beefield

11 ms·

61.
by beefield 3y ago
In P/NP case I would need to assume that I can solve a problem in polynomial time, and then find a contradiction. To me it is really hard to see a useful path forward from there.
62.
by beefield 3y ago
Well, I guess you can prove that there are no odd numbers in set (2,4,6), so some negatives you can prove, but in general case, I do not know how to prove that something does not exist and will not exist ever.
63.
by beefield 3y ago
> Despite having done my undergrad in CS, I never understood what NP-hard really meant Not having done CS undergrad, I have never really understood even P/NP. The naive explanation of problems that are easy to solve and check vs pro
64.
by beefield 3y ago
> Why do you assume malicious intent? Not GP, but to me, admittably a complete non-expert on search, there are so many low-hanging fruits if search result quality was anywhere on Google's radar that it is really difficult not to ass
65.
by beefield 3y ago
That is a nice theory many (especially google) would like me to believe, but if that is the case, why majority of my google first page results can be classified as SEO spam?
66.
by beefield 3y ago
> LLMs will not achieve superintelligence: Although these models can process information quickly and access a wide range of knowledge, they lack the superior reasoning or inference abilities that would classify them as superintelligence.
67.
by beefield 3y ago
I have 64gb on my 5 year old thinkpad. What kind of performance (tokens per sec) I could expect on that nowadays for a 70B model?
68.
by beefield 3y ago
> Sorry, what? I did not say the doctor would need to trust the answer. Only that they should be required to ask.
69.
by beefield 3y ago
Having spent some time with a hallucinating ChatGPT and having spent some time with doctors during my life, at this point, in my humble opinion, it should be made illegal for a doctor to make a diagnosis without consulting an LLM finetuned
70.
by beefield 3y ago
You might want to check almost anyone else searching something on google and see if they get past the first few paid links/ads.
71.
by beefield 3y ago
1. Did you calculate in your per capita share of commercial/industrial use? 2. Feel free to provide more accurate numbers with a source.
72.
by beefield 3y ago
> It's almost silly how simple the real solutions are. Simple? Real ? Let me give you some numbers. Let's use electricity as a proxy for consumption[1]. Average global per capita electricity consumption is currently about 3000
73.
by beefield 3y ago
Knowing nothing about UUID v4 generation, I have likely a stupid question. What makes you so confident that all implementations and their entropy sources are flawless enough to make actual collision probability close enough to theory?
74.
by beefield 3y ago
Out of curiosity, have you tried "extreme" sports like skydiving or downhill biking? I have a theory that people into extreme sports have often muted emotions compared to ordinary folks.
75.
by beefield 3y ago
Some time ago I was listening a guy talking about (operational) risk management in financial industry. One of his main points was that the systems in financial organizations should not be completely bug-free and automated. Because when so
76.
by beefield 3y ago
I have long thought there should be a virtual sandbox available for apps that generates fake data to apps according to preset stories. Like, tell one app I am a rich bachelor spending time in french riviera and another I am homeless in New
77.
by beefield 3y ago
Are those typically blocked by adblockers?
78.
by beefield 3y ago
So, a dumb question. What's the easiest way to run privacy friendly analytics on static github pages? "Privacy friendly" as in unambiguously no need for cookie/gdpr permission popups. "Analytics" can be as simp
79.
by beefield 3y ago
I kind of understand many of the books in the front page, but is there a message for example in harry potter or game of thrones that I do not get?
80.
by beefield 3y ago
Hiding the option to allow "human creativity" behind two layers of menu hierarchy would mean 99% of users would never allow it.
81.
by beefield 3y ago
It's also time to regulate ebikes to have standard and open interfaces to batteries and motors (you are welcome to have multiple standards and even create new ones if technology evolves). They are ridiculously expensive to replace for
82.
by beefield 3y ago
The one mentioned in one of the earlier comments of this thread?
83.
by beefield 3y ago
I have occasionally thought that someone should make a "death app". An app that would have all kind of stuff you should do because you die one day. Not just a check list (which would be a good start), but you actually would be abl
84.
by beefield 3y ago
As a beginner book, no there are no substantial changes on what a beginner should read. However, while the simple discounting formulas described (likely, haven't read other than the list of contents) in the book were at the time actua
85.
by beefield 3y ago
> I really get the feeling you didn't think it through at all.... I said the structure can be discussed. There is no problem giving first number of years free to see if the IP is commercially viable. The aggressive progression (doub
86.
by beefield 3y ago
> If the owners/companies no longer exist, then I would think the IP should be free by default. Perhaps that's another law to work on... I think I have even more elegant solution. All IP should be subject to a tax (or maintenan
87.
by beefield 3y ago
I don't know about ability but it is obvious they are not willing to even try. Why? If search result quality was anywhere in their radar, there would be a button next to all search results letting me block the domain from my personal
88.
Ask HN: Useful ChatGPT Plugins?
2 points by beefield 3y ago|0 comments
89.
by beefield 3y ago
I'd be happy to have autonomous driving in just a few roads, namely highways. I really do not understand why there is practically zero interest in getting me out of the wheel where it would be both the easiest and the most useful. I&#x
90.
by beefield 3y ago
My pet peeve with bathroom scales is that it is impossible to find one that actually does what it claims and has accuracy of +-0.1kg. They do show the measurement to 0.1, but just shifting your weight differently can cause a difference of h

More ›