Tarlos25
tarlos25.bsky.social
Tarlos25
@tarlos25.bsky.social
I exist. I know things. I do things. For reasons.

Happy Today!

Keep on keeping on.
We already have password sharing in place. I used myself as an example. I have seen this situation come up for others in my life, it factors into some of my decisions.

I am far from a security professional. My expertise is in High Performance Computing, where security is frequently a liability.
September 15, 2025 at 4:49 AM
And I don't expect anyone to help me remember my passwords. There are other edge cases where passkeys cause problems. How does my wife get into my accounts if I die? That's not an edge case, EVERYONE will eventually die.

By the way, I would enjoy having more ads like that.
September 15, 2025 at 2:50 AM
Then why avoid truth in advertising? Just be honest about there being situations where the passkeys are less convenient. More secure, but less convenient.
September 15, 2025 at 12:39 AM
I never said they weren't better. I am suggesting a way to improve the ecosystem around them further.
September 15, 2025 at 12:38 AM
Considering that this has degraded to a point where tracking any individual thread is difficult, which point is "crazy talk" and why?
September 15, 2025 at 12:32 AM
It isn't more convenient for a significant number of people, even on modern, supported systems. That is what's wrong with. Safer, sure. Faster and more convenient, not universally. I'm asking for a more detailed approach to the advertising.
September 15, 2025 at 12:30 AM
"High"? 95% is not high. And it does not compromise the goal at all. If expecting truth in advertising is compromising the goal of the advertising, then there is likely something unethical going on.
September 15, 2025 at 12:25 AM
You seem to be under the impression that I am saying we shouldn't use passkeys. Which is incorrect. I am only talking about how they are advertised.
September 15, 2025 at 12:23 AM
You are saying that supporting 95% of a user base is too high of a success metric and unachievable. I provided a real-world example where it is actually a very low target and easily exceeded regularly. Quite relevant.
September 14, 2025 at 6:15 PM
A better analogy would be to say that new roads must support all existing cars.
September 14, 2025 at 6:12 PM
So it should be a reasonably large dataset, but biased somewhat. Fair. I still maintain that passkeys should not be advertised as exclusively easier, but it is a reasonable assumption that the full dataset is able to pass.
September 14, 2025 at 3:46 PM
Actually, in my team at work (building developer tools like compilers and libraries), our success metrics were significantly higher and over the last decade, those goals were met by around 99.4% of releases.
September 14, 2025 at 1:58 PM
That is an interesting data set. Am I reading it correctly that they are inly measuring on their site? Or are they measuring general web usage through something more broad like Akamai or Google Analytics?
September 14, 2025 at 1:51 PM
Incorrect analogy. I am not saying everyone needs to use passkeys. Your analogy assumes every vehicle is required to use electric powertrains. I am not looking for full backward compatibility. As I said, only 95% of devices. That is a reasonable target.
September 14, 2025 at 1:47 PM
Then maybe we should adjust for something that works for more devices. The majority is simply over 50%, which is considered failing in every class I was ever in. Frankly, 70% as "passing" is a really low bar. Computer tech is too quick to race forward and abandon too many people.
September 13, 2025 at 8:57 PM
Majority is not sufficient.
September 13, 2025 at 4:36 PM
And passkeys work well enough for many scenarios. But my point remains that they are NOT easier. Until such time as at least 99% of use cases are actually easier, then it should not be called easier without qualifiers.

Side point, good software design handles edge cases well.
September 13, 2025 at 1:17 PM
Windows 11 also requires TPM, which isn't on my processor. Any login solution needs to support older technology as well as new.
September 13, 2025 at 3:55 AM
Meant to say that it *has* nothing for biometrics.
September 13, 2025 at 2:15 AM
I am the device manufacturer. Windows 10 Pro, desktop nothing that can read any biometrics. And I did go through the enabling steps for Windows Passkey support. This is the fallback, which is the default for a LOT of people.
September 13, 2025 at 2:14 AM
I tried it. Now, it wants me to insert a security key, which I don't have. And if I did, now it requires yet ANOTHER device I have to carry around with me. This is NOT easier than a password. It is more secure, but not easier.
September 12, 2025 at 7:40 PM
Then why does every site that prompts me for a passkey send me a one-time code? What kind of passkey are you talking about?
September 12, 2025 at 3:58 PM
In that situation, passkeys are no easier than passwords. The passkeys I have encountered all involve some type of one-time code. Which requires waiting for a text (hope you have good enough signal) or waiting for an email (which requires another login that might need ANOTHER passkey...).
September 12, 2025 at 12:58 PM
To address with a little more detail, scalable incorrectness would be quickly telling incorrect information to more people with less effort. Such as publishing lots of articles claiming that passkeys are easier. Hence my original point: stop claiming passkeys are easier.
September 11, 2025 at 9:45 PM