Forgot your password?
typodupeerror

Comment Re: Reviewing code is more effort than writing cod (Score 1) 78

I've tried LLM agents for coding. I find they they do an OK job at getting a small project off the ground. Afterwards they're useless for me. However, it really depends if you're writing code that's essential a copy & paste of common code on the internet or if you're writing something new. What I haven't tried is using agents only to unit test my code. That could possibly work out ok.

Comment Reviewing code is more effort than writing code (Score 5, Interesting) 78

I've been writing software for 30 years. Honest code reviews take more mental effort than writing the code yourself, unless the changes are small and clearly and verifiably well tested. Proper design for unit testing is hard and beyond the capabilities of AI. Hence, you can't really do better than a human software engineer, yet.

Comment Reddit's other bot problem (Score 2) 116

Their other bot problem is that they frequently flag for TOS violations when you're saying the opposite of a violation. Like, if you say "I object to the notion that women shouldn't have equal rights" the bot will flag you for the phrase "women shouldn't have equal rights." So the only safe thing is to not even touch topics that relate to their TOS.

Comment I don't care anymore (Score 1) 152

Nothing I do compares to idiot politicians starting wars and burning millions of gallons of oil, gas and tons of ordinance. I'll pick up trash in my neighborhood and make sure my car doesn't ruin my local air quality. Otherwise, nothing I do matters for global warming, even voting as the last blue guy blew up a gas pipeline too.

Comment One month ago (Score 1) 83

One month ago Amazon announced huge AI related layoffs and reduction of red tape for decision making. Anyone could have predicted this outcome. There a new phenomenon, AI induced psychosis, which I think is much more widespread than people realize. It's almost unavoidable if you use AI frequently and will quietly creep into your thinking in whatever domain you use LLMs for.

Comment Theoretical result? (Score 1) 34

I looked through the paper. It seems like they haven't run this on a real quantum computer.

The preparation of the state | by the referee is computationally more expensive as it requires the implementation of pseudorandom permutations. As a near-term toy example, one could implement permutations using the Simplified Advanced Encryption Standard (S-AES) protocol. S-AES uses a block size of 16 bits (as opposed to 128 bits for AES) and a key size of 16 bits as well (128, 192 or 256 for AES). A system size of 16 can display features of quantum advantage: from Theorem 1 with, e.g., =216, =215 and =1/6, any classical algorithm solving complement sampling requires 214 =16384 distinct samples. A quantum circuit implementation of S-AES based on [45] requires only 48 qubits, 168 Toffoli gates, 364 controlled not gates, and 75 not gates.

Why didn't they take the next step and run their algorithm?

Slashdot Top Deals

"Virtual" means never knowing where your next byte is coming from.

Working...