And if you look at other languages, you a whole lot of security problems that have completely different causes that have nothing at all to to with memory safety. In fact, OWASP Top 10 does not even have an item for Buffer Overflow anymore.
In addition, it is pretty much as easy to make web-applications effectively as insecure in a memory-safe language. Just screw up permission checking, for example. Otherwise we would not see the flood of vulnerabilities in things written not in C or C++.
Indeed. And it is not like applications written in memory-safe languages are more secure in reality. The problem is incompetent developers, plain and simple. In a memory-safe language, they just screw up something else with pretty much the same result.
Energy generation is not the point at this time. The point is creating and maintaining the plasma and 10 seconds is pretty impressive at this stage for Tokamak.
Well, I do agree that deanonymizing "anonymized" data is routinely very easy, especially when you only want 95% or so in accuracy.
However, I do not get your point. Are you saying this latest development changes nothing and they were directly lying before and are just maybe a bit more honest now?
Pretty much what more perceptive people have predicted is happening. Also, anybody working at Google should think very hard about what it means to be complicit and whether that is something they want to be.
Not in an age were every preliminary grande claim gets funding, while solid, slow and often boring, but long-term hugely profitable research gets overlooked...
This is really stupid. Google is changing and endangering working internet standards, just so they can safe a few bucks on connectivity. This should be resisted decisively.
I think the problem here is both that Google has stopped caring about anything than themselves (if they ever did) and that they actually lack experience. They may just come with yet another new protocol in a few years, because this one did not do what they want after all. This is not good at all.
Slashdot Mirror
And if you look at other languages, you a whole lot of security problems that have completely different causes that have nothing at all to to with memory safety. In fact, OWASP Top 10 does not even have an item for Buffer Overflow anymore.
Ah, the Rust fuckups are at it again. That explains it. They do not understand what makes code insecure.
In addition, it is pretty much as easy to make web-applications effectively as insecure in a memory-safe language. Just screw up permission checking, for example. Otherwise we would not see the flood of vulnerabilities in things written not in C or C++.
The whole article addresses the wrong problem.
Indeed. And it is not like applications written in memory-safe languages are more secure in reality. The problem is incompetent developers, plain and simple. In a memory-safe language, they just screw up something else with pretty much the same result.
The problem is not C or C++. The problem is incompetent developers. They manage just fine to make things in other languages just as insecure.
Energy generation is not the point at this time. The point is creating and maintaining the plasma and 10 seconds is pretty impressive at this stage for Tokamak.
Well, I do agree that deanonymizing "anonymized" data is routinely very easy, especially when you only want 95% or so in accuracy.
However, I do not get your point. Are you saying this latest development changes nothing and they were directly lying before and are just maybe a bit more honest now?
Such beliefs come easy when you are arrogant and ignorant enough.
Naaa, just another false God. We already have more than enough of them.
And if you believe that ... never mind.
Pretty much what more perceptive people have predicted is happening. Also, anybody working at Google should think very hard about what it means to be complicit and whether that is something they want to be.
No backup - no pity. Seriously, not having a backup _is_ asking for it. No, this is not victim-blaming, this is pointing out extreme stupidity.
Running without backup is gross negligence by any sane standard. This person should get nothing.
I do not share your perverted idea of the Star Wars universe!
Not in an age were every preliminary grande claim gets funding, while solid, slow and often boring, but long-term hugely profitable research gets overlooked...
Aaaaaaand fail. There are tasks that are inherently single-thread and these are not exotic ones.
Nobody sane has real workloads on Windows servers...
Well, you will get fu**** by Intel. No idea whether that counts for you.
If they did, they could not fake benchmarks like this.
As soon as you look at prices and availability, Intel is utterly naked.
Also, UPnP is just another extreme fuckup by Microsoft.
Excessive use of UDP breaks TCP. Has been known for 20 years or so...
You know, for changing IP addresses, I just use mosh and tunnel over it. I think Google is solving the wrong problem here.
Seeing that requires experience. I begin to think Google engineering is lacking that.
This is really stupid. Google is changing and endangering working internet standards, just so they can safe a few bucks on connectivity. This should be resisted decisively.
I think the problem here is both that Google has stopped caring about anything than themselves (if they ever did) and that they actually lack experience. They may just come with yet another new protocol in a few years, because this one did not do what they want after all. This is not good at all.