haberman

53 ms·

331.
by haberman 4y ago
Yes, proto3 "singular" fields are a big problem. But now that proto3 supports "optional" fields (which remember the difference between unset and explicit 0) you can use true optional fields for all new messages going fo
332.
by haberman 4y ago
Validation is important, but the serialization layer is the wrong place to put validation logic. A protobuf is a low-level abstraction, like a struct or record type in your favorite programming language. You want validation logic to be a s
333.
by haberman 4y ago
Hi there, I am the primary maintainer of the PHP library as of the last few years. I have heard that there used to be a lot of crashes; the code was almost completely rewritten in 2020 and is in a much better state now. If you find a segf
334.
by haberman 4y ago
> Their behavior was a part of a harassment campaign against someone who is largely a private citizen in this instance. What you call a "harassment campaign" happens constantly on Twitter, driven by Twitter's own site and
335.
by haberman 4y ago
What is the actual benefit of this? The cost is that conversion to/from civil time is far more complicated, and worse, cannot be computed for future dates for which leap seconds have not yet been determined. I think that 86,400-second
336.
by haberman 4y ago
> We are also working on a special backend, one that produces C source code. Progress on the C backend has recently shot forward (87% and counting) thanks to an amazing recent contribution. I love this. It's great story for people
337.
by haberman 4y ago
Offer people food and shelter. Hopefully they will take it. If they decline because they like the streets better, there has to be some recourse. We can't cede the health and safety of public spaces because we are too afraid of havin
338.
by haberman 4y ago
I just spent a few days in Birmingham, Alabama. The downtown streets were so clean you could eat off of them. It is really hard to come back to Seattle and see all of the trash everywhere. In the Seattle airport one of the bathroom stalls
339.
by haberman 4y ago
Blind orchestra auditions are a great idea that is sadly being called into question for not being progressive enough: https://www.nytimes.com/2020/07/16/arts/music/blind-audition... https:/&#x
340.
by haberman 4y ago
I don't think you can call Zig's choice for #1 the cause of the problems in #2. C++ made a different choice for #1 and yet still suffered the problems of #2 until C++20.
341.
by haberman 4y ago
You can't use "if constexpr" to vary the type signature, for example (as in my link). Or to choose a different representation for a type based on a type parameter. Those will require metaprogramming at the template level. A
342.
by haberman 4y ago
By #1 I mean Zig's design of expressing all generic-related logic using normal control flow constructs that just happen to be evaluated at compile time. For example, suppose you want a generic function or generic type to have some spec
343.
by haberman 4y ago
I feel that this post conflates two different things: 1. The model for expressing and instantiating a generic function. 2. The model for constraining a generic function such that type errors are as local and comprehensible
344.
by haberman 4y ago
Do you mean Rutherford B Hayes? Andrew Jackson died in 1845, 16 years before the civil war began.
345.
by haberman 4y ago
It's astounding to me that you can get intelligible speech at seven 700 bps. 700 bits is less than the ASCII of this comment.
346.
by haberman 4y ago
> Change from stainless steel hardware for doors and travel touch surfaces back to "brass/copper". I have never heard of this. I had to Google it to even understand your meaning. It's eye-opening to learn that diffe
347.
by haberman 4y ago
Folks seem to be comparing xz to zstd, but if I am understanding correctly the true competitor to xz is the article author’s “lzip” format, which uses the same LZMA compression as xz but with a much better designed container format (at leas
348.
by haberman 4y ago
> Is it fair to say it's just less optimised than before? I would say it's worse than that: it has an actual bug. Yes, a program that depends on tail call optimization, despite having no guarantee of tail call optimization bein
349.
by haberman 4y ago
I wrote an article about my motivation for it: https://blog.reverberate.org/2021/04/21/musttail-efficient-i...
350.
by haberman 4y ago
Most optimizations only offer constant factor improvements. Tail call optimization is the difference between O(1) and O(N) stack size.
351.
by haberman 4y ago
You don't want it to be implementation-specific whether tail call optimization is performed or not. The nightmare scenario is that you spend months writing a program that works great in browsers A and B, but when you roll it out to pro
352.
by haberman 4y ago
I didn't bring race into it, the article did: > The staff, predominantly white, is recruited overwhelmingly from the universities of Oxford and Cambridge, and a disproportionate number of the most important editors have come from ju
353.
by haberman 4y ago
> But you at least get something of some value at the end. That makes it worse honestly. You can go hundreds of thousands of dollars into debt getting a BA in the humanities from a university whose tuition is increasing at 2x inflation,
354.
by haberman 4y ago
The Economist was founded in 1843, and the article goes to great lengths to paint the writers and editors of the magazine throughout its history as not merely wrong on matters of policy, but personally privileged and bigoted.
355.
by haberman 4y ago
Intellectual superiority and moral certainty are precisely what this article is selling.
356.
by haberman 4y ago
This article is a highbrow variety of the two minute hate, in which an ideological enemy is identified (in this case a newspaper), and 179 years of history are cherry-picked in order to present a simple, satisfying narrative that this enemy
357.
by haberman 4y ago
Can anyone offer a concrete summary of what Wolfram Language and Mathematica are, from a technical perspective? The marketing material alone makes it difficult for me to situate them in the landscape of languages. - Is "Wolfram Langua
358.
by haberman 4y ago
> You have the same restrictions in C++ except with worse consequences. In C++ it is safe because the arena is only used from one thread at a time. To model this C++ pattern in Rust, what I would really want is: 1. Arena should be Sync,
359.
by haberman 4y ago
I invested a lot of time porting some parsing code I had written to Rust, with the vision that Rust is the memory-safe future. The code I was porting from used arenas, so I tried to use arenas in Rust also. Using arenas required a bunch of
360.
by haberman 4y ago
> sort of like an Electron competitor How would this work? Does the user start and stop the server manually? One nice thing about an electron app is that you open and close it like a normal application.

More ›