BL4CKP1XX13 [they/it]

why the fuck

Because, despite us being in the Information Age, and computers surrounding us everywhere, only a very small portion of the population consider that learning how they work may be useful.

Username is for the project, so again, just waited until they were ready.

Well… Yeah? Doesn’t say anything about how long this has been in development for, could’ve just waited until they were ready before buying a domain.

It really is.

I love the Linux community, because this memory shortage has prompted two quite significant developments. Firstly, quite a substantial speedup of the Zram/Zswap (whichever one it is) systems (and a little improvement on compression), secondly, a potentially whole new memory compression stack referred to as CRAM that could allow reading compressed memory directly at near-DRAM speeds.

While making current memory management more efficient won’t in any case effectively solve the memory shortage, as none of these savings are comparable to effectively giving you a free extra stick, it is awesome that people have spent significant effort on this front.

Ew no what the fuck why would I want Teams to run in my brain it’s bad enough on a Windows desktop.

Oh look another reason for my “reasons to not buy modern cars” pile.

No way burning the lumps of carbon increases the amount of emitted carbon. :surprised_pikachu

Because only a smart president would be insecure enough to put in paranthesis that they have a “High IQ!”

Yes, unfortunately they are, and I do think efficiency is going to be a significant research front for open-weight models. The nature of this topic is highly speculative as our compute capabilities have only recently reached what is required to consider running generative AI models, what we have today are very crude first implementations of what I personally believe will become an everyday tool for developers, and more.

And we have seen this, there are models now capable of running on an individual’s hardware (and not particularly expensive hardware either) that can outperform what ChatGPT initially launched with.

Unfortunately not yet, no, true FOSS models are likely many years away, but I would argue that that follows typical FOSS lifecycles. Emerging technology is typically outperformed by proprietary endeavors, which creates an audience, and then that audience undertakes a FOSS implementation that initially underperforms, then just about competes, then eventually overtakes (i.e, GNU/Linux).

If the model is adequately FOSS, i.e. open weights, and can run on a single consumer GPU (or NPU), and the “author” (quotations because I’m personally undecided if one can claim code generated by an aforementioned model is theirs) understands it, then I really don’t see what the problem is…

Personally I have nothing against SLMs/LLMs as a technology, to me my grievances against ChatGPT or Claude are mostly about their environmental impacts and selling us back our own art, also keeping knowledge behind a for-profit black-box - if those aren’t appropriate for a specific model, then I say using that model is fair, and good for productivity.

Not for those of us on the dbzer0 instance 😭

Damn right it’s a national security concern, let the US keep it’s spyware to itself.

Not at all, it’s a Markdown file, plaintext. You just instruct your agent to assimilate that file into it’s context.

Trust me my code aint worthy of being “art”.

If yours is, I totally get where you’re coming from, AI-generated code is pretty bad right now, albeit it does often work.

The way I often do things is have an agent create a rough first implementation according to my own architecture, so I have done all that high-level thinking that it currently struggles with; then I have a dedicated improvement agent come and clear that up substantially, and then I review whatever is left.

Which, yeah, isn’t great, but I stand with my intsance’s viewpoint on generative AI:

In other words, we’re not against Generative AI as a technology but we are against Generative AI as promoted by capitalism and corporate interests. To put this into perspective, it’s like saying: We’re not against using camera equipment, but we’re against the surveillance industry.

The full wiki page.

Just a lot homonculi people see AI and start screaming “AI slop” from the hilltops.

Grow up.

Of course, NASA ain’t modelling actual missions in KSP.

I imagine KSP is simple Newtonian physics which is accurate at macro-scales, whereas obviously NASA is going to use Einstein’s equivalent. Far more computationally intense but really important once you scale up the small errors in Newton’s equations.

That makes sense, floating-point precision falls apart with large numbers but is pretty good with small numbers.

Perhaps there’s a mod that can swap out the position system to use doubles instead of floats? But I’m pretty sure that’s a Unity thing with its Transform positioning system so that’d be quite a challenge to change.