sovok_x

It's what betas are released for - to check whether something breaks or not. What's the point of reposting this to a news site?

This doesn't work though. I was banned there without as much as a single-line explanation after 10+ years of being a normal rule-abiding user which didn't change even up to that point. Maximum stuff they might not like I did is visiting through a proxy due to my country's circumstances. Right after that I tried appeals daily nearly for a week with no reaction whatsoever. So I just deleted my acc, less value for them is better.

"Incompatible hardware" is their corporate speak for any hardware that weakens vendor lock-in and propagation of MSFT-centered consumerism.

The comments section there says everything about how many people process such questions:

- historical ignorance;

- economical ignorance;

- repeating mass media propaganda narrative word for word;

- hubris.

I do hope europeans will start working as hard and as low-waged as chinese people did last two decades though, so those tech transfers will actually be on point. They lived too comfortably recently, forgetting that you always bear the hidden price for that in some way. No cheaply available products reality _is_ the norm, after all.

AFAIK there are two ways for this, neither of which is convenient to use: install all extensions from the source (you can unpack an existing crx for it or use their clonned repo if it's opensource) or use a group policy to disable extensions autoupdate and update each of them manually when the new version has something you want.

That reminded me of how Windows abuses my HDD by regularly writing dozens of logs I won't ever need unless I'm both knowledgeable enough and debug for specific errors.

And also about WinSxS/Temp/InFlight folder that gets filled with an undeleted garbage of unknown purposes. You can't safely delete files there (with elevated System privileges) even after finishing delayed patching and dism cleanup operations because some of them are still used by the system somehow despite it being a "temp" folder. While thousands of empty subfolders there slow dism/TrustedInstaller runs to a crawl on their enumeration.

More like false advertising and platform abuse, but given that this doesn't incur any loss unless you were actually hired and then dismissed, it's also not really actionable. The best measure here is for job search platforms to curate the whole process and deprioritize the companies caught doing that.

The thing is it was pretty obvious from the update description that something like this will happen. But the update in question is cumulative so even among advanced users that know how to restrict autoupdates many were forced to install it to prevent even more scary exploits for which there are no separate patches.

And now people need to disable Secure Boot entirely to avoid the side effects. Way to improve worldwide security, MSFT.

If you have uBlock Origin installed it should already protect you from this attack vector with "Block outsider intrusion into LAN" filter list, though.

And you also don't need this on Windows since 0.0.0.0 IP is blocked on a system level there.

Ignoring moral and sanitary considerations of accessing a dead body, it just won't work. Most modern phones use capacitive sensing during the process. Once tissue is dead, it loses its electrical charge and will fail to activate the fingerprint sensor, making it impossible to unlock the phone this way.

You just use bad tools for that. When using custom AutoPagerize UserJS mod that utilizes pushState, endless scrolling works like a charm for me for years already. Like I can scroll to thousand-numbered items of HN seamlessly. It remembers the exact page you were on after you close the tab and overrides "target" property of added pages' links to open everything in new tabs regardlessly, preventing accidental page reloads. IDK why developers don't reproduce this.

Kinda sad more and more Web sites switch to pointless AJAX app model recently and it stops working for those.

Heads up map digitizing seem to refer to the method where computer operator digitized an aerial image map by manually tagging its features displayed on the screen, in contrast to heads down digitizing where they used a special tablet.

Anyway I really need such articles from time to time not to lose my faith in humanity.

This site is currently a counterproductive trash that will simply waste your time on using it. There is some promise in LLM comparison but the limits of this make it utterly useless, even as a litmus test, since you can't check anything practical with it.

It has a very restrictive one-time prompt that stops working pretty fast sending you to buy their subscription and there is even no way to ask the models to fix or continue printing the unfinished code since the input field gets locked right after the inital prompt.

