You do realize that 1968 was a long time ago??

Maybe if RCS actually existed, then people would move to it?

Once proof of stake goes through, doesn't it make the problem worse? Now instead of getting money from both minting coins and doing transactions, people running the blockchain rigs will only get transaction money. I would think that this means either a) transaction costs increase or b) it becomes unprofitable for many miners to run their rigs, leading to more consolidation of giant operations with preposterously cheap electricity.

Who is clamoring for that, and are blockchains + POW the optimal way to do that? I'd argue the frameworks of Certificate Transparency could probably get you most of the way there

> I mean, it's not like facebook provide identity services, allowing a facebook login to be used for accessing other services run by third parties. If they offered that service, things might be different.

They do.

If the problem was certificates, we would've pushed new certificates. It's the ancient SSL/TLS implementations that are the problem.

I agree that IPMI controllers, in general, don't have their security well-implemented. That's the whole reason we have a physically isolated network for the IPMI traffic in the first place.

My pet peeve -- all the IPMI management consoles whose SSL causes firefox/chrome to refuse to even allow you to request to ignore the weak encryption, then the java console access applets which new versions of Java refuse to let you load.

All of our IPMI interfaces are connected via a physically separate network, which only a single locked-down machine has access to. It ends up being quicker to open a ticket to ask someone down in the DC go touch the hardware instead of trying to manage some things remotely over IPMI

Have any keywords I could search for about the OGL mess? I remember it being stagnant for a while, but I don't remember any of the details.

This is me being naiive, but if it's a stationary, land-based installation, do you really care if you need flywheels that are 20x as massive as comparable Li batteries? A slug of iron's gotta be a lot cheaper than a mountain of lithium, right?

net energy consumption was pretty negligible

As long as you have infinite empty cars uphill :)

The docs roughly say, "POSIX is crufty and a bad API, we're not afraid to not implement those parts in order to get a cleaner interface".

Okay, but what does this new API look like instead? Will the features enabled by these not-implemented APIs be reimplemented in some other, cleaner fashion? Or will that functionality just not be provided at all?

Question from a layman: Do GPUs have a physical-virtual memory mapping? Ex: Could process A get space on the GPU, then when process B requests memory, the GPU would give process B the physical pages process A used to have (while copying the actual data over) to defragment the physical pages, leaving B with a copy of what A used to own? Or, perhaps process B requests so much memory that the texture space requested by process A gets paged out to main memory?

That's true. I wonder what the $/kg for mining lunar He3 compares with building a ton of nuclear bombardment facilities works out. Not that any of it is at all practical anyway...

1) the vast majority of earth helium is He4, which has some undesirable properties for fusion

2) even if it was all He3, there's not a lot of helium. He is light enough that it simply floats away from the earth

Isn't that the 80/20 rule? I know I'm guilty of doing the "exciting" stuff on little libraries, then when it comes to the bits like packaging/documentation/etc.., putting it off to "later". It's just not as fun to do those little maintenance stuff as it is the neat parts