dist1ll

21 ms·

241.
by dist1ll 4y ago
It was a lame attempt at humor, a roundabout way of referring to the simplifying assumptions that *nix systems generally make of the underlying machine.
242.
by dist1ll 4y ago
Building an OS is hard. There's no way around its complexity. But closing your eyes and pretending everything is a file is a security disaster waiting to happen (actually, happening every day). And furthermore, OS research is not onl
243.
by dist1ll 4y ago
There's this great USENIX talk by Timothy Roscoe [1], which is part of the Enzian Team at ETH Zürich. It's about the dominant unholistic approach to modern operating system design, which is reflected in the vast number of independ
244.
by dist1ll 4y ago
Fair enough, I agree that a global x32 switch seems misguided, especially since efficient data structures are already written in cache-aware ways.
245.
by dist1ll 4y ago
I'm not sure I understand your point. First you say: > In general if shrinking the pointer size is a huge optimization, your data structures have too many pointers in them. And then you go on to say > having sibling references (e
246.
by dist1ll 4y ago
Pointer compression is a pretty nifty and simple optimization. While working on a high-performance wait-free MPSC queue, pointer compression allowed me to fit the tails of each ring buffer queue together into a single cache line, reducing c
247.
by dist1ll 4y ago
If your unsafe code is sound, you can't cause memory corruptions with wrapping integer overflows, no matter what you do with them. If your Rust code is engineered correctly, overflows are primarily logic bugs, not memory safety issues.
248.
by dist1ll 4y ago
Holy god what a burn! I absolutely agree though.
249.
by dist1ll 4y ago
As long as the game is zero-sum or even negative-sum, you won't be able to make money. But , when you're placing tote bets, you're often competing against a variety of different actors in a prediction market - and not all of
250.
Avoiding benchmarking pitfalls with black_box in Rust(alic.dev)
1 points by dist1ll 4y ago|0 comments
251.
by dist1ll 4y ago
That's a culture problem. It boils down to documentation being a second-class citizen which is not included in the "definition of done".
252.
by dist1ll 4y ago
Just because something is a social construct doesn't mean it's worthless, impure or required to be rejected. Going from self-reflection to nihilism is a pretty big overreaction.
253.
An Unbelievable Demo(brendangregg.com)
3 points by dist1ll 4y ago|1 comments
254.
by dist1ll 4y ago
That depends on what domain you're interested in. Web? Embedded? Graphics? OS dev? Overall, I generally recommend conference talks by experienced engineers; Strange Loop[1], USENIX[2] and CppCon[3] to list a few. [1] https:/&#x2F
255.
by dist1ll 4y ago
Go's interfaces are an example of structural subtyping , not duck typing. The difference is that the structure a function expects is defined at compile time (i.e. in the function's signature) and not dependent of runtime usage of
256.
by dist1ll 4y ago
I thought a bit more about it. I stand by my statement that federation is not decentralization. Federation is a pre-requisite for decentralization, but not a sufficient criterion. (I said weaker form, what I meant to say was necessary condi
257.
by dist1ll 4y ago
Federation is a very weak form of decentralization. In practice, many federated platforms turn into cartels of giant centralized providers. It's better than Twitter, but not something I'd call decentralization.
258.
by dist1ll 4y ago
Mastodon is not decentralized, it's federated. So no, it wouldn't fit any common definition of web3.
259.
by dist1ll 4y ago
If I had to pick the two worst sources of bugs I can imagine, it would be (1) concurrency and (2) caches. Those functions often require extensive performance and load testing and have tons of non-deterministic failure modes.
260.
by dist1ll 4y ago
> C is fast because it's close to how CPU and memory actually work. Out-of-order execution, cache hierarchies, branch prediction, virtual memory, pipelining, vector instructions, ILP, NUMA are all pretty transparent to the C spec. T
261.
by dist1ll 4y ago
> but it puts a lot more pressure on the GC than simple local stack variables. Do you have evidence for this claim? AFAIK the Go compiler does escape analysis, and allocates pointers that don't escape on the stack.
262.
by dist1ll 4y ago
I feel conflicted. The FOSS label is an extremely powerful statement and draws in lots of users. To switch the license is very poor taste in my opinion, and not much different from a bait-and-switch. It should be a maintainers responsibilit
263.
by dist1ll 4y ago
That's referred to as "source-available" to distinguish it from free/libre software.
264.
by dist1ll 4y ago
Does anyone else find the typography on this website remarkably beautiful?
265.
by dist1ll 4y ago
(1) Knowing how to use crypto safely is hardly "expertise". It's something that can be learned quite quickly from experts. (2) Having a general idea of many areas in life is important for a functioning society and becoming a
266.
by dist1ll 4y ago
Great, we don't need speculation in crypto. We need people taking a chance on the technology, and empower democratization of finance.
267.
by dist1ll 4y ago
Do you think it's a good thing that we as a society are collectively extremely ignorant about how money works? Don't you think that maybe people should start becoming more educated about how our economy runs? Or what protects thei
268.
by dist1ll 4y ago
Yeah, but the conversion happens very quickly. Deposit, trade, withdraw. Unfortunately, people leave tons of money on exchanges, for long periods of time. That's the problem. That's what the majority of exchange users are doing.
269.
by dist1ll 4y ago
That's what people don't get. FTX' spectacular failure is literally an advertisement for decentralization.
270.
by dist1ll 4y ago
You forgot operations, tooling, observability, SRE, incidence response, etc

More ›