preasket

joined 1 year ago
[–] preasket@lemy.lol 0 points 2 weeks ago

We need a fully community run password manager with row-level server synchronisation between devices and shared vaults. Maybe a new client for the Bitwarden protocol with Vaultwarden or something new. E.g. 1password's secret key as a second factor is, imho, their best feature. It pretty much eliminates the possibility of the vault being decrypted due to a weak master password.

[–] preasket@lemy.lol 1 points 3 weeks ago

Yup, makes sense, since Android is also Linux

[–] preasket@lemy.lol 14 points 3 weeks ago (2 children)

AFAIK, the main difference is that Firefox's process isolation on Linux specifically is incomplete. They're working on fixing that.

[–] preasket@lemy.lol 1 points 4 months ago

Honestly, a very impressive move. Makes me way more confident in the trajectory of the company and I'm happy to have been a visionary user for multiple years.

I wonder, though, just how much of Proton A.G. does the foundation now own? They say it's the largest shareholder, but they didn't say "majority shareholder".

[–] preasket@lemy.lol -1 points 11 months ago

Make phone numbers optional and add a setting to allow/forbid accounts with no phone number to message you. I bet phone numbers have zero effect on the level of spam.

[–] preasket@lemy.lol 0 points 1 year ago (1 children)

You’re taking this too seriously lol

[–] preasket@lemy.lol 0 points 1 year ago (1 children)

I think people misunderstand what BlackRock does. It's a proxy for other investors. Investors all around the world buy assets, produced by BlackRock and BlackRock routes that money into corresponding shares/bonds/real estate, etc.

[–] preasket@lemy.lol 0 points 1 year ago (3 children)

What if the bugs are linux-specific? lol

[–] preasket@lemy.lol 0 points 1 year ago (1 children)

The idea of backpropagation and neural nets is quite old, but there's some significant new research being done now. Primarily in node types and computational efficiency. LSTM, transformers, ReLU - these are all new.

[–] preasket@lemy.lol 0 points 1 year ago (3 children)

Sure, tbh, I think ChatGPT is overhyped. It can be useful, but it's nowhere near AGI. I even have a controversial opinion that the rate of progress will not be exponential - it will be logarithmic, because, I think, the data will be the constraint.

 

I'm sure there are some AI peeps here. Neural networks scale with size because the number of combinations of parameter values that work for a given task scales exponentially (or, even better, factorially if that's a word???) with the network size. How can such a network be properly aligned when even humans, the most advanced natural neural nets, are not aligned? What can we realistically hope for?

Here's what I mean by alignment:

  • Ability to specify a loss function that humanity wants
  • Some strict or statistical guarantees on the deviation from that loss function as well as potentially unaccounted side effects
 

That version hasn't been released yet... 😅