The more I play around with #Plan9 the more I think we took a serious wrong turn back in the 90s.

Well I guess we get to see if Claude Sonnet is useful at all, since neither Opus or Fable are working!

When did the term rogue-like become associated with all kinds of games which are not randomized CRPG dungeon-crawlers? Randomly generated worlds seems to be a stretch.

Haha! Let's make everyone in the U.S. use the nerfed models, so the rest of the world can wreck us with the exploits the Chinese open models can find.

I struggle to find any of my projects where Fable is actually a useful product.

OpenAI's Bot is a blight on the Internet. No other LLM crawler hits this often or hard.

Seriously.. I’ve seen horrors you cannot believe. LLM slop is honestly MUCH better in many cases than legacy I’ve encountered.

So many API errors in Claude Code in the last 24 hours. Really killing velocity on this project. Fortunately, not under a timeline.

New model dropped from Anthropic, and 3/4 of my projects give me: ⏺ Fable 5's safety measures flagged this message for cybersecurity or biology topics. They may flag safe, normal content as well. These measures let us bring you Mythos-level capability in other areas sooner,

The reason software eats RAM is the same reason factories used to dump chemicals in rivers. The cost is externalized. Every mass of inference compute shows up on an engineering manager's AWS bill, broken down to the cent, reviewed quarterly. Every mass of RAM consumed on YOUR machine shows up nowhere in anyone's budget. Chrome could cut memory usage by 60% tomorrow and Google's revenue wouldn't move a single basis point. Docker's 2GB idle footprint costs Docker Inc. exactly $0. Electron's 500MB todo list costs the Electron team exactly $0. The user paid for the RAM. The user pays the electricity. The user deals with the fan noise. The company ships faster because they chose the laziest possible runtime. The token-optimization obsession makes this even clearer. Companies optimize inference cost because inference cost hits their margins. They'll spend six months shaving 200ms off a model response. They won't spend six days reducing a desktop client's memory footprint because that memory belongs to someone else's hardware. This is why the 16GB vs 32GB debate is a trap. You're asking consumers to buy more expensive hardware to subsidize the software industry's refusal to optimize for a resource they never have to pay for. The market will never fix this on its own. The people writing the checks and the people running out of RAM are on opposite sides of the transaction.

Ugh.. why did I forget the prime directive when coming back to this hell site? Just block. Don't engage otherwise your algorithm becomes a cesspool.

What is it with all these CLI projects? Why are we trying to turn our personal graphical super computers back into VT100s?

I just had an epiphany. All these AI Accelerationists actually believe the Capitalists building towards AGI will actually usher in a Utopia. Such naivety.

I liked the old "Fail Whale" better.

Oh look everyone! The Fail Whale graphic's been changed.

All of you ninnies calling for more laws for this or that... How about we enforce the EXISTING laws that already deal with your concerns instead? We need fewer laws not more!

The great thing about the Linux ecosystem is that you can find tools and libraries to do almost anything. The bad thing about the Linux ecosystem is that there's almost zero consensus and you wind up with multiple versions of libraries doin the same concepts installed.