Also:

- Allows switching non-utf8 zip filename encodings from the menu, it's a mess with 7zip;

- Its context menu works inside Sandboxie without bugs;

- GUI is really much better, so I will reiterate the point. 7zip basically froze in time with it.

I will switch to it in a flash if it'll allow creating embedded recovery/par2 records for 7z and zip archives without breaking format compatibility somehow. Otherwise it's just an alternative GUI or things I don't need in practice.

Technically sites can prevent that by disallowing anything that looks like long baseXY-encoded blobs in their postable sections like pastebin does already.

Can someone explain why WinRar installers of the same exact versions and languages but downloaded from their different sites www.rarlab.com vs www.win-rar.com have different sizes/hashes while being signed by the same signer?

Well, the man clearly wants to kill the thing so crossing it out is not anything unusual.

For ads and online-dependent annoyware a third-party firewall works too.

Google Windows IoT. And yes, it supports WSL.

Use IoT version of the OS instead. It's quite difficult to remove some of the bloat from mainstream versions even having Windows debloater scripts or O&O's tools because of how deeply they're intertwined into the system. Removing some unneeded stuff sometimes causes either OS functionality or updateability to break too.

It's not a question of should or not. People just willingly give up privacy for convenience. For example merely using a phone (not to mention a smartphone with their location awareness capabilities, or a plain mobile phone with their multilateration, any one that works at all, even a landline if you use it at that very moment) is a breach of privacy. But you need it to function in a modern society. You can avoid such things by taking some extra steps but each of those steps takes progressively more effort and bites a huge chuck of the aforementioned convenience, not only for you but also for the other side of your social interaction.

The whole original topic is more about prevalent abuses of authority than privacy itself, though. And that's a different thing because protecting your privacy from authorized and/or dedicated actors is always quite difficult, if possible at all, even with a proper privacy "hygiene", and needs a totally different approach.

Also I suggest to use:

  ):xpath(following-sibling::tr[1])

  ) + *

At this point I just wonder whether Firefox userbase will flee the browser in droves if someone will make its unmozilled tracking fork sustained by easily payable tips, that has all the BS disabled and uBlock + couple of other top 10 extensions like Stylus and Tampermonkey natively implemented in it, with a good unified UI and all, for added value.

Also some of the settings are more of a silent cry "please stop doing this" because the browser will disregard them a few versions later.

Nice, today I learned of :upward CSS pseudo-class in uB, so I will change my old rules to that.

Still, it's better to add

  html:not([op="item"]) 

Nicely written. I want to add some usability ideas to this:

- Don't show multiply modals of the same error if possible. Add something like "(# errors)" to the existing modal instead;

- If some error is not critical but possibly recurring add an option to stop showing it for this session to its view;

- Don't send users to support for errors neither they nor the support can do anything about;

- If there is an error reporting option, provide a common report template for each type instead of forcing users to explain it as they see fit with a single "explain what happened" textarea.

It's nice as a vague intro but leaves more questions than it answers. Are there any actual insights into accessing and (ab)using this underlying RISC-like architecture, IME, or other mysterious stuff, somewhere?

If you use uBlock you can just filter out those posts with a couple of rules like I do:

  news.ycombinator.com###hnmain .athing:not(.comtr):has-text(/corona|covid|\bAI\b|GPT|wordle|Ukrain|Russi|\bChin[ae]|death|quarantin/i):xpath(following-sibling::tr[1])

  news.ycombinator.com###hnmain .athing:not(.comtr):has-text(/corona|covid|\bAI\b|GPT|wordle|Ukrain|Russi|\bChin[ae]|death|quarantin/i)

I'm not okay with this because its author cherry picks images to compare. Those might as well be an outliers. It'd be more convincing when comparing 10k images gallery with a variety of image types, like photos, paintings, sketches, manga, real objects, textures etc and then averaging.

Also subjectively the AVIF lossy compressed flag pic looks better to me because it has less noise, I don't really care about it being more "true" to anything in such use-cases.