Retr0id

10 ms·

241.
by Retr0id 2mo ago
If everyone's giving everyone the same gift, it's not really a "gift" any more, more of a p2p buy-now-pay-later arrangement (which isn't a terrible idea, but it's suboptimal for those with later bithdays).
242.
by Retr0id 2mo ago
Also, as more batteries get added to the grid, the difference between the "cheap" and "expensive" energy prices ought to reduce.
243.
by Retr0id 2mo ago
> Create a battery birthday club. Buy each friend a battery on their birthday until everyone has one. At that point, wouldn't it be easier if everyone just bought their own? And then whoever has the latest birthday doesn't have
244.
by Retr0id 2mo ago
Interestingly this dashboard says there are currently no issues. Edit: I just realised it's hosted on gh pages and therefore probably can't be updated
245.
by Retr0id 2mo ago
This was a coursework problem in my CS course, back in the day. For larger canvases, the state-space blows up and it gets slow/intractable to solve.
246.
by Retr0id 2mo ago
I think the most charitable theory is that they're seeing rapidly growing usage (more users churning out higher volumes of code, via LLMs), and are struggling to scale to meet demand. Alternative theories include "azure sucks&quot
247.
by Retr0id 2mo ago
What a strange take. LLMs themselves are a perversion of writing. I couldn't care less about the implementation details of the PRNG they use for next-token sampling (well, as long as they're not stuffing a user ID in there).
248.
by Retr0id 2mo ago
I actually don't think ZKPs solve the age verification problem (for a variety of reasons that won't fit here), but they solve plenty of more interesting problems. Private transactions in Zcash, for example.
249.
by Retr0id 2mo ago
A great way to announce that you don't understand ZKP
250.
by Retr0id 2mo ago
It fits the "naive deplagiarization" explanation perfectly, however. Many of them are pre-LLM, when people had to plagiarize things the good old-fashioned way, and subsequently hide that fact via synonym replacement.
251.
by Retr0id 2mo ago
It looks very stealable, which is amusing given "boost" is slang for "steal".
252.
by Retr0id 2mo ago
Right, that's the problem I'm describing.
253.
by Retr0id 2mo ago
They're ok. But when I use a search engine it's because I want search results. If I wanted an LLM's interpretation I'd have just asked a search-enabled LLM in the first place.
254.
by Retr0id 2mo ago
I think it'd be more informative to see total users per provider, rather than % shares.
255.
by Retr0id 2mo ago
I find the amount of things interesting to me is about the same, or reduced.
256.
by Retr0id 2mo ago
I wrote an RV64IMA emulator recently. I just needed a virtual CPU core that could boot linux, and RV64IMA seemed like the simplest way to do that - and I think that's more or less true. But then I wanted to be compatible with off-the-s
257.
by Retr0id 2mo ago
Bootstrap, what a blast from the past! They just don't make bad websites like they used to.
258.
by Retr0id 2mo ago
I suspect airports could be a whole lot simpler if the notions of "security" and "border control" didn't exist or were otherwise relaxed.
259.
by Retr0id 2mo ago
The simpler angle is that more text lets them bill you more. I don't think that was necessarily their intent, but it does mean they have a negative incentive to fix it.
260.
by Retr0id 2mo ago
> A leaked token enables connections, but not impersonations. The token is addressed to one specific endpoint's public key So now the problem is moved to "how do you decide who to issue tokens to?", which every project mus
261.
by Retr0id 2mo ago
I've tried telling it to "fix" comments with varying degrees of specificity and in my experience it just... fundamentally doesn't get it. Presumably using a different model for it would help. My theory is that Claude&#x2
262.
by Retr0id 2mo ago
It's a surprising change from my perspective, because in the past it felt like they understood that Claude should be pleasant to interact with.
263.
by Retr0id 2mo ago
I'm not sure how much the harness affects things, but the Deepseek web chat keeps trying to talk to me in Chinese. I tell it to use English, and it "forgets" a few turns later. I wonder if I'd get better results if I cou
264.
by Retr0id 2mo ago
I assume it's deliberate - you're not supposed to know what it's doing. It's a black box that either completes the task or spins forever trying.
265.
by Retr0id 2mo ago
This is also my opinion. However, I still think it's worth raising the bar for plausible fakes. Minimising processing is another way to raise that bar (or at least, lower the floor). It's also not something you can do by default,
266.
by Retr0id 2mo ago
It's also something I've been looking into. I've completely broken Google's implementation and I can sign any file as if it were real, more details will be public in the coming weeks. Signatures are nice, but the content
267.
by Retr0id 2mo ago
Apple is apparently working on the latter, although it'll be opt-in and it's unclear whether it will also reduce the processing (I hope it does!): https://www.macrumors.com/2026/08/10/ios-27-apple-re
268.
by Retr0id 2mo ago
Often the "RAW" output is still DSP'd, just less (e.g. only debayering + noise reduction).
269.
by Retr0id 2mo ago
Yes, but that doesn't mean defense-in-depth isn't worth doing. The article discusses how known gadgets were removed in the past.
270.
by Retr0id 2mo ago
Right, that is in addition to mere encryption.

More ›