This isn’t the sort of thing I expect people to “get in trouble” for at the time of filing. I expect that years later, if the startup turns out to be worth something and the taxpayer is audited, the auditors will go through old paperwork and discover a very large amount of taxes they can collect.
This was on an iPad. My experience before had been that I was able to register an iCloud passkey for Google 2SV when using Safari on iOS/iPadOS, but I was unable with Chrome.
At that time, I inspected the registration request Google sent to Chrome and found it was passing a private option that Chrome recognized. According to what I found in web searches for it, the option created a legacy U2F key, and they needed to do that because there were existing Android devices that they could not upgrade and that would not support log-in with WebAuthn keys.
I wrote about my experience with passkey support on other services here: https://news.ycombinator.com/item?id=35758918. That experience was mostly negative. For anyone implementing this, the user experience matters and requires usability testing of a lot of combinations.
In comparison to those, Google’s support seems better. It worked, was transparent about what was going on, and gave me the option to create the key on either the device I was using or another one if I wanted. The one hitch was that when I already had a 2FA key on the same platform authenticator, it just said I already had a registered key on this device and didn’t do anything. I would have expected some sort of upgrade flow for people who previously registered their devices for 2FA, or at least to more directly tell me to delete the existing security key on the device (which is what I did, and which worked).
Update: I decided to go back and look at the settings pages for the two that worked.
Kayak shows a "Set up passkey" button with no indication that I already have one or any apparent way to revoke the existing one. Very bad.
Shop Pay does clearly show what passkeys exist, and allows revocation and addition. However, when I try to add one on Chrome on macOS, it only lets me set up a Chrome passkey. (These don't sync between devices. It should have let me set up whatever kind I wanted.) Also, the list just indicates passkeys by what browser type last used them, like "Chrome on macOS." So according to the list I have two passkeys that are just described as "Chrome on macOS" and they are indistinguishable.
I can't imagine recommending anyone try this unless they are consciously an early adopter.
I want to be enthusiastic about this technology and at first I was, but it's getting harder. I tried to set up passkeys with the sites listed here that I already have accounts on. Results:
1. PayPal: I could not find an option to do it on either a phone or a web browser even after reading the linked instructions. Perhaps it thinks neither device is compatible (they are).
2. Microsoft: After eventually I found an option inside its settings to go "Passwordless" which sounds right, it instead tries to ram an app called Microsoft Authenticator down my throat. On passkeys.directory I check the instructions for insight and I see it says "All you’ll need is a device running Windows 11 and the Microsoft Edge browser." So much for cross-platform standards. If I had been using Windows 11, I assume I would have still needed my password to log in on my phone or anywhere else.
3. Kayak: Works fine, although the log in option is not as discoverable as it could be (I had to click to log in via email, and then click in the email field, and then an option appeared).
4. Shop Pay: Works fine, the only one about which I have no complaints.
I think that’s reasonable if you are on a recent JVM version and expect to upgrade very regularly. Many are in organizations where that’s not the case.
knaekhoved your comment is dead which I think is unfortunate because it seems better to inform you than bury your comment so I'll reply here: the page did mean N -> N, given the context of the previous paragraph. You may have thought what you did if you did not know that Card(N)^Card(N) is equal to 2^Card(N) by the laws of arithmetic for cardinal numbers. If you have not encountered those before there are enough on Wikipedia to show that equality https://en.wikipedia.org/wiki/Cardinal_number#Cardinal_arith....
It incentivizes the welfare-maximizing result. I don’t have a short “intuitive” explanation of why handy but the math is covered in section 8.3 of Public Choice III (should be easy to get in college libraries, pdf copies can be found online if needed).
I wasn’t there but I have seen multiple histories say that there were servers that accepted connections from anyone (most famously eris.berkeley.edu but not only that one). For example, https://about.psyc.eu/IRC
It doesn’t have *. What it has is listed in the declaration starting with “C vt[]”. Those are the names of verbs and the following lines give the function pointers for their monadic and dyadic uses.
It worked fine for me when I was careful to stay within what it actually supported, bearing in mind it’s a fast and rough implementation not something for production use.
In the US, federally funded research institutions have to follow certain rules when doing human subjects research. When I was a researcher the training I took said prisoners generally could not be used as subjects just because they were convenient, and that the two situations where they could typically be studied were cases where prisons, crime, or something else that was inherently specific to prisoners was being studied and cases where some condition very highly over represented among prisoners was what was being studied (the example given was HIV).
Cochrane seriously misconstrues what the WSJ op-ed is about, which I think is understandable since it adopts a pro-regulatory tone while in the underlying Second Circuit appeal that the op-ed is about the OCC is adopting a (sort of) deregulatory position.
In brief, if you want to set up a fintech company now, you may need to comply with 50 states laws. That isn’t just being basically law-abiding, but involves serious compliance work. I started looking up licenses for Square as an example and got bored after the first few as state web sites are all different and inconvenient:
Each of those, and also for all the other states, involved satisfying a state agency that Square was adequately solvent and would comply with a host of laws. You may notice that the first two have license numbers and the numbers are small. That says something about how easy it is to get the licenses. This is a meaningful part of these companies’ moats.
The OCC action that New York is challenging is to provide a federal alternative structure in which companies would get one federal license and receive nationwide permission to engage in a host of banking-like activities. They would still be able to get state licenses instead, if they wanted. This system (of picking between a state and federal regulator) is called dual chartering and has existed in the banking industry for a long time; it establishes a check on both state and federal regulators by allowing the regulated entity to switch to an alternative regulator.
It might be that the LLC is liable for the taxes and not you personally. Do you really want to pay $1600 for something you aren’t really using or do you want to abandon it and walk away?
Nothing related to screen resolution. Wifi worked via the command-line tools mentioned in the manual, or the KDE interface. Similar for other things. I didn't stick with KDE--I switched to sway--so there were a number of things to deal with because of that, but that won't matter if you do use one.
If you do encounter an issue it will be more work to deal with, because you both have to figure out the underlying issue and then figure out how to fit that into the Nix way of doing things.
The biggest factor is handling downloaded pre-built binaries that assume various libraries will be in typical locations. Those generally require patchelf and then they work fine. I would only recommend NixOS for now for people who are willing to tinker a bit and learn how it works.
I do. When I installed it the default configuration had no desktop enabled but had commented-out lines to enable KDE. I think it's best to think of it as not having a default or preferred one, and you are instead required to pick one.
(I haven't tried everything, so I don't have firsthand knowledge that they all work equally smoothly.)