modrobert

@icing@chaos.social sudo apt-get install coreutils-from-gnu

@jerry Yes, abandoned, last admin post in 2024, lots of error 500 when using. The final straw which made me move was when my profile couldn't be viewed, get error, same problem for others I tested; e.g. https://qoto.org/@barefootstache , one of the listed admins. I'm suspecting the server is running out of disk space for certain data, or similar problem.

@dalias @jerry The scientific method? Anyway, I noticed shortly after joining that some pushed the "If we don't believe in free expression for people we despise, we don't believe in it at all" mantra hard sometimes which led to blocking on several other servers, but these were problems while the website still worked without errors, which it hasn't for the past year or so.

@jerry I had a similar experience with qoto.org.

@loke Lazy and lame, makes me question the underlying work as well.

"Cross sectioning insects in an electron microscope with a femtosecond laser" https://www.youtube.com/watch?v=NwhVJ7cv9B4

Here's a JavaScript port of Eliza; the computer psychotherapist from 1966 using a token system with a simple lexer: https://modrobert.ddns.net/eliza This was circulated a lot back in the day (early 80s) when you typed in Basic code mainly from magazines, using slow microcomputers with only a few kilobytes of RAM and 8-bit processor. You can find the original paper here: https://cse.buffalo.edu/~rapaport/572/S02/weizenbaum.eliza.1966.pdf Eliza is pretty coherent considering how little resources it uses compared to genAI.

Another "quantum secure" encryption scheme bites the dust... https://eprint.iacr.org/2026/1630

@0x00string@infosec.exchange The marriage of corporations and government.

@0x00string@infosec.exchange As long as we allow cartels that might be the case.

@RueNahcMohr@infosec.exchange Looks similar to the ones that used to pop in my US-Robotics Courier modems.

@RueNahcMohr@infosec.exchange Sometimes Motorola 68060. 🤤

@astralia@infosec.exchange Starvation is also a destroyer of humanity. If the planet is more important than humanity then the solution is on the horizon, according to the paper 4 billion people dead by 2050 at +3C temp increase ("Appendix I: Risk impact matrix").

@astralia@infosec.exchange All life is precious, including plants.

@briankrebs@infosec.exchange I definitely thought the craziest thing would be gov related, because they never cease to amaze and neither do their shills, but perhaps you keep the worst ones to yourself? ;)

@briankrebs@infosec.exchange Not sure what's worse; the shills or the ignorant.

@cR0w@infosec.exchange https://c.tenor.com/gpx4Kh5RlisAAAAM/puppet-puke.gif

@harrysintonen@infosec.exchange Don't they ever get tired using the same bullshit excuse? We know it's not about children.

@lcamtuf@infosec.exchange Scorched earth approach.

@pancake@infosec.exchange No prompt, was just reading this: https://actuaries.org.uk/media/ni4erlna/planetary-solvency.pdf Specifically "Appendix I: Risk impact matrix". Building more data centers for genAI is the last thing we need.

@pancake@infosec.exchange The data is from scientists funded by insurance companies called actuaries, their business is making profit from trying to predict the future accurately, so no political bias. The data center part was my take on current events, it is 4% of global energy production and climbing fast.

@jukkan@mstdn.social I remember Richard Feynman mentioning something like this back in the day as well.

@MLE_online@social.afront.org Chopper?

@MLE_online@social.afront.org Making sure people stay divided under the pretense of serving the public, seems like a truthful motto then.

@MLE_online@social.afront.org I can imagine the part about the "gold cans", the "year 3500" was harder, at least with humans in it.

@MLE_online@social.afront.org Would humans survive on Venus?

@MLE_online@social.afront.org Some context: https://en.wikipedia.org/wiki/Runaway_greenhouse_effect

@susankayequinn@wandering.shop Unfortunately training AI works a bit like a one way hash, you can't really prove what's going in there once the dataset is complete, which makes it difficult to verify that opting out has any effect.