My Emacs config heavily used the Hyper key: although I'm not japanese I'm using japanese keyboards (I've got several HHKB Pro JP, since years and years and years) and they've got more modifiers than regular US/EU keyboards. So I remapped the Henkan/Muhenkan (or whatever modifier whose japanese name I forgot) thinggies to be Hyper.

I can't even begin to imagine how complicated it'd be to get that working in "emacs -nw" (Emacs No Window, aka terminal)...

Thankfully I don't have no real need for terminal Emacs.

[1] Space-cadet keyboard with "Hyper" keys and, overall, really a lot of modifier keys: https://en.wikipedia.org/wiki/Space-cadet_keyboard#/media/Fi...

Regardless, that’s kind of the beauty of this project: term-keys generates the terminal config file based on all your bespoke keybindings for you. Would probably be easier to setup than you imagine.

> As someone who loves immich, a VM is overkill but also without vGPUs or external configuration loses you access to some the best features, local AI searching akin to what google and apple photos offer.

I installed immich in a VM. And the VM is using GPU passthrough. I don't see how it's overkill: immich is a kitchen sink with hundreds if not thousands of dependencies and hardly a month goes by without yet another massive exploit affecting package managers.

I'm not saying VM escapes exploit aren't a thing but this greatly raises the bar.

When one install a kitchen sink as gigantic as immich, anything that can help contain exploits is most welcome.

So: immich in a VM and if you want a GPU, just do GPU passthrough.

I agree that the search using facial recognition is nice in immich that said.

> For general computer usage, SSDs really were a once in a generation "holy shit, this upgrade makes a real difference" thing.

The last one were I really remember seeing a huge speed bump was going from a regular SSD to a NVMe M.2 PCIe SSD... Around 2015 I bought one of the very first consumer motherboard with a NVMe M.2 slot and put a Samsung 950 Pro in it: that was quite something (now I was upgrading the entire machine, not just the SSD, so there's that too). Before that I don't remember when I switched from SATA HDD to SATA SSD.

I'm now running one of those WD SN850X Black NVMe SSD but my good old trusty, now ten years old, Samsung 950 Pro is still kicking (in the wife's PC). There's likely even better out there and they're easy to find: they're still reasonably priced.

As for my 2015 Core i7-6700K: it's happily running Proxmox and Docker (but not always on).

Even consumer parts are exceptionally reliable: the last two failures I remember, in 15 years (and I've got lots of machines running), are a desktop PSU (replaced by a Be Quiet! one), a no-name NVMe SSD and a laptop's battery.

Oh and my MacBook Air M1's screen died overnight for no reason after precisely 13 months, when I had a warranty of 12 months, (some refer to it as the "bendgate") but that's because first gen MacBook Air M1 were indescribable pieces of fragile shit. I think Apple got their act together and came up with better screens in later models.

Don't worry too much: PCs are quite reliable things. And used parts for your PC from 2014 wouldn't be expensive on eBay anyway. You're not forced to upgrade to a last gen PC with DDR5 (atm 3x overpriced) and a 5090 GPU.

It’s motherboards that tend to fail the most. I had one fail this year, albeit for the first time.

fyi someone or something is downvoting your recent posts to oblivion, and i didn't see any obvious reason.

> Google will do just what Microsoft did with Internet Explorer and bundle Gemini in for 'Free' with their already other profitable products and established ad-funded revenue streams.

Just some numbers to show what OpenAI is against:

    GMail users: nearing 2 billion
    Youtube MAU: 2.5 billion
    active Android devices: 4 billion (!)
    Market cap: 3.8 trillion (at a P/E of 31)

So on one side you've got this behemoth with, compared to OpenAI's size, unlimited funding. The $25 bn per year OpenAI is after is basically a parking ticket for Google (only slightly exaggerating). Behemoth who came with Gemini 3 Pro "thinking" and Nano Banana (that name though) who are SOTA.

And on the other side you've got the open-source weights you mentioned.

When OpenAI had its big moment HN was full of comments about how it was game over for Google for search was done for. Three years later and the best (arguably the best) model gives the best answer when you search... Using Google search.

Funny how these things turns out.

Google is atm the 3rd biggest cap in the world: only Apple and NVidia are slightly ahead. If Google is serious about its AI chips (and it looks like they are) and see the fuck-ups over fuck-ups by Apple, I wouldn't be surprised at all if Alphabet was to regain the number one spot.

