They typically don’t. They do proxy it if there is something preventing a direct connection, but the proxy bandwidth is super limited and results in pretty terrible playback quality.

Lol, and what would the ransom be for taking down someone’s money-burning hobby project?

Tape is still the cheapest option for mass amounts of storage since the actual tapes are so cheap. You just need to store enough data to offset the cost of the drive. Drive cost increases very quickly the higher you go in storage density.

Interesting that that is the workflow that works best for you. I’ve personally always found it a much better experience to do my searching/browsing off of the server and wget whatever I need to download. If that’s truly your situation, then you may just need to use another browser that supports JS or use a different search engine. I prefer DDG anyway, lol. Not a huge deal.

You’ve seriously been in situations where you had no access to the internet except through a terminal, and you had to do a google search? No phone or other computer that you’re remoting in from?

Even so, there are terminal-based browsers that support javascript like brow.sh or links (not lynx).

I doubt the nothing-but-terminal users comprise a significant enough portion of Google’s userbase to justify the extra costs to test and maintain non-JS functionality.

I think this isn’t a case of if Google can, but rather of why they should. Do enough people really use the modern web without JavaScript to justify spending the resources to test and maintain functionality without JS? And they probably don’t want to let the few people that don’t have JS to open support tickets or write articles about how google.com is broken. Easier to just block it on purpose than to let it decay.

It makes more sense that a government website would support it, since they can’t let even a single person fall through the cracks, and changing laws/regulations is more difficult than making a company decision.

Search suggestions require JS. Also, why would Google spend the resources supporting the 5 people that block JS when virtually all websites and users rely on JS. This is a nothingburger of a story.

Google is a lot more than just the one google.com page. And even if it were, JS adds some nice features like predective text / suggested searches.

Tracking, ads, and AI can be done without JS. They may be slightly less granular in the same way as the user experience will be slightly worse, but disabling JS won’t stop it.

I’d bet the biggest reason Google decided to do this is so that they don’t have to support a version of the site that virtually nobody uses.

Imo, the most compelling reason for non-JS versons of typically JS-driven sites is to support lower power devices. But it’s 2025 and even a 10 year old phone you found in a dumpster behind a decaying Radio Shack can run modern websites without issue.

Even the article is grasping at straws for why this might be bad. “It might make accessibility more difficult or add security issues”. One of the most valuable companies in the world, with some of the best engineers in the world, is going to have problems adding aria attributes and updating dependencies? Give me a break.

If you want to block tracking, ads, and “AI”, there are plenty of ways to do that without disabling literally all JS. If you want to construct your google search request without the rest of the stuff on google.com, use your browser’s search bar.

I’m as anti-google/tracking/etc as the next guy, and I’ve been using DDG almost exclusively for years, but I’m not going to pretend like asking companies to make HTML/CSS-only versions of their sites is a reasonable request in the modern web environment. It can be really fun and cool to build a site without JS, but there aren’t many scenarios where it’s actually beneficial.

The replies in this thread are just plain ignorant. Basically every website uses JS heavily and disabling all JS with something like noscript is just a plain bad time.

Even in your comment, every sentence is wrong. Google searches are done with GET requests, and there are plenty of reasons to force JS other than tracking, ads, and ai.

Honestly, JS is such a core part of the web, I’m surprised it took this long.

Oh, and “other AI CEO agrees, saying just imagine a world where everyone uses our AI, isn’t it wonderful?”

“Company selling AI snake oil claims that they will maybe consider possibly using AI to run their business maybe and not immediately rehire the engineers they laid off only a few months ago, definitely because of the AI and not for any other reason.”

And

“Our custom AI has revolutionized the way we do business and saved us millions! No, you can’t see how.”

But… they aren’t even really in the same market. There’s some overlap, sure, but not entirely.

This is laughable.

And then at BrightonSEO, Marcus Tober from Semrush shared a slide that assumes ChatGPT has continued growth of 13% MoM growth rate. It would then catch up to Google in four years.

Idk, a tablespoon is the largest spoon-based unit

Thing is, it already was ubiquitous before the AI “boom”. That’s why everything got an AI label added so quickly, because everything was already using machine learning! LLMs are new, but they’re just one form of AI and tbh they don’t do 90% of the stuff they’re marketed as and most things would be better off without them.

deleted by creator

So they’re adopting a similar structure to OpenAI, a for-profit company majority controlled by a non-profit organization.

Outrage, yes, but what about decreased usage? What’s the effect on revenue and stock price? C-suite pay?

If they said or implied anything else, they would lose all leverage. The public couldn’t care less about who owns tiktok, so they need people to think they’ll lose it to have any public support.

If you’re worried about unauthorized access to the physical machine, you could always just do disk-level encryption instead or store the app’s data in something like a Veracrypt virtual disk. They’d still be able to access the data if they go through your OS/user, but wouldn’t pick anything up by accessing the drive directly.

Nothing short of E2EE can truly stop someone from accessing your data if they have physical access to the server, but disk encryption would require a targeted attack to break, and no host is wasting their time targeting your meme server. I seriously doubt they’d access it even if you had no encryption at all, since if they get caught doing that they’d get in a heap of legal trouble and lose a ton of business.

Podman is purposefully built to rely on systemd for running containers at startup. It ties in with the daemonless and rootless conventions. It’s also nice because systemd is already highly integrated with the rest of the OS, so doing things like making a container start up after a drive is mounted is trivial.

Podman has a command to generate systemd files for your containers, which you can then use immediately or make some minor tweaks to your liking.

I use podman for my homelab and enjoy it. I like the extra security and that it relies on standard linux systems like systemd and user permissions. It forces me to learn more about linux and things that apply to more than just podman. You can avoid a lot of trouble by running the containers as root and using network=host, but that takes away security and the fun of learning.