So, it’s not really a problem I’ve run into, but I’ve met a lot of people who have difficulty on Windows understanding where they’ve saved something, but do remember that they’ve worked on or looked at it at some point in the past.

My own suspicion is that part of this problem stems from the fact that back in the day, DOS had a not-incredibly-aimed-at-non-technical-users filesystem layout, and Windows tried to avoid this by hiding that and stacking an increasingly number of “virtual” interfaces on top of things that didn’t just show one the filesystem, whether it be the Start menu or Windows Explorer and file dialogs having a variety of things other than just the filesystem to navigate around. The result is that you have had Microsoft banging away for much of the lifetime of Windows trying to add more ways to access files, most of which increase the difficulty of actually understanding what is going on fully through the extra layers. But regardless of why, some users do have trouble with it.

So if you can just provide a search that can summon up that document where they were working on that had a picture of giraffes by typing “giraffe” into some search field, maybe that’ll do it.

I don’t use OpenHAB or Home Assistant, but I’d be extremely surprised if they don’t have existing functionality for connecting microphones, speakers, and LLMs to set up voice-controlled stuff.

searches

https://heywillow.io/

Willow Is a Practical, Open Source, Privacy-focused Platform for Voice Assistants and Other Applications

Willow is an ESP IDF based project primarily targeting the ESP32-S3-BOX hardware family from Espressif. Our goal is to provide Amazon Echo/Google Home competitive performance, accuracy, cost and functionality with Home Assistant, openHAB and other platforms.

100% open source and completely self-hosted by the user with “ready for the kitchen counter” low cost commercially available hardware.

https://rhasspy.readthedocs.io/en/latest/

Rhasspy (ɹˈæspi) is an open source, fully offline set of voice assistant services for many human languages that works well with:

• Hermes protocol compatible services (Snips.AI)
• Home Assistant and Hass.io
• Node-RED
• Jeedom
• OpenHAB

We’d have achievement parties, work on multiplayer cheevos, and just compete for the highest gamescore in general.

I take it that “cheevo” is what people are calling achievements now.

My understanding from a very brief skim of what Microsoft was doing with Copilot is to take screenshots constantly, run image recognition on it, and then make it searchable as text and have the ability to go back and view those screenshots in a timeline. Basically, adding more search without requiring application-level support.

They may also have other things that they want to do, but that was at least one.

EDIT: They specifically called that feature “Recall”, and it was apparently the “flagship” feature of Copilot.

Not the position Dell is taking, but I’ve been skeptical that building AI hardware directly into specifically laptops is a great idea unless people have a very concrete goal, like text-to-speech, and existing models to run on it, probably specialized ones. This is not to diminish AI compute elsewhere.

Several reasons.

• Models for many useful things have been getting larger, and you have a bounded amount of memory in those laptops, which, at the moment, generally can’t be upgraded (though maybe CAMM2 will improve the situation, move back away from soldered memory). Historically, most users did not upgrade memory in their laptop, even if they could. Just throwing the compute hardware there in the expectation that models will come is a bet on the size of the models that people might want to use not getting a whole lot larger. This is especially true for the next year or two, since we expect high memory prices, and people probably being priced out of sticking very large amounts of memory in laptops.

• Heat and power. The laptop form factor exists to be portable. They are not great at dissipating heat, and unless they’re plugged into wall power, they have sharp constraints on how much power they can usefully use.

• The parallel compute field is rapidly evolving. People are probably not going to throw out and replace their laptops on a regular basis to keep up with AI stuff (much as laptop vendors might be enthusiastic about this).

I think that a more-likely outcome, if people want local, generalized AI stuff on laptops, is that someone sells an eGPU-like box that plugs into power and into a USB port or via some wireless protocol to the laptop, and the laptop uses it as an AI accelerator. That box can be replaced or upgraded independently of the laptop itself.

When I do generative AI stuff on my laptop, for the applications I use, the bandwidth that I need to the compute box is very low, and latency requirements are very relaxed. I presently remotely use a Framework Desktop as a compute box, and can happily generate images or text or whatever over the cell network without problems. If I really wanted disconnected operation, I’d haul the box along with me.

EDIT: I’d also add that all of this is also true for smartphones, which have the same constraints, and harder limitations on heat, power, and space. You can hook one up to an AI accelerator box via wired or wireless link if you want local compute, but it’s going to be much more difficult to deal with the limitations inherent to the phone form factor and do a lot of compute on the phone itself.

EDIT2: If you use a high-bandwidth link to such a local, external box, bonus: you also potentially get substantially-increased and upgradeable graphical capabilities on the laptop or smartphone if you can use such a box as an eGPU, something where having low-latency compute available is actually quite useful.

