Dave.

Owned by Akami now for 4 years, I think.

I’ve several small linodes and have been using the platform since 2018 or so. No difference seen so far in the linode plans or features I’m using with Akami, except for more in-depth monitoring available (for a price).

Now if you turn anything off (like that stupid autobrake that brakes for cars in the left hand turn lane beside me), they’ll use that to say “it’s not our fault, the driver turned it off!”

Don’t try and spin this as a convenient liability-limiter for manufacturers.

Unless it is an actual failure of a safety critical system, with no warning, that then directly leads to harm, manufacturers are in no way responsible for how you operate the chunk of metal and technology that you bought from them.

All this stuff absolutely helps to reduce the overall risk of injury when operating a vehicle. Is it also a pain in the ass sometimes? Sure it is. Can you turn it off? Sure you can. But sometimes that shutoff is just temporary, because governments - not manufacturers - mandate that some things are now required, because they’re sick of hosing body parts off of highways.

Even slow RAM is a thousand times faster than swap. Search for “memory hierarchy” for a better idea of why.

Oh man that’s sweeeeet. This is the kinda shit I want to see at trade shows , not yet another iteration of “world’s thinnest phone with a chunky camera bulge”.

Oh your new phone can fold in half and it’s got a 600MP camera and it has a 37Ah battery that lasts for 3 days even when it’s churning on its internal navel-gazing AI 24/7? That’s greaaaaat.

But I want crazy shit!

I want a row of clicky buttons down one side as direct shortcuts to apps.

I want a TV tuner and a pull-out antenna so I can watch Judge Judy serve some proper smackdown to some entitled neighborhood dispute when I’m on the bus, nearly poking someone’s eye out when we hit a pothole.

I want a phone with a laser pointer and an IR blaster (again).

I want a phone that looks like the palmtops of old, with a clamshell case and a little hardware chicklet keyboard and it runs on AA’s, except it’s also a phone.

I want a phone that slots into a boom box and fully controls and plays tapes with auto-reverse.

I want a phone that’s a drone, throw it in the air and it takes group photos.

I want a phone that has gyros and can self-balance on one of it’s corners, and do little party tricks, and can hop along behind you if you leave it behind somewhere.

I want a phone that’s also a mini 1080p laser projector.

I want its calculator to also have an I Ching mode and print “a suffusion of yellow” instead of any answer greater than 4.

Where’s all the FUN stuff?

The article is quite fixated on how dialup was around in the early 2000s, and how Americans used dialup in the early 2000s with screeching modems, and how AOL turned off their dialup last year and only a thousand people were using dialup by then.

But there’s no indication that this has anything to do with the application, except for maybe everyone used a popular means to connect to the Internet to use it when it was first built.

Now if someone maintained a bank of modems and people legitimately called them up and connected to the system like it was an ISP, and it was secure by way of that was the only way to access it as opposed to over the Internet, then sure, I can see that maybe it’s time to work on that part.

Firstly, suspend might be an option compared to hibernation for your subsequent startups if you can suffer a slow startup the very first time.

Otherwise look up “Linux from scratch” .

A lot of boot time is decompressing and mounting initrd, loading modules there, mounting the root filesystem, jumping to that.

Build a kernel that has exactly the minimum modules needed for your system, skip initrd and jump straight to a root filesystem and call your init binary (it can be BusyBox even) do exactly what’s needed there to get your network+graphics card up and running with a single shell script, done.

Also look at COW filesystems. Read only base with an overlay file for writes. Discard that file on each boot and you basically have a system that is always at the same state on startup.

Complains about incomprehensible short commands in legacy shells.

Builds own shell with a new set of incomprehensible short commands, minus the hookers and blackjack of existing shells.

(Insert xkcd standards comic here)

Consider only exposing a VPN service, like Wireguard.

Then when you are away from home, it’s one tap / click to be back on your LAN again.

You’re not thinking black-swan enough.

You’re typing your comments using a blob of goo with about a hundred million neurons in it that cycles under a hundred hertz and draws less than 20 watts.

I don’t think that we’ll be running packs of goo in our PCs any time soon. But I do think some entirely different way of looking at the problem will emerge that will reduce computational requirements by many orders of magnitude. And it won’t involve gigantic statistical engines trying to find the best average response to a question.

I’m just waiting for the closed-source AI industry to have their Black Swan moment.

Something’s going to come out of left field from the open AI community and all that investment in proprietary models and mega-compute is going to be rendered useless.

Here’s a little feedback for you: Never trust the user’s browser.

Ctrl-Shift-I -> Console -> "this.proManager.enablePro()" -> <Enter>

It’s worth nothing if it hasn’t been graded and placed in a sealed plastic box by a grading company that asserts that grading is awesome and things are worth nothing if they aren’t graded.

(sad \s)

Later versions of windows recognise that the device is removable and don’t cache writes. From the users point of view the copy dialog box only closes when writes are complete.

I’ve been working on updating all my old software projects lately, and as part of that process I feed the source code into a LLM for review.

The amount of simple security errors, logic flaws, and code smells that it reveals is quite embarrassing.

This was all “good functional code” that’s been used internally for years. Clearly it worked well enough, but a simple pass through a LLM reviewer made it a lot more robust and secure.

Every large AI startup is racing to be the last one standing. They are all gobbling up compute resources to ensure that they have the best competitive advantage when others go broke. That includes soaking up all the future supply they can to restrict their competitors ability to do the same. In < 5 years time there is only going to be one commercial AI company left, and all the rest will just be skeletons picked clean of all their resources.

Be prepared to live with whatever hardware you have now for about 10 years. If you’re lucky things will be back on an even keel by then.

I work on heavy mining equipment and I tell the techs who are looking for a laptop to just buy something cheap because when you drop it off the top of a machine it’s still just as broken.

Statistically your weird internal corpo program that was made bespoke over a decade several decades ago …

Won’t work at all because it lives on a shared drive that needs your windows login and your PC to be joined to the domain to reach it, and IT just said “lol no” to your request to join your home made windows VM to the domain.

Get some M-Discs and burn a couple of copies of your docs and photos. The theoretical lifetime of M-disc is ~1000 years. Good luck finding a Blu-ray drive in the 30th century though.

That’s quite an odessy. I put MX Linux on my 13 year old dell laptop, and there was an option in the MX utilities app to install the NVIDIA 340 drivers for it, and it merged them into the 7.0.something kernel that I had in about 3 minutes.

This isn’t meant as snark. Get settled into things for a few months with what you’ve got and then boot a few distros off USB and see if there’s something that better suits you or your hardware. You’ll have a good idea by then if what you really want/need.