HN user

y7

1,644 karma
Posts7
Comments283
View on HN
Signal Secure Backups 11 months ago

Without paying for remote backups, can I just manage my own backup on my own hard drive, and restore it when I want to?

Honestly I feel like you should only use kernel anticheat on a dedicated machine that's kept 100% separate from any of your personal data. That's a lot to ask of people, but you really shouldn't have anything you don't consider public data on the same hardware.

Yes, and at that point, you may as well use Windows for that machine.

I guess, from a Western-European perspective, the problem is that with the choice of Democrats and Republicans you get the choice between right-wing and ultra right-wing. Having right-wing politics that funnel money from the poor to the rich, or the tenants to the landlords, is in the interest of the financial backers of both parties. Messaging-wise, the Democrats have always been "more honest" (low bar, it's hard to be more dishonest/convoluted than Trump anyway), so maybe that's why Trump seems to come out ahead there.

What I don't get is how the bar for the Democrats seems to be so much higher than for Trump. Sure, "the typical man" is more easily validated by Trump than Harris, but at the same time Trump says much worse things about women than Harris about men. I can see how the Harris seems more "elitist" in a way than Trump, but to me that seems like a subtle negative versus Trump's long list of very obvious flaws.

How does the hatred for the Democrats get so big?

M4 MacBook Pro 2 years ago

When it stops working great. My 2014 Macbook is about due for an upgrade, mostly due to the GPU struggling with a 4K screen.

Synthetic data can never contain more information than the statistical model from which it is derived: it is simply the evaluation of a non-deterministic function on the model parameters. And the model parameters are simply a function of the training data.

I don't see how you can "bootstrap a smarter model" based on synthetic data from a previous-gen model this way. You may as well well just train your new model on the original training data.

The interesting question is not the choice between diversity and merit, but how you attract and select the most meritorious candidates in a setting where your company overwhelmingly skews towards one demographic, given that people are more likely to hire others that are similar to them (and conversely, potential candidates are not likely to join companies when there are no people similar to them).

The reason you're getting downvotes is because your comment is irrelevant. The point of the article is the transition of passwords to passkeys. Minimizing the number of accounts you have has absolutely nothing to do with that. If the point of the article was about optimizing your online security posture, and passkeys as a method for that, then your comment would be more relevant.

The reason no one replies and just downvotes, is because we don't want to clutter up the discussion with even more irrelevant comments. Usually I'd do the same, but for the chance that you are really commenting in good faith and not a troll, I thought I'd present you with a learning opportunity.

Besides the actual hardware, I think the biggest problem is the software implementation of trackpads in Linux (at least in libinput/Wayland). Even running Linux on my Macbook, the trackpad experience is vastly inferior to macOS.

I don't care too much about gestures, but mostly about stable palm rejection and proper acceleration curves. The current implementations sadly provide neither.

You don't think that having children is a selfish decision in itself? I have thought long and hard about whether I want kids or not, and ultimately I concluded that I should only have kids if I myself really wanted to raise kids (regardless of say, whether my partner wanted to have kids). Kids do not ask to be born, and life is not a guaranteed net positive.