lbrandy

7 ms·

31.
by lbrandy 8y ago
All of msvc, gcc, clang, and the isocpp committee have active work ongoing for C++ modules. We'll have them Soon™.
32.
How 4 plucky Englishmen held mighty AlphaZero to draw(chessable.com)
3 points by lbrandy 8y ago|0 comments
33.
by lbrandy 8y ago
"zero overhead abstractions" is a principle of the language, not the library. The STL containers are clearly designed to be on the "safer" side than the performant side. You are right that, perhaps, we've extracted
34.
by lbrandy 9y ago
"terminal level", in the sense being used by the OP, doesn't mean "end of the IC track". It means a level that you can be at "forever". Often times the junior engineering levels have some expectations arou
35.
by lbrandy 9y ago
Just so I'm clear, what this appears to be doing is the moral equivalent of a function that asks if some constant equals what we know to be its value (true at compile time)? And then at program-init time (via __attribute__((constructor
36.
by lbrandy 9y ago
> Folks said the same thing before for every industrial revolution. I'd like to see more facts than religion on this matter. I agree with you that the "jobs are running out" is an argument that requires evidence, given its
37.
by lbrandy 9y ago
> A major pain point probably is that people never really followed the "conventions" of C++ resource management as rigidly as Rust enforces them. That's exactly right. Rust does a great job moving idiomatic C++ best practi
38.
by lbrandy 10y ago
libfuzzer (and many of the fuzzers used widely today) is a coverage (as in, code-coverage) directed fuzzer. So it's not just randomly mutating the input, but "selecting" for random inputs that happen to light up new code path
39.
by lbrandy 10y ago
Do you really want my thoughts? The question isn't asked in good faith, so it doesn't deserve a good faith answer. The account was also created just to make that post. And yet it was upvoted by the community. And that speaks poorl
40.
by lbrandy 10y ago
No.
41.
by lbrandy 10y ago
All code needs to be reviewed and approved. The official technical rule is something along the lines of anyone can approve anyone else's change (to anything) so long as they both aren't interns. Socially, it ends up being more lik
42.
by lbrandy 10y ago
Well, let's see. Today I worked 8:30 to 5:30 or so. I spent some time cleaning out emails. I did two interviews (which is unusual). I reviewed several changes to central C++ libraries in the codebase. I spent awhile with some people de
43.
by lbrandy 10y ago
This is clever but it can't be a good captcha. Computers are really good at understanding a visual chessboard and even better at actually finding the best move.
44.
by lbrandy 10y ago
If you search either of these numbers on google you see a ton of errors and people asking befuddled questions. We're literally doing a public service to future versions of ourselves by juicing the google results for this post. For once
45.
by lbrandy 10y ago
> Spend 90% of the time writing performance-neutral code in a modern high-productivity language and 10% on performance-critical code in a modern high-performance language. What about the other 80% of your time spent trying to get those t
46.
by lbrandy 10y ago
First time I ever went to Pittsburgh, driving w/o a gps, we were trying to find 6th street. We were on an avenue hoping to see numbered cross streets. We ended up at the intersection of 5th avenue and 6th avenue ( https://goo
47.
by lbrandy 10y ago
> But why? Go ahead and post a link in a message (I chose this one: http://imgur.com/gallery/wjANVCD ). A little visual preview will appear. Where do you think this comes from? > And classify you as a good citizen
48.
by lbrandy 10y ago
What could be more sinister than aliasing issues?!?
49.
by lbrandy 10y ago
Our (facebook's) folly library has lots of components for doing stuff using fibers[1]. That sits with folly::futures[2] and wangle[3]. That said, without proper language/compiler support, though, it's difficult to make them &
50.
by lbrandy 10y ago
> I never quite understood why unique_ptr and move semantics... It's not those types, in particular, it's because C++11 allows you move your ownership model into the type system. And then, and this might sound familiar, writing
51.
by lbrandy 10y ago
Hi. I worked on Facebook's anti-abuse infrastructure for awhile (I'm still at Facebook, but working on different things now). So while I didn't personally fight spam/fake accounts, I worked closely with those who did. I&
52.
by lbrandy 11y ago
As my 3 year old began to recognize Minecraft, I decided to sit down and play it for the first time. The base game was ok but the modded versions... let's just say that was 3 months ago and I'm completely hooked. Like factorio, it
53.
by lbrandy 11y ago
I've been following modules pretty closely and attended the meeting in Jacksonville. I think the story here is actually pretty positive. I understand people are upset it's not making C++17 but those expectations were always unreal
54.
by lbrandy 11y ago
Pointers (smart or raw or whatever) get used when you need reference semantics. It's true that value semantics are usually better but not always (eg inheritance). To be sure, though, being able to make proper value types is a strength
55.
by lbrandy 11y ago
> arguing that people have said this bubble was about to burst and that it hasn't yet isn't an argument that it won't. The reverse isn't an argument, either. There's essentially zero information content in a clai
56.
by lbrandy 11y ago
> any example using a lambda in C++ I find to be 10,000× worse than the functional equivalent. > I naïvely assumed that the keyword `lambda' was involved. I don't understand how this two sentences co-exist. If you are un
57.
by lbrandy 11y ago
The top comment in HN shouldn't be overly cynical and negative sniping at some tangentially related comment in an otherwise interesting article. That's the kind of thing that might make people want to stop reading.
58.
by lbrandy 12y ago
It's weird how this justification for "look a bubble!" has been consistently voted to the top of HN for its entire existence. Smart people have been chatting about their being or not being a bubble longer than Hacker News has
59.
by lbrandy 12y ago
> Many of these people are now forced to sell out to big corporate record labels... Mostly look at the word "now". At what time did musicians not need to "sell out" to "big corporate" record labels?
60.
by lbrandy 12y ago
OP is correct. You've quoted the general language around move semantics but unique_ptr itself has a stronger gaurantee. From 20.7.1 Additionally, u can, upon request, transfer ownership to another unique pointer u2. Upon completi

More ›