I know open ai bought ~40% of microns memory production.

IIRC Micron was the only Big Three DRAM manufacturer that OpenAI didn’t sign a contract with. I think that they signed contracts with SK Hynix and Samsung for their supply, and didn’t with Micron.

searches

Yeah:

https://techcrunch.com/2025/10/01/openai-ropes-in-samsung-sk-hynix-to-source-memory-chips-for-stargate/

OpenAI ropes in Samsung, SK Hynix to source memory chips for Stargate

Not signing was actually probably to Micron’s advantage; I understand that OpenAI didn’t let Samsung know that they were negotiating with SK Hynix and didn’t let SK Hynix that they were negotiating with Samsung and signed both deals concurrently. That is, each of Samsung and SK Hynix probably sold the DRAM that went to OpenAI for less than they could have gotten on the open market, since neither was aware at the time of signing that the supply on the open market outside of themselves would sharply decrease during the period of the contract, which would be expected to drive up prices.

I mean, they still made a lot more money than they had been making. Just that they could have probably managed to get even more money for the DRAM that they sold.

IIRC the 40% number was OpenAI signing for 40% of global production output, not for any particular company’s output.

According to TrendForce, the boom is expected to continue, as conventional DRAM contract prices in 1Q26 are forecast to rise 55–60% QoQ, while NAND Flash prices are expected to increase 33–38% QoQ.

And that’s in an environment where DRAM output is significantly ramping up:

https://www.theverge.com/news/847344/micron-ram-memory-shortage-2026-earnings

Micron aims to ramp up production and expects to increase its shipments of DRAM and NAND flash memory by 20 percent next year

https://www.tweaktown.com/news/109011/sk-hynix-to-boost-dram-production-by-a-huge-8x-in-2026-still-wont-be-enough-for-ram-shortages/index.html

SK hynix to boost DRAM production by a huge 8x in 2026, still won’t be enough for RAM shortages

EDIT: Also:

The memory sector has surged into a full-blown seller’s market, with both South Korean giants rejecting long-term agreements (LTAs) of two to threes years and sticking to quarterly contracts, anticipating stepwise DRAM price increases each quarter through 2027, the report suggests.

Just a few years ago, they were losing a ton of money due to low DRAM prices, so I imagine that rejecting long-term contracts at current (already high) prices drives even more home that they expect demand to increase further relative to supply:

https://www.marketscreener.com/quote/stock/SK-HYNIX-INC-6494929/news/SK-Hynix-reports-Q2-loss-as-chip-glut-continues-44417161/

SK Hynix reports Q2 loss as chip glut continues

SEOUL (Reuters) -South Korea’s SK Hynix posted a quarterly operating loss on Wednesday, as the company said the memory chip market is beginning to recover from a deep downturn.

Published on 07/25/2023 at 07:29 pm EDT

“Open source” really isn’t the right term here, if they’re just releasing API specifications. “Open sourcing” the speakers would be releasing the source code to the software that runs on the speakers.

Like, all of Microsoft’s libraries on Windows have a publicly-documented interface. That hardly makes them open source. Just means that people can write software that make use of them.

He’s still not sure Niemannn cheated, though. “It is of course suspicious,” he said. “But it could be luck or it could be that Magnus had a bad day… maybe it’s not even possible to do this. That’s why I thought to make this program. Let people try. Maybe if people figure out it doesn’t even work at all, then this whole theory of butt plugs was just a waste of time.”

Hmm. Actually…you probably can actually determine it. Assuming that the device’s radio is talking Bluetooth, which not all do, if anyone had a cell phone near the environment, and made use of Google’s or Apple’s Location Services, said companies probably have a log of it responding to beacons. Those location services work by broadcasting a beacon to Bluetooth and WiFi devices and uploading their MAC address and signal strength of the response to Google and Apple, who then compare it to prior position reports and IDs and strengths to determine a position, so there’ll be a log of devices and their active periods floating around in their databases.

looks confused

searches

Ah.

https://www.vice.com/en/article/did-hans-neimann-cheat-at-chess-with-a-sex-toy-this-coder-is-attempting-to-find-out/

A cheating controversy rocking the chess world just won’t let up. One conspiracy theory promoted by Elon Musk without evidence is that young chess wiz Hans Niemannn defeated world chess champion Magnus Carlsen in early September with the aid of a vibrating set of anal beads.

It’s an intriguing idea, but is such a thing even possible? Ron Sijm, a software engineer in the Netherlands, wants to find out and has developed software to test the theory. He’s posted the code to open-source coding platform GitHub, and all he needs now is the right sex toy.

