jsilence

Or get a good old LaserJet 4050 used somewhere. Indestructible workhorse.

It works on people with e specific birth gene defect named OTOH. So likely not helping with tinnitus.

Calibra might help to put some benchmark numbers on your prototypes: https://calibra.swival.dev/

Great news! Expecting this to get implemented in all the major inference runners pretty fast. See also: https://news.ycombinator.com/item?id=47637422

User @thin_signal developed a tool for mixed precision quantization on MLX. They perfoemd a sensitivity analysis across the model layers and applied less radical quantization on the layers more sensitive and more quantization tomlayers that are more robust.

The tool, which is documented here (https://mlx-optiq.pages.dev/) also implements the recently aanounced TurboQuant KV-Cache optimization, so in total this should greatly improve the quality of locally run LLMs.

Looking forward to an OptiQ release of the Gemma 4 family.

This is so annoying.

Along the same line is that you can watch any hour long video without interruptions unless it is music where you will get interrupted every couple of minutes with "are you there?" dialogues.

Why not Quarto? Genuinely curious.

The Tick live action series with Patrick Warburton is unmatched. Too bad it got cancelled after nine episodes. But those are pure gold.

Maybe teach zhem how to cook? Much cheaper.

Wondering if it were feasible to vibe-code a Bot that would engage and invest the little sums that yield those bait gains, and harvest those. Eject at the right time.

Wondering why they are not using uv instead of pip in the container, give that uv is much faster than pip.

That is part of the ‚first mover advantage‘. Sometimes operating and experimenting in grey zones before they become regulated.

The European and especially German approach of regulating pre-emptive might be more fair, but apparently it also stifles innovation, as we can observe. Almost no significant players from Europe and Germany.

A wonderful gift idea for nerd friends.

Maybe consider evaluating Marimo?

Amazing what well trained small neural networks can achieve.

Using this trick with Marimo.io notebooks in app-mode.

Instant reactive reproducible app that can be sent to others with minimal prerequisites (only uv needs to be installed).

Such a hot combo.

Also let's not forget that Quarto is the spiritual successor of the widely successful R Markdown ecosystem. By the same developers: https://quarto.org/docs/faq/#who-are-the-developers-of-quart...

Have been using Spacemacs and Doom Emacs some time ago, but turned off evil mode, bc I was not used to vim style editing. Lately I understood that vim key bindings is actually an editing language and finally found joy in vim [0]. Surely going to return to Doom-Emacs with evil mode on. Maybe switching to evil god mode later.

[0] https://www.youtube.com/watch?v=wlR5gYd6um0

I have been using a similar tool, Clipea, on a regular basis and I find it really helpful. I am CLI proficient enough to know whats possible on the shell, but I use it too infrequently to remember all the ins and outs and flags. At this point these tools really improve productivity.

Have not used shllm, but from the docs it looks good. Clipea and others have not been updated for quite some time, except llm from Simon Willison, the datasette developer. His llm implementation has a llm-cmd plugin which provides what shllm does. Works like a charm.

Over all a very useful category of AI programs.

Is there any kind of copyright or other protection on the generated Podcast audio? Or can it be redistributed at will?

Not an expert, but I think it does not have pentagons because it is in a non-euclidian space. Might be wrong, just a guess.

I have a similar situation, managing a lot of different threads in very different contexts (teaching and projects) at our university.

I miss the state of flow very much, not being able to focus on one thing and diving deep for a couple of hours or even a couple of days. The constant context switch and firehosing of incoming messages I have to fend off is draining.

Just as you say, some days feel like having been completely busy while not achieving anything of substance. I know I am lubricating others work, but I'd love to be with my own toughts for a while also.

For the next project I am attempting to dedicate full and exclusive days to the different contexts, so that at least on that single day my mind is not concerned with all the contexts, threads and their TODOs.

I am already tagging my incoming emails with "TODO" and "waiting for reply" and sorting them into their respective folders and then afterwards heading for the tag stacks and working on them. I think I will add tags for contexts and set up filters, so that I can open the right TODO stack for the topic of the day I am working on, to conceal any "not today" work.

Internet via Ham Radio aka Packet Radio is a thing: https://themodernham.com/ip-over-ham-radio-via-new-packet-ra...

Pair it with the Gemini protocol and you're there: https://en.wikipedia.org/wiki/Gemini_(protocol)

Just saw this by chance in the new section and I think it is really compelling. Curious why it is not getting any traction here.

No mentioning of the energy requirements. I can see they are using artificial light exclusively. I am certain they will not be economically competitive on the markets.

Anybody know of a similar service for weather information?

If only it would support Quarto flavored Markdown, that would be awesome.

Was recently kajaking in KayakVR in the Arctic scenery. Wife snuck by and opened the window. Felt real.

Anyone pitting the Biden-Bot and the Donald-Bot against each other and curating the most hilarious parts? For funsies?

Hoping that future drought events don't have a massive impact on forests. The 2018 drought in Germany already induced some heat and drought stress on trees. If the trees die, they will not take up 20% more CO2.