One possible way to deal with this and very nearly return to the former freedom-to-tinker status quo is to send the bank your custom OS along with a computer-checkable formal proof that the bank’s app, while running on your OS, behaves as it would be expected to under the stock OS. With homomorphic encryption, it might be possible to do this without revealing your custom OS, only its one-way hash. The bank can then verify that the proof is correct and then accept transactions with attestation from your custom OS. This would enable installing a custom ROM that can be used for online banking without having to go through some cabal/consortium. The only caveat is something of this magnitude has never been done before. It’s a research project for sure. It would take many man- and compute-hours. But it would be very cool.

https://danluu.com/diseconomies-scale/

Please share.

https://boingboing.net/2011/12/27/the-coming-war-on-general-purp.html

Anyone have a more in-depth technical description of how that works? I’m interested.

The NoScript extension will properly do this. The extension blocks domains from running scripts except those you’ve whitelisted. There’s a drop down that displays a list of domains from which the page wishes to run scripts. It makes much of the web a pain to use, though. I sometimes have to go through a loop of whitelisting a subset of domains which want to run followed by a page refresh until the page works. Javascript is often not optional. If you had to live like Richard Stallman professes you should, you’d probably have to join the Amish.

I think the trouble young people have with using desktop computers is overstated. It’s a bit of a satanic panic situation. You can learn it pretty quickly. A common complaint is that “they don’t know hierarchical file systems” because the mobile devices have only flat file systems presented to the user or something. A tree structure is not a challenging concept and the basic things you can do in a file system you can count on 2 hands. Open a file, save a file, rename a file, delete a file, move a file, copy a file, create a directory, enter a directory, move up a directory. The physical interface is the mouse with 2 buttons, a primary and a secondary for opening context menus; and the keyboard which has the characters printed on them. There’s a bit more to it, but it can be explained in, like, a page of text. And the rest you can learn through experimentation. Touch typing is another thing entirely, though. That takes dedicated time to learn.

I wonder if ineptitude with tech shared between the young and old are different kinds. Maybe the old are just completely inept, but, for the young, it’s just temporary. It’s a shock when we find out they don’t know something, but, after explaining it, they’re productive within minutes. A 20-year-old still has plenty of mental plasticity. Having to teach somebody the desktop metaphors isn’t a huge bottleneck.

I’ll end by contending that I don’t think schools should not be teaching computers. Rather, they should be teaching computers in more depth. Teach students basic programming and they will have to learn the desktop metaphors along that journey anyway. Computers are way too important to leave the future stewards of the Earth in the dark about how they work. I had to learn how the energy of a photon relates to its wavelength and I had to read and analyze the Canterbury Tales. Not entirely useful. But it’s at least a little interesting. Kids are very capable. They won’t all be programmers. They should learn it all anyway. Don’t let Silicon Valley have it all to themselves.

Something like an introduction to unix and programming should be mandatory. They seem to think that kids need to “learn to use a computer and the internet.” It’s a fucking point-and-click interface. What is there to learn? The software industry is very skilled at making it all so easy that a chimpanzee can use it. You don’t even need to read a manual. I wonder if this is all a holdover from the 70s when the computer interface was likely to be a paper teletype which is naturally difficult to use without instruction. We’re living in the future. Teach the difficult stuff. The teachers need a wetware update.

Also, other sites on the Net can collate links posted to reddit and the users and times that they were posted with users accessing pages on their own site. For example, if you’ve posted a sufficient number of youtube links, youtube can possibly link your reddit account and youtube account, say, as belonging to the same person. Just about any popular site can link your reddit account with your IP address. And your facebook account with your IP address. Etc. And thereby link reddit accounts with facebook accounts in cases with long-lived IP addresses. If X is a scrapable site on which users with fixed pseudonyms can submit links and Y is a site popular enough and has enough content, then Y can collate user accounts on X with IP addresses. This is not to say anyone is doing this. There are not many sites that meet the criteria to be a Y. I would say it includes mainstream news sites. The authorities could compromise some of the Ys or create their own through astroturf and thereby link users on the Xs with IP addresses and then with location.

These days, this might be a bit harder because reddit closed up when the AI companies started scraping. I don’t know the situation with Facebook with regard to scraping. If it’s scrapable, they’ve got your name linked to your reddit account.

What I’ve said about reddit equally applies to lemmy, even more so because lemmy is open by principle.

You can’t find anything on youtube these days. Uploads from ordinary people don’t appear in search results. They’re crowded out by the mainstream. It was called youtube for a reason. It isn’t that anymore.

Build your home as a Faraday cage. They can’t bypass physics.

P.S. Holy crap. The guy on the radio is on lemmy?

They’re not competing with you or me in the conspicuous consumption game. They’re competing with all the other rich people.

I swear bro, just let me add one more gigabyte. I swear, we’re gonna get to AGI. Just add one more gigabyte. Just let me add one more gigabyte. Just let me add one more gigabyte. I swear. I swear. I swear. I swear, we’re gonna get to AGI. Just one more gigabyte and just make it bigger. Just make it bigger. We’re gonna get to AGI. We’ll get there. We’ll get there. We’ll get there. Just make the model bigger. I swear.

They turned Windows into an IoT device. It’s your refrigerator with a TCP/IP stack and a touch screen bolted on the front. How many watts does the fridge use? Oh, I don’t know, but look, it has a digital calendar! How long does it take to cool items down? Who cares! You can use it to set reminders! When will I need to replace the gasket? What? I don’t know. But it can scan barcodes and send it all to the cloud. Isn’t that neat?! Cool, cool, but why does my fridge need to do that?

The administration didn’t threaten to take down the IA or investigate it or anything like that, so it’s not similar at all.

It’s conspiratorial to think the FBI is doing this to censor or hide something. archive.is is primarily used to get around paywalls. The most likely explanation is news sites complained to the FBI that their copyrights are being violated (which is true), so the FBI is investigating. They’ve had a problem with falling revenue for a decade or more at this point as everything went online and people expected to get instant access for free in contrast to print media.

It really highlights just how centralized so much of the internet is on like three companies (Amazon, Microsoft, and Google)

Cloudflare: What am I? Chopped liver?

I’m similarly skeptical it’s a genuine outage given how so much stuff nowadays is done with cloud computing which enables dynamic on-demand procurement of server resources. Or whatever the correct terminology is to describe that.