keysersoze33

Read the github readme instead - much more to the point:

https://github.com/geo-tp/ESP32-Bit-Pirate

Most of the energy consumption comes from moving data between memory and compute units rather than the math itself. While HBM stacked on a silicon interposer was a signification efficiency improvement over traditional DDR, there's still room for improvement. The high energy costs will drive this further, e.g. near-memory/in-memory compute (PIM) or tighter 3D packaging and possibly optical interconnects. (And of course training/model optimisations)

A couple of points:

- Pipewire or Pulseaudio has nothing to do with XWayland, let alone Wayland (Pulseaudio is somewhat legacy and may not be maintained as well as it used to be)

- Zoom (the last time I checked, a while ago) wasn't really well implemented on Linux (came before xdg-desktop-portal, which has standardised screen sharing across DEs and only really prioritised the GNOME specific extensions.

- The engineering effort to properly support screen sharing on Linux is a _bit_ more complicated than other platforms as we have many DEs

- Browser based screen sharing is generally the best option on Linux imho

For those unware, this is Keir the Cabbage, distant relative of https://en.wikipedia.org/wiki/Liz_Truss_lettuce

The (Arch) community is moving quickly to release scripts/tools.

Right now, this is the most up to date, consolidated utility to check for infection:

https://github.com/lenucksi/aur-malware-check

Also, the aur-request mailing lists has many delete/orhan requests coming through to undo the malicous commits:

https://lists.archlinux.org/archives/list/aur-requests@lists...

Interesting to see at the bottom/Step 5:

> Before providing the final response, create a compliance checklist to verify that every constraint has been met.

I wonder if just this statement causes the Gemini to ensure compliance or there's a separate post validation function

Admit it, deep down, our inner engeering child also wants to build a semiconductor clean room ;)

Link to the preprint paper: https://arxiv.org/pdf/2604.04721

Worth reading the conclusion - makes a good point or two regarding the cumulative effect of using AI and not only the loss of the learning through struggle/time, but also the reference point of how long tasks should take without AI (e.g. we are no willing longer afford the time to learn the hard way, which will impact the younger generation notably).

The takeaway is to vet new companies one is dealing with - even just calling them up and asking if they've AI generated any system which deals with customer/patient data.

This is going to get more common (state sponsored hackers are going to have a field day)

Unless political lobbying becomes significantly harder (globally), I don't think the growing gap between the wealthy and poor will change anytime soon.

Problem is, to tax the ultra wealthy more, it needs to be coordinated globally. Otherwise, the ultra wealthy just move to where the tax system is more attractive. Not sure if this will ever happen to be honest.

However, awareness of the problem is still important

I think most distros are waiting to see if upstream can rethink this first and evaluate what makes sense. Forking something like systemd is no easy task. Perhaps this just starts out as a patch removing the DOB changes. But in reality, distros don't want to face possible legal ramifications either. It's all a bit messy, and time will tell (I suspect for the better, but who knows).

Being 46 (and quite an active 46 year old, just finished a Skitour), am curious why the cutoff for these things tends to be 45?

This is more of a Linux kernel criticism of KASLR, but perhaps it's related as to why it's not been a priority in FreeBSD (i.e. it gives a false sense of safety and rather focus on 'proper' security hardening): https://forums.freebsd.org/threads/truth-about-linux-4-6-sec...

That's a really good point - thanks for chiming in

This smells like the beginning of entshitification at ARM. I'm not saying AMD or Intel are a whole lot better, but the move to compete with licencees of ARM tech and to cheekily use AGI in the name is not going to ensure confidence in the short or long term.

Just the basic things, like viewing the complete json (with syntax highlighting) to then determine the filter, that is '.' becomes '**'

I was a bit skeptical at first, but after reading more into jsongrep, it's actually very good. Only did a very quick test just now, and after stumbling over slightly different syntax to jq, am actually quite impressed. Give it a try

I had the same problem (brave browser)

If you think sauerkraut and kimchi make things move more, wait till you try homemade kefir!

I'm quite sensitive to it and can only have a spoonful or so and generally plan to work from home the next day. Note that the store bought stuff is often heated or similar and doesn't contain anywhere near the same level of bacteria.

I've been a Kagi Pro user for several years now, and to be honest, the models from the pro tier aren't that useful compared to the free versions of Gemini & ChatGPT.

For me, I pay for Kagi pro for search without Google/Bing enshittification, their Translate (which I use quite often while I'm learning German/working in German - better for me than Google Translate), and their Summarizer. I pay for Claude, and also occasionally use OpenRouter for my AI needs.

Thanks for the kind offer - no urgent rush though

Interesting that Weave tries to solve Mergiref's shortcomings (also Tree-sitter based):

> git merges lines. mergiraf merges tree nodes. weave merges entities. [1]

I've been using mergiraf for ~6 months and tried to use it to resolve a conflict from multiple Claude instances editing a large bash script. Sadly neither support bash out of the box, which makes me suspect that classic merge is better in this/some cases...

Will try adding the bash grammar to mergiraf or weave next time

[1] https://ataraxy-labs.github.io/weave/

I've contributed 32 edits (1 new page) in the past 10 years, so despite being stable, there are still many things to add and fix!

Sadly, the edit volume will likely drop as LLMs are now the preferred source for technical Linux info/everything...

I'm sure it was a GoPro - was mocking the journalist's use of the Ye olde term 'videotaped'

Anyone who base jumps with a bulky videotape recorder instead of GoPro deserves to arrested ;)

Sadly the OP is a newly registered account which has already submitted 2 stories

That's surprisingly relaxing – thanks for sharing

Stumbled across this second alternative: https://sites.elliott.computer/one-minute-park/

(And the original HN thread: https://news.ycombinator.com/item?id=40323785 )

If all else fails, ask for BJ