That's the company OpenAI is fighting: a company that's already been the biggest cap in the entire world and that's probably going to regain that spot rather sooner than later and that happens to have crushed every single AI benchmark when Gemini 3 Pro came out.

I had a ChatGPT subscription. Now I'm using Gemini 3 Pro.

You just made it clear who needs to acquire openai.. it's going to be Apple! (Jonny Ive already there).

And great points on the Google history.. let's not forget they wrote the original Transformers paper after all

the branding is all wrong. I could see Apple buying anthropic, but OpenAI is exactly the wrong ai company for Apple. openai is the tacky, slop based ai company. their main value is the brand and the users, but Apple already has a strong brand and billions of users. Apple needs an ai company with deployment experience and a good model, but paying for a brand and users doesn't make sense for them.

Maybe it'd then be a good idea to have labs secretly funded by a joint venture half-US, half-Chinese, in China, doing gain-of-function research on these?

And then maybe that if some shit hits the fan, it'd then be a great idea to ask someone neck and tie deep in that funding and in that research to act as the "expert" to tell us if we should put masks on or not once it leaks?

So basically if you like to put SSDs on shelves (for offline backups), you should read them from scratch once in a while?

I rotate religiously my offline SSDs and HDDs (I store backups on both SSDs and HDDs): something like four at home (offline onsite) and two (one SSD, one HDD) in a safe at the bank (offline offsite).

Every week or so I rsync (a bit more advanced than rsync in that I wrap rsync in a script that detects potential bitrot using a combination of an rsync "dry-run" and known good cryptographic checksums before doing the actual rsync [1]) to the offline disks at home and then every month or so I rotate by swapping the SSD and HDD at the bank with those at home.

Maybe I should add to the process, for SSDs, once every six months:

    ... $  dd if=/dev/sda | xxhsum

I could easily automate that in my backup'ing script by adding a file lastknowddtoxxhash.txt containing the date of the last full dd to xxhsum, verifying that, and then asking, if a SSD is detected (I take it on a HDD it doesn't matter), if a full read to hash should be done.

Note that I'm already using random sampling on files containing checksums in their name, so I'm already verifying x% of the files anyway. So I'd probably be detecting a fading SSD quite easily.

Additionally I've also got a server with ZFS in mirroring so this, too, helps keep a good copy of the data.

FWIW I still have most of the personal files from my MS-DOS days so I must be doing something correctly when it comes to backing up data.

But yeah: adding a "dd to xxhsum" of the entire disks once every six months in my backup'ing script seems like a nice little addition. Heck, I may go hack that feature now.

[1] otherwise rsync shall happily trash good files with bitrotten ones

> It’s the cheapest way to build effective drama, but if you don’t fully dissolve yourself in the movie logic, the whole time you want to scream, “can’t anyone just talk about what’s happening directly?!”

Yup it's insane. At the end of a very long series of three movies I told my father: "OK so this all basically happened because the person who sent the letter considered the (snail) mail service to be flawless and didn't bother to make sure the recipient got the letter in the first place".

Doesn't matter which (french) movies: some dumb plot where relatives don't know they're relatives because the only person who knew didn't bother to make sure the letter explaining they were relatives arrived.

Not naming the movies otherwise we'll get nitpicking.

TFA is right: it happens all the time in movie plots and really doesn't help with the suspension of disbelief.

And a poster of Lenna is on the wall of the Richard Hendricks character in the Silicon Valley series. Which makes sense as he's working on a compression algorithm.

Just a heads-up: you seem to be shadow-banned, all your comments are auto-dead.

I can see them?

I just vouched for a few of them, maybe that's why.

do you have showdead turned on?

Refusing to pay a ransom and instead giving the money to the "ennemies" of the attackers isn't "virtue signaling" (as someone already commented: it's a "fuck you" to the attackers).

In french we call that a "pied de nez". "Turning the table" / "Poetic justice" / "Adding insult to injury" would all be more correct than "virtue signalling".

If there was no attacker and the company gave half a mil out of nowhere to a security company (or a charity) and boasted publicly about it, that would be virtue signalling.

But refusing to pay the ransom and giving the exact same amount to security researchers is just a big, giant, middle finger.

And a middle finger ain't no virtue signalling.

If they wanted to meaningfully give a middle finger to the attackers they’d be spending the money lobbying for a ransomware payments ban, not throwing away money by giving it to universities that have a plenty of money and will probably do absolutely nothing to reduce ransomware attacks in the foreseeable future.