Not gonna happen until all the private debt we've accumulated in the last 50 years is gone.

Make sure you aren't running out of RAM for some reason. For example, I've seen the "Server" windows service leaking memory like a sieve in Windows 8. I've seen the BIOS limit available memory to 4GB, even though there was 16GB installed. Combined, these issues caused the machine I was using to practically grind to a halt.

Everyone works with their files locally, changes are synced via a common server. Everyone has a compressed backup of the complete history of the entire filesystem for disaster recovery. Everyone should be able to browse and recover any version of any file without adding load to the server, though usability might be slightly lacking. You could also setup a FUSE filesystem on a linux box to browse the history.

You may need to partition the file storage into multiple repositories, so that people don't need to synchronise folders that they don't use.

This sounds like a symptom of how android deals with clocks and timeouts in native code. If you call poll with a timeout, and the CPU suspends, the poll timeout stops decreasing.

You can hide the videos with a simple adblock filter; "slashdot.org##article#firehose-000".

You could also make the share button go away (slashdot.org##div.popularity), but that does break the tags css.

And in C not just the result, but the behaviour is undefined. If you divide by zero the compiler, runtime libraries and the CPU can do whatever they like. They could ignore you, crash, format your hard drive or kill your pet.

For speed reasons, this is a good thing. If it looks like you might run into undefined behaviour, the compiler can assume that the inputs to the program won't trigger that behaviour. This allows all kinds of optimisations to be performed, from dead code elimination, to hoisting invariant code out of loops.

At least other high level languages define precisely what a divide by zero should do. That way you run into platform or compiler specific heisenbugs far less frequently.

From the SSD Endurance Experiment;

To me, that case layout looks like it was designed for easily displaying and identifying the card in their presentations. Though there are a large number of small form factor cases with some kind of riser / 90 degree bend.

Then there are games where holding TAB will display the scoreboard, and holding Alt will show health bars...

In the logjam paper, they speculate that the NSA has the funds to run the first part of a number field sieve on a small number of 1024bit primes. So long as we keep using software implementations with these well known primes hard coded in their source code, HTTPS SSH & VPN connections may be vulnerable.

Not putting all of our eggs in one basket reduces this risk considerably. In response to this threat, we should periodically publish and use a new set of primes that are appropriate for DH exchanges. Though I would be happier if it were possible to generate a new prime on the first boot of a server.

Or we could swap to using a different method for producing session keys.

Good. But it doesn't go far enough. How about some kind of deprecation warning if DH is using any well known prime number?

FTFY. Look, I like the idea of a single cable and port standard. But there should also be a single protocol. At least you'll have USB3 to fall back on, but I imagine many sales drones and users getting confused on the finer points of compatibility issues.

Humans only have = two hands, and they take up a fair amount of vertical space too. Stack the picking arms 10-20 high, build a conveyor belt that can bring shelves to them and they start to look much more interesting. Sure each robot may take a minute to pick one item. But this is an embarrassingly parallel problem.

And add a user preference to control this. Some people might want to see a new poll in the stories list, in the sidebar they might not notice the change. Probably not many people though...

Sure you need a directory service. But it needs to be tamper proof. OP is suggesting that all service names should be public keys. So all DHT records that would be published / fetched can be signed. And the connection to the service can also be signed.

That way noone can guess the current key for a known service. Then the only chance of a sybil attack is to convince someone that your key is the service they are looking for. Something that should only be possible by intercepting the first request.