With the code built, Sijm started hunting for a butt plug or set of anal beads to test his theory. He’s turned to a community that knows the systems best, the butt plug sex toy control project Buttplug.io. Sijm has been talking with the folks on Buttplug.io Discord server in an effort to find someone who already has a device and is willing to test the software.

Sijm said coding out the basic software took about four hours and that, hypothetically, it would be easy for someone like Niemann or his team to put together. The list of compatible anal vibrating devices is long.

https://hypebeast.com/2026/1/razer-project-ava-ai-companion-3d-hologram-kira-zane-faker-sao-release-info

Razer has officially pulled the curtain back on Project AVA, a “Friend for Life” AI desk companion featuring a 5.5-inch 3D holographic display. Moving beyond simple voice assistants, AVA utilizes human-like vision and audio sensing to provide full contextual awareness, acting as a real-time gaming wingman, professional consultant, and personal organizer.

The hardware is a sleek cylindrical unit equipped with a dual far-field mic array, an HD camera with ambient light sensors, and Razer’s signature Chroma RGB. At its core, the device currently leverages xAI’s Grok engine to power its “PC Vision Mode,” allowing it to analyze on-screen gameplay or complex documents via a high-bandwidth USB-C connection to a Windows PC.

For some time, man had suffered in a world lacking a smart speaker with a camera, tits, short skirt, and ability to monitor everything he did on his computer. That world was about to end.

Web devs need hardware integration support too.

They said they thought they were within Github’s acceptable use guidelines; even though they make mods for hentai games and things like interactive vibrator plugins, they took care to not host anything explicit directly in their repositories.

A developer who goes by Sauceke, who Github suspended in mid-November without explanation, said their open-source adult toy mod users are now encountering broken links or simply can’t find any of their work.

Hmm. Buttplug.io’s GitHub repositories are still up, and I’d think that that’d be rather-more-prominent if the issue is sex toy code.

According to ShinyHunters, the records contain extensive data on Premium members including email addresses, activity type, location, video URL, video name, keywords associated with the video and the time the event occurred. Activity types include whether the subscriber watched or downloaded a video, or viewed a channel and events include search histories.

This sort of thing is one of those examples why “no log, no profile” service is probably a good idea. The service could have offered the option to charge a fee for access, but not retain customer activity data. They didn’t do that. At some point down the line, someone got ahold of the data, which I imagine that their customers are not really super keen on having floating around attached to their identities.

Probably a lot of companies out there that log and retain a lot of data about their customers.

I’m not familiar with FreshRSS, but assuming that there’s something in the protocol that lets a reader push up a “read” bit on an per article basis — this page references a “GReader” API — I’d assume that that’d depend on the client, not the server.

If the client attempts an update and fails and that causes it to not retry again later, then I imagine that it wouldn’t work. If it does retry until it sets the bit, I’d imagine that it does work. The FreshRSS server can’t really be a factor, because it won’t know whether the client has tried to talk to it when it’s off.

EDIT: Some of the clients in the table on the page I linked to say that they “work offline”, so I assume that the developers at least have some level of disconnected operation in mind.

The RSS readers I’ve always used are strictly pull. They don’t set bits on the server, and any “read” flag lives only on the client.

Secondly, is there a benefit to creating an LVM volume with a btrfs filesystem vs just letting btrfs handle it?

Like, btrfs on top of LVM versus btrfs? Well, the latter gives you access to LVM features. If you want to use lvmcache or something, you’d want it on LVM.

The good news is that single-player games tend to age well. Down the line, the bugs are as fixed as they’re gonna be. Any expansions are done. Prices may be lower. Mods may have been created. Wikis may have been created. You have a pretty good picture of what the game looks like in its entirety. While there are rare cases that games are no longer available some reason or break on newer OSes with no way to make them run, that’s rare.

With (non-local) multiplayer games, one has a lot less flexibility, since once the crowd has moved on, it’s moved on.

It looks like it doesn’t support ARM architecture systems at all, or anything other than x86_64.

https://docs.bazzite.gg/Gaming/Hardware_compatibility_for_gaming/

Minimum System Requirements

• Architecture: x86_64

Lead image seems a cyanide and happiness cartoon,

It doesn’t claim to be a Cyanide and Happiness cartoon anywhere in the cartoon.

Systemd is unnecessary bloat. Actually, having an init system is unnecessary bloat. All one really needs is the Linux kernel and emacs.

http://informatimago.free.fr/i/linux/emacs-on-user-mode-linux.html