There should be an easier way.
Music is encoded at 44khz * 24-bit samples. That's about 1Mbits per second of audio.
Why not just run through all the permutations of possible 1s and 0s and copyright that? You'd literally have every possible 1-second sound that could exist.
Now, as to the computational power you'd need, well, there's the rub.
Can't speak to anything other than what I've seen: People aren't retiring.
At my last place, we had a workers comp claim by a mid-level manager; she tripped on a power cable and broke her hip. She was already in her late 60s at the time of the accident. And after taking 2 years off to recover, SHE CAME BACK!
We had to make crazy adjustments to our workflow so that OSHA/workers comp would be satisfied.
I was scowled at for suggesting we *not* hire people too old to see the cables on the floor. Or people with such advanced bone density loss that a simple slip/fall means 2 years of physical therapy.
How can a junior manager (or technician) seriously consider staying in a job where his/her boss is only a few years older and is literally planning on working till they die?
I used to put yellow stickie notes inside the RS-232 (530 and 449 connector shells were too full of wires) cables I built.
Thought about it after tearing apart a bundle of 25-pair cables and found a 25-year-old note expressing my predecessor's displeasure in having his work undone.
IIRC, you are using the term NAT when you really mean PAT. In true NAT, you will have X internal addresses mapped to Y external addresses.
If X>Y, then you may have requests get dropped or mangled.
PAT is 1 external to many internal shifting/translating the port numbers to create a unique channel.
As long as Internal32768, then you should be okay ; you need to reserve a port for each end of the channel. Realistically, most channels will have 80\443 as an end point. On those types of networks, you can get much closer to 65535. Still, a few badly.configured torrent clients can easily exhaust ports and bring the network down with almost no utilization.
You'll see a lot of references to defense in depth. If you browse a CISSP syllabus, you'll see they talk about everything from parking lot lighting to ring 0 code. Between an adequately lit parking structure and ring 0, there are a lot of things you can do. Each one adds a bit more security. You do hit diminishing returns quickly, but host-based firewalls are quick and cheep.
To harden a host based fw, turn on remote logging and have the logging server flag configuration changes as critical.
No one should be doing a configuration change without notifying your change mgmt team. If they get a red line on their monitor, they contact and chew out the offending employee. If no one feses up, nuke the server, restore, and re-harden.
It is important to know that your server administration can also be the change manager on small teams. You just need to have him/her mentally firewall the two jobs.
That seems like a bad idea. Roads take a lot of abuse.
Why not make mile markers and guardrails with small windmills on them. The drafts from traffic would drive the generators.
They do this in Japan; kinda. The reflectors on the roadside use the drafts from cars to spin a protective disc that cleans the reflectors. It's not 100%, but nothing ever is...
Not the OP, but it *can* be that way sometimes.
I dropped $12 on each of 2 titles from DCS:World on a special a few weeks ago. They are normally $40, but Steam was running a big sale to promote a new DCS release.
Anyway, it's a $12 game that probably needs $2500 of hardware to run properly. The graphics are part of it, but the background AI really crushes the CPU as well.
Truly.
Lots of people hunt. I'd never call a gun a toy...
Clean, old-fashioned hate rears its ugly head. Oh well, I guess it *is* almost September...
Go Dawgs!
Defense in depth. Even if we let a terrorist walk into a silo or hangar and take a nuke, he still would never be able to detonate it without the PAL.
http://en.wikipedia.org/wiki/Permissive_Action_Link
Even when there is a wealth of information on the device (like the B-61 http://www.glennsmuseum.com/controller/controller.html ) there is very little chance someone could recover a working nuclear core. There are multiple layers of failsafes which (probably) fire a small charge inside the sealed core to destroy the pre-detonation subsystems.
So, given the option of ordering a fully capable B-61 off of ebay, or stealing an old USSR/Pakistani/Indian nuke, the terrorist would be better off with the second option.
I don't know anything about PALs or failsafes for their nukes.
I think you hit on the major problem with (US) gun laws. There are something like 300M+ guns in the USA. Supposedly, those are guns in circulation. i.e. Not sitting in a warehouse or on in a gun shop display case.
It is an easy thing for anyone to obtain a gun illegally for illegal purposes. It is somewhat less easy to obtain a gun legally for legal purposes.
And it's only moderately difficult to modify a legally obtained gun to exceed the parameters that make it legal.
Larger magazines, modified lowers and uppers, mil-dot scopes, and general tweaks can turn any gun into an "assault weapon".
As to the mental stability, the US has a bad track record for dealing with mental issues. Even today, it's hard to get good mental help. You could self-identify today as a potential danger to yourself or to others and it could be 2+ years before you are stable.
And that's for someone who wants to get help. Most don't.
I disagree that we need a gun that can determine if you are mentally compromised. It's just too "sci-fi" to be viable. What we do need is a way to track guns from cradle to grave and hold the registered owners accountable for illegal use.
If the plane crashes due to pilot error, then the pilot will never be charged with a crime.
With a UAV, you can immediately test the pilot for drugs/alcohol or any other condition that altered his mental capacity.
If the plane broke, then it sucks to be on the plane. But if the pilot screwed up, then his peers will get to see him charged with manslaughter and dragged through the legal system. Ultimately leaving him a broken shell of a man. Probably in jail.
3 or 4 of those and the other pilots will either become very clever or stop screwing up.
Information? Or in formation?
If the research develops into a marketable product, patents would protect them from other companies.
If the research doesn't pan out, then nothing is lost.
As for fund sharing, if Cali funds 20% of research that goes on to become the next Cisco, Google, etc, then Cali should get 20% of the company stock. Or get a 20% cut of quarterly revenue.
An adequate bootstrap is a contradiction in terms.
