Yeah, I also get that vibe from the article.

To me the article isn’t convincing. There are some interesting points raised. Like “the llm sharing memories between a team of developers”.

I mostly use llms individually, so this is a real blind spot for me. But I might convinced to share a corpus of memories between developers if it ever becomes practical.

But for now aren’t context windows still sometimes smaller than the task at hand… so those “llm memories” take the form of literal documentation???

The closer llm memories get to native format, i.e: stored tokenized content, don’t we lose compatibility between setups anyhow. What if you’re using fp16, and I’m using nvfp4?

I'll take my father's HP 49g to my grave. but if TI wanted to flirt with me, all it would take is a setting to enable Reverse Polish Notation. (I did check the features, and no mention).

According to the wiki, Quarkdown supports cross-references in the same document, but not cross-references across many documents.

The wiki explains Quarkdown’s concept of subdocuments. A subdocument gets created when linking to another Quarkdown/Markdown file and populates a graph. Subdocuments inherit the linker’s properties but are then independent sandboxes that get compiled independently.

A cross-subdocument cross-reference is such a niche use case that I don’t feel like it should break this design choice.

>> Does not describe a way to prevent [dendrites], so solid-state batteries don't seem like a realistic product for now.

Every lithium battery in use today is susceptible to dendrites. One of the promises of solid state batteries is more safety, not less.

Agree, but as of now it's seems like it's still a promise, not a reality.

This is an article about an academic paper investigating a mechanism behind this effect, it doesn't say anything about the relative viability of the technology in practice.

Thank you for supporting Phoronix. They are the #1 review I keep an eye on for linux laptops.

This article is a horrific bait and switch. The title is "Reading is Magic", but towards the end the tone is literally:

> … all struggles will be powered by outright sexual sadism.

If that’s where we’re heading, how about a different title and at least a little heads up in the introduction?

Personally, I wouldn’t have wasted my time.

I'm sure you know about pandoc for translating markdown into html (and all it's other tricks).

Yea but that still requires a non-zero amount of time to setup and most importantly maintain and keep up to date.

Pure HTML generated from a text file just works and probably will forever.

I mean it’s just my personal website which is mostly just for me to look up things quickly / personal wiki

But the text file has some markup syntax beyond human language? Point being LLMs are subpar for acting on formal grammars, like cracking a nut with a sledgehammer. That's why its important tools like 11ty and pandoc remain.

That’s somewhat true (in my case it’s it’s laughably simple though).

I also never said that tools like pandoc are obsolete now. Just in my case they are already overpowered and I might migrate to something simpler soon. Otoh i might just run the current version of 11ty indefinitely and never upgrade.

To repurpose an old idiom: Not even a dozen AGI agents could make a baby in 6 months.

But yeah, your point stands.

FYI: muppetman’s profile says “don’t take me seriously”, so I think this is a joke?

While tone often portrays poorly over text, I think this is an example where the sarcasm is very overt. I don’t think anyone would think the comment is serious.