zzzcpan

39 ms·

331.
by zzzcpan 7y ago
I don't think the conventional wisdom of building on top of filesystems exists. In distributed systems you always naturally gravitate towards using raw storage devices instead of filesystems, it becomes obvious very early on that files
332.
by zzzcpan 7y ago
Network is under user control, pretty much all devices require active cooperation to connect device's internal network to any external network, it doesn't happen on its own.
333.
by zzzcpan 7y ago
Connecting to a centralized DNS provider over tor defeats the purpose of tor and makes it easy to deanonymize your traffic completely.
334.
by zzzcpan 7y ago
It's possible a single entity was behind massive world wide promotion campaign. But unlikely. More likely many players who had to gain something on speculation all contributed to the campaign on their own.
335.
by zzzcpan 7y ago
The author pushes one sided Cloudflare's line on both client subnets and DoH (given APNIC maybe he should disclose affiliation). There is nothing accurate about it. Client subnet extension doesn't violate any privacy in any threat
336.
by zzzcpan 7y ago
> A wealth tax is a tax on capital, period - and making capital more expensive hurts everyone, but it hugely hurts those who need capital the most. This is some weird ideological bullshit. There is plenty of evidence that capital hurt
337.
by zzzcpan 7y ago
Terminals don't render colors and fonts the same way, displays don't display them the same way, people don't see them the same way. You are also not going to colorize output properly keeping consistent perceived brightness an
338.
by zzzcpan 7y ago
> The title is a bit confusing, actually means things they attribute Go's success to. As opposed to what? I can't see another possible interpretation of the title.
339.
by zzzcpan 7y ago
I'm assuming by success you mean some adoption, like top 20 languages maybe? Marketing is, of course, necessary for that, but far from the whole story. Because we also had Perl, PHP, Python and Ruby that don't fit your explanation
340.
by zzzcpan 7y ago
If you know a bit of background on how these things work, it is also an obvious intelligence agency level murder, not just someone hiring someone to kill him.
341.
by zzzcpan 7y ago
> Surely someone who can do all of this stuff would be considered skilful. Or merely someone who guards report to.
342.
by zzzcpan 7y ago
> I'm fascinated by the idea that someone could have killed and successfully covered up a murder this high profile. What reasons could there be to cover up non high profile murders? High profile is where all the covering up happen
343.
by zzzcpan 7y ago
> And if the program can't discard messages, then it's a DoS. If the program can instead rely on the receiver to keep up so it doesn't need to make this choice, then there's a trust assumption between these processes
344.
by zzzcpan 7y ago
It's not about selling the data for ISPs, never was, never will be. For ISPs almost all websites are uniquely identifiable by looking at clear text SNI and Host: headers. But it's much worse, 95% of top 1 million websites are uniq
345.
by zzzcpan 7y ago
There is no assumption of mutual trust between processes, and there is nothing synchronous about it, you still don't know whether any of the messages reached their destinations. This is just backpressure, asynchronously propagated (or
346.
by zzzcpan 7y ago
You are over simplifying it and dropping too many important details so that it ceases to be a real system and becomes some weird system susceptible to DoS. In high performance scenario it would be more like this: process shares fixed ring-l
347.
by zzzcpan 7y ago
> Sure, if you want to introduce inherent DoS vulnerabilities into your IPC subsystem, not to mention slow down IPC so much that it's practically unusable. This is nonsense. There could be DoS vulnerabilities in implementations,
348.
by zzzcpan 7y ago
> And keep in mind that both WASM and JS now have spectre-defenses, so there is no need for a PL from scratch for this. As far as I remember they weren't able to defend from side channel attacks within the same process completely
349.
by zzzcpan 7y ago
> But I DO think it's difficult to design an IPC mechanism that everyone is happy with. Some people want only synchronous IPC ... Every form of IPC can be implemented on top of asynchronous message passing. Interface is not the pr
350.
by zzzcpan 7y ago
With the advent of multi core CPUs monolithic kernels jumping in on doing slow shared memory concurrency kind of killed the idea of monolithic kernels to be fast. They can't be fast unless they do actor model, but it goes against the i
351.
by zzzcpan 7y ago
Trusted code has to be free from vulnerabilities to be immune, so it's still an issue even for trusted code. And I'm pretty sure neither webassembly nor other sandboxing methods can fully mitigate speculative attacks on out-of-ord
352.
by zzzcpan 7y ago
> Again, the choice of algorithm is one thing, but what adds to the accidental complexity is the requirement that that choice be reflected in the type signature, and so it affects the consumer of the code as well. As long as you have s
353.
by zzzcpan 7y ago
These are not aesthetic preferences, these are essential parts of concurrency models. And they cannot be decoupled from algorithms.
354.
by zzzcpan 7y ago
With "async/await" you see exact places where you give up control, it's part of the model, without it you don't and it's a different model. First one lets you program as if you can always access shared memory l
355.
by zzzcpan 7y ago
Type declaration is supposed to signify the choice of concurrency model. Either way algorithmically the choice has to be explicit, even it it's not a type declaration, it still has to be an explicit declaration somewhere or there is no
356.
by zzzcpan 7y ago
It's a convention that cannot really be enforced by compiler or any tooling, it ultimately relies on humans following it, hence it will always be broken in some way for some users.
357.
by zzzcpan 7y ago
If your concurrency model only allows bounded nondeterminism, you can't express an algorithm that assumes a concurrency model with unbounded nondeterminism in it. Furthermore, the choices concurrency models make hugely affect performan
358.
by zzzcpan 7y ago
"async" is required from the algorithmic point of view, so the algorithm can be expressed in terms of specific concurrency model. And in Rust, it's not just "async" that's required, but also a specific executor
359.
by zzzcpan 7y ago
> Now I am seeing the same craziness with “a sync” and continuations and wonder if I will be lucky enough to outlive it to. Maybe you will. It's an attempt to address shortcomings of shared memory, just like borrow checker. Once lan
360.
by zzzcpan 7y ago
Isn't the whole point of CPUs is that you don't have to run code on GPUs and other special purpose processing units? You leave it to special circumstances when you actually can't avoid doing it.

More ›