ardel95

While you are right, we are where we are and there is tremendous leverage in making an extremely common (even if undesirable) thing more efficient. It kind of doesn't matter why millions of lines of code perform memcpys - they do, and we're unlikely to change majority of them anytime soon.

P.S. I work on the type of systems where we obsess over unneeded memory accesses and copies. Minimizing those is not as trivial as it may seem, especially if you care about memory safety.

So New Zealand is about to prop world’s first an underground cigarette market.

So the whole 5 year process with thousands of human years of effort boiled down to: We’ll use some cloud.

Fascinating.

Oh, you are right. For some reason I thought I saw unsigned in there.

Tweet said unsigned ints, so don’t think this would work. Even with signed I think this can overflow.

The tweet said unsigned ints

Eh, that’s what I get for writing code on a cell phone. Fixed.

The correct way most people would write this for positive integers is:

if (a < b)

  return a + (b - a) / 2;

else

  return b + (a - b) / 2;

This method is just more efficient (for places where it matters) as it avoids divisions and branches. But for a vast, vast majority of use-case that tiny efficiency gain doesn’t really matter.

I don’t know about others, but our Alexas became really hard to use lately. Basic commands like “Alexa, turn on the lights” or “Alexa, play Frank Sinatra” get misunderstood and need to be repeated multiple times. Combined with unsolicited “by the ways”, we are generally unlikely to ever buy another one.

I think smart assistants could be a useful gadget eventually, but alas this isn’t it.

This line of reasoning assumes that all physical processes are deterministic. If they are not (as is the current mainstream belief) then I don’t think that line of reasoning would stand. Instead, the current state is the result of that initial set of inputs, and countless other nob-deterministic dice rolls.

50% savings to replace a managed service with a self-managed instance (trust me, even if things are fully automated there will still be gray failures, weird hiccups, congestion events, etc) doesn’t sound like a great idea for most companies.

I think the left side of the Drake equation is pretty challenging too. If we assume that faster than light travel is impossible (which is likely) and all galactic civilizations are subject to the tyranny of the rocket equation, then it’s quite likely that there are no alien spaceships wizzing around, even if our galaxy is full of intelligent life forms.

And then our best way of detecting advanced alien life forms is through direct observations of remote star systems, a field that is in its infancy at best.

Live migrating a process from one host to another sounds fun on paper, until you run into all the caveats. And it gets even more fun when you try to do it on a cloud provider, where your process state is coupled to the instance on which it runs (such as security group or instance metadata state).

If there is a housing crash coming, the builders around me tend to think otherwise. Lots of new construction starts.

I think as it often happens, the most likely outcome is that lower priced areas will see the brunt of the value decrease, while higher priced neighborhoods may at best see a moderate speed bump.

That sounds plausible. But also, most software (browsers, web service SDKs, RPC frameworks) treat TCP connections as fallible by setting read/write timeouts and aggressively reopening broken connections. So, I’m totally not surprised this issue went unnoticed for this many years.

I'm glad folks are starting to look at memory efficiency of Java programs - there's just so much improvements to be had there. For a fun exercise, take a look at memory footprint of storing a million entries in a HashMap<Integer, Integer>. Between "object header" overhead, "everything is a pointer to a heap allocated object" overhead, "GC needs enough headroom to remain effective" overhead, the actual memory consumption can be as much as 10x of a comparable C, C++, or Rust implementation.

That’s the thing with fraud. The goal is never to stop it all, but to keep it at some local minimum. The rest is just the cost of doing business.

I’m afraid this might be too little, too late. But regardless whether I’m right or wrong, we all should welcome a widely used language trying to become a little more safe.

They do claim to be the safest cars!

Rust is significantly better than C++. And I say that as somebody who has done many years of C++ development.

I think they can, but they may need to evolve away from a goods marketplace and more towards entertainment centers. Most of the times we find ourselves going to a mall it’s to do something (eat out, watch a movie, engage in some kids activity) rather than to buy something.

Not an iOS developer, but curious what’s the current state of the art for distributing and storing your credentials if the app needs to make AWS service calls?

Or is that considered an anti pattern, and you want to proxy all AWS calls through your own authenticated backend?

I wasn’t familiar with Kiwifarms, so I read up about it - and oh, boy. The fact that something like Kiwifarms can exist in the United States without legal or civil liabilities is really problematic.

May I introduce you to a long list of Linux breakout and privilege escalation CVEs.

It's way past time we accepted that all current operating systems are single user environments. Any access to the system means full access to the system.

DPDK is common within network appliances, where you don't need most of the features within kernel networking stack.

This is wishful thinking, but I really hope HDMI protocol was replaced by something more reliable one day. I’ve never had this many problems with a simple task of having device A connect to device B over a short physical cable.

Having to flash firmware on a TV just so that it can successfully connect to a DVR is peak techno dystopia.

Because humans are too complex to fit into schemas.