ot

13 ms·

121.
Mike Krieger: I've joined AnthropicAI as their Chief Product Officer(threads.net)
40 points by ot 2y ago|11 comments
122.
by ot 2y ago
My overall point is that neither your function or my function are actually O(1). Whenever you see the notation O(...), there is an implicit context "As input size n grows arbitrarily, ...". You can check the formal definition on W
123.
by ot 2y ago
> This function is O(1) I think I addressed that in my comment, but to be more explicit, this function is O(1) too: size_t find_sorted(void* object, void** list, size_t size) { for (size_t i = 0; i < SIZE_MAX; ++i) { if
124.
by ot 2y ago
Radix trees are not O(1), they're O(log n). Data structures that support constant-time predecessor lookup do not exist, there is a super-constant lower bound, even in the RAM model [1]. Often I hear people say "well pointer size i
125.
Formal methods: Just good engineering practice?(brooker.co.za)
2 points by ot 2y ago|0 comments
126.
webgraph-rs: a Rust port of the WebGraph framework(github.com)
4 points by ot 2y ago|0 comments
127.
by ot 2y ago
In general the compiler/linker cannot assume that derived classes won't arrive later through a shared object. You can tell it "I won't do that" though with additional flags, like Clang's -fwhole-program-vtables
128.
by ot 2y ago
Or "something is wrong with my benchmark setup", which is also a possibility :) Without a comparison of generated code, it could be anything.
129.
Looking for AI Use-Cases(ben-evans.com)
3 points by ot 2y ago|0 comments
130.
New Assistant: Meta AI, Built with Llama 3(about.fb.com)
42 points by ot 2y ago|6 comments
131.
Async C++/Rust Interoperability(medium.com)
1 points by ot 3y ago|0 comments
132.
Ray Cat(en.wikipedia.org)
2 points by ot 3y ago|0 comments
133.
by ot 3y ago
Surprised that neither the article or this thread link to https://xkcd.com/570/
134.
by ot 3y ago
It is worth noting that this does not come for free, and it would have been nice for the article to mention the trade-off: reconstruction is not cheap on CPU, if you use something like Reed-Solomon. Usually the codes used for erasure coding
135.
by ot 3y ago
But in the case of HN the RNG was seeded by millisecond-granularity timestamps. This was universally considered bad already in 2009 (it would have been even in 1999). The Debian SSL key scandal happened in 2008. It is quite a leap to say &q
136.
by ot 3y ago
I wish articles like this described the technique for what it is, that is a fun gimmick, or maybe even an art installation, rather than misleading the reader into thinking that this is produces better randomness than any other common entrop
137.
Meta previews 'Fediverse sharing' for Threads(engadget.com)
2 points by ot 3y ago|0 comments
138.
Norfolk County Council beats Apple in £385M iPhone row(bbc.com)
10 points by ot 3y ago|1 comments
139.
How a cheap, generic drug became a darling of longevity enthusiasts(washingtonpost.com)
4 points by ot 3y ago|1 comments
140.
by ot 3y ago
Did you read the article? It is specifically about how the series looks simple, but the error is actually very bad if you do things naively.
141.
by ot 3y ago
`numThreads` is written before the child threads that read it are started, so there is an explicit happens-before relationship and no data race. Before `numThreads` is reset, the child thread are joined. There is no bug in the program, it i
142.
by ot 3y ago
> Just for example, if you wanted to dynamically extract a field at query-time based on the difference of two values and group by that, that's something you can do trivially in splunk/sumo. You can trivially do that in Scuba to
143.
by ot 3y ago
I wouldn’t count that as a transparent “display” since it needs a light source when detached, or it wouldn’t be of much use. So if we only consider self-contained transparent displays, this is indeed a first?
144.
by ot 3y ago
"Big Mac's a Big Mac, but they call it Le Big Mac"
145.
Gemma.cpp: lightweight, standalone C++ inference engine for Gemma models(github.com)
3 points by ot 3y ago|0 comments
146.
Every Linux bug fix will get a CVE regardless of security impact(twitter.com)
2 points by ot 3y ago|0 comments
147.
by ot 3y ago
I don't know how you did that math, but you cannot convert kcal to mAh, you need a unit of energy like watt hour 2000 kcal * 4184 (J/kcal) / 3600 (seconds) ~= 2324 Wh which is... a lot of batteries (a large laptop battery is
148.
by ot 3y ago
You can do that, Java did recently. You need full control of all the blocking primitives so they can yield the thread. Where this breaks is with FFI: if you cannot intercept blocking calls in foreign functions, you block useful threads or e
149.
by ot 3y ago
Comparing debt to GDP is very much mainstream economics, not MMT. https://en.wikipedia.org/wiki/Debt-to-GDP_ratio > is usually unable take their income streams It takes a percentage of that, through taxes. https:&#
150.
by ot 3y ago
What novel ideas did you see here? To me this looks like a standard 16-way compressed trie. The node encoding is quite natural if you consider the limitation of 64M entries (which is really not a lot). Did I miss something?

More ›