It simply uses the “Not Hotdog” algorithm from Silicon Valley.

On a more serious note, it likely does on-device pattern recognition, similar to how other apps like TIDY works and will put a warning/blur if the image matches specific keywords.

Not sure why they didn’t just merge the capability into the Private Compute Services app instead.

The first Senators to actually do anything about the Trump admin

Yeah I’m not saying Bridgefy is better, just that it’s available on both major mobile platforms while Briar isn’t. I do prefer Briar on technical and privacy levels.

Bridgefy was used more during protests since it’s available on both iOS and Android, while Briar is Android only.

One way to avoid recurrent leaks is to foster a culture of trust and not be a ginormous dickwad.

There’s CryptPad that has a form feature which you could use for surveys.

On the other hand, NOT using MFA on an online password manager is just poor opsec.

Maybe it’s already the case, but may I suggest sending a PM to site-wide banned users including the reason and an out-of-band method for appeals (ie: email adresss).

You could even prebuild the email URL so that the email’s subject, body, etc to already contains the user’s info to facilitate the communication.

Stored in blue barrels.

You’d be better of security-wise looking for a secure alternative to the Yubikey, such as the NitroKey, Feitian, and some others. As long as their hardware is FIDO2-certified and the key has the capabilities you want (FIDO2, GPG, etc), you should be in business.

!support@lemmy.world would be the place to ask, moderator is simply applying the community rules.

The greatest deception of all: make them think it was their idea all along.

I used a platform called Aether which I used for a while but it seems the developer kinda abandonned it.

The idea was interesting, all participants would hold a copy of the network data locally and sync between eachothers (kind of like a blockchain, no it’s not related to cryptocurrencies).

Writing a comment, a post, upvoting and downvoting required conputational power, which limited the ability to spam the network. There was the idea of having elections to decide who could act as a moderator in each communities (and the ability to impeach an existing mod), but it never came to be.

Same, even my young kids if I ask them a yes/no question while they’re eating to avoid talking with their mouth full.

I have a knowledge rebuilding package, mostly in the form of the Kiwix (Wikipedia en full) archive and some other archives, as well as some other stuff like maps data from OSM. Next step is to buy a mobile device with eInk display with MicroSD support and load it there, to make it energy-efficient and that can be recharged using a solar panel.

There’s also Tuta if you’re looking for an alternative.

Of course.

On our end we’re monitoring the situation, if it ends up being spam that overwhelms the All feed and AI-generated slop we may just defederate from this instance.

And also relies on ActivityPub, which is the heart of Lemmy & Mastodon.