direct customer (even though in some cases your name is legally on the account for FDIC insurance reasons).

Open banking tech is way more about the former than the latter.

New comment by devman0 in "Do not download the app, use the website"

devman0 — Sat, 26 Jul 2025 03:46:34 +0000

The principle issue with hardware keys as implemented today via FIDO2 or U2F is that you can't enroll them without having them in your physical possession, which means if you have a backup key stored offsite, you have to fetch it anytime you sign up for a new service.

New comment by devman0 in "Why English doesn't use accents"

devman0 — Mon, 07 Jul 2025 03:12:55 +0000

Blame the Normans for that one...well English was already kind of a mess, but the the conquest of England by the Normans really sealed the deal.

New comment by devman0 in "DiffX – Next-Generation Extensible Diff Format"

devman0 — Wed, 04 Jun 2025 05:37:13 +0000

git does have word diffing if you need something more granular than line diffing, the default delimiter being whitespace.

New comment by devman0 in "Don't just check errors, handle them gracefully (2016)"

devman0 — Tue, 03 Jun 2025 21:42:58 +0000

This is, in theory, the difference between checked and unchecked exceptions in languages that support such distinctions. Checked exceptions are basically part of the control flow and the API contract used for things that can happen but are not necessary programmer errors (IO Exceptions being the classic example). Whereas unchecked exceptions are supposed to be used for programming errors (think Illegal Arguments, Null Pointer, Array Index Out of Bounds). Unchecked also usually cover other fatal conditions like stack overflow or OOM. Checked, handle with care, Unchecked, let the stack unwind and fail fast.

New comment by devman0 in "Show HN: SVG Animation Software"

devman0 — Sun, 25 May 2025 21:36:38 +0000

I think JetBrains model is best with thier IDE. You purchase, and get one year of updates, if you don't renew you get a perpetual license at whatever update level your update license ended at, and you can restart at anytime.

This avoids games about what is or is not a "new version" or shipping minor updates as versions or any of that nonsense, it's just updates generally.

New comment by devman0 in "Ending TLS Client Authentication Certificate Support in 2026"

devman0 — Sun, 18 May 2025 19:46:38 +0000

Header passthrough is nice to have, but there isn't really a standard for it for TLS, and it isn't well supported by most applications that are interested in doing mTLS. Additionally there is a trust component required between proxy and application and while can be accounted for in the architecture between the two a JWT instead passes through nicely and can be independently validated by the application.

New comment by devman0 in "Ending TLS Client Authentication Certificate Support in 2026"

devman0 — Sun, 18 May 2025 05:53:24 +0000

Client certs are strongly bound to the TLS handshake, and thus a connection, which makes them pretty unfriendly to a lot of proxy driven architectures, where as JWTs and Cookies can pass through proxies trivially, and are bound to requests not connections. There is some work being done around standardizing pass through of client cert information through proxies, but I'm not aware of any widespread support.

New comment by devman0 in "Moody’s strips U.S. of triple-A credit rating"

devman0 — Sat, 17 May 2025 04:10:40 +0000

The treasuries are effectively zero-risk assets to the trust fund and they also pay interest. Just having the fund sit the on cash would not be efficient.

The money goes to the trust fund, the trust fund buys treasuries and meets it's outlays using the maturing treasuries.