It's totally unscientific, but I just ran Xbench on my 2009 Mac mini and got around 3 GigaFLOPS. It's not very accurate, but probably good to within an order of magnitude.
According to Wikipedia's supercomputer article, that compares roughly with a 1985 Cray-2, which cost about $25 million at the time and was the size of a large closet.
Or, if you believe we're all about to have our personalities uploaded to the great singularity in the sky like Ray Kurzweil, you could have an instance of you uploaded to a tiny computer-starship, and live in a virtual environment for the entire journey.
For an interesting and entertaining take on this concept (and other singularity-related ideas) check out the novel Accelerando by Charles Stross.
It's a great book by a fellow Slashdot user, and you can download it free!
There's going to be a whole lot of pissed off Navy pilots if they make a UAV that can land on a carrier deck at night in crap weather. Their main reason for superiority over all other pilots will be shot to hell.
I'm the senior Landing Signal Officer for the US Navy's Atlantic Fleet, and we've actually had fully automated landing systems on carrier aircraft for a long while. The first test of any Automatic Carrier Landing System (ACLS) was in August 1957, and after extensive development the system was regularly used in Vietnam. The current AN/SPN-46 is the latest iteration, but essentially it's just a glorified missile tracking radar that feeds into the airplane's autopilot via a simple UHF datalink. It's all old tech.
While not all aircraft since Vietnam have done it well (my old F-14B Tomcat was actually worse at "Mode I" (fully coupled) ACLS approaches than the F-4 Phantom it replaced) today's Hornets and Super Hornets are very smooth when coupled up -- much smoother than the typical manual landing.
The problem comes when the system fails (something that can happen in any large automated system - in the air or on the ground). Pilots regularly practice landing by hand, because they never know when the ACLS might not be there for them. They could perform coupled approaches every pass, but they wouldn't have the skills to confidently get aboard if the system ever went away. Those skills require lots of practice to stay sharp, and landing at sea is really hard. I've been doing it for ten years, and it's still just as challenging as ever.
Sometime in the next decade the N-UCAS is supposed to demonstrate truly autonomous UAV operations in a carrier environment. It will rely on a draft version of our next-generation GPS-based replacement for the SPN-46: JPALS. It's stated goal is to fully integrate with our normal manned carrier air traffic procedures. Having seen highly trained aviators struggle with the challenges of operating around the boat, I'll be impressed if it lives up to its goals.
Examples of prohibited uses include, without limitation, the following: ...
(ii) as a substitute or backup for private lines, landlines or full-time or dedicated data connections;
Someone correct me if I'm wrong, but doesn't this seem to imply that you *must also* have dedicated home telephone *and* data service or you're violating their TOS?
A big name like HP offers a Linux laptop that you can buy on Amazon.com and in BestBuy, and then suddenly the public will see Linux in a very different light.
If you want Ubuntu, and you want this device, you can simply go and buy it. In other words, they'd rather you buy the Ubuntu-branded version than buy the Windows version, then download and install Ubuntu.
I wonder if this will successfully force the sales numbers to reflect Ubuntu's popularity?
The poster asked a very legitimate question that points out the real issue here, namely that people are addressing their mildenhall.af.gov mail to mildenhall.com.
Thanks for your insightful post, and if I hadn't already commented on this thread I'd mod you up.
Those very systems you mention exist and are used religiously. I have no doubt that, had the misrouted information been truly sensitive, there would have been a much bigger reaction.
The challenge is the part you mention here:
The other (available only to those who really need it) is able to send stuff to the outside world.
Unfortunately, it's the other way around. The system connected to the outside world is available to nearly everyone, since nearly everyone has legitimate needs in that arena. Not surprisingly, these are also your less trained, less experienced, less trusted users. As another user comments below, PEBKAC.
Thanks for the response and I agree completely. Sensitive information has to stay on the high-side.
But what about the guy who sends a "Hey John, what's up?" e-mail to his bud, Airman John Doe at john.doe@mildenhall.com. What happens when that happens a few thousand times?
There was obviously a traffic problem in addition to a sensitivity issue with mildenhall.com. That much well intentioned but incorrectly addressed e-mail could easily overwhelm a small site.
And what do you do as a network administrator of a site with a customer at widgets.net or widgets.us or widgets.info when a bunch of your users keep inadvertently sending e-mail to your (or his) competitor at widgets.com?
It's easy to poke fun at the Air Force, but this is a serious IT question.
How do you keep (sometimes stupid) users from sending proprietary (or even run of the mill) e-mail to addresses with the wrong.tld?
It's not as easy as blocking all.com mail, or rerouting that mail to.mil addresses, since they certainly have users with legitimate e-mail needs that send mail to.com accounts. Even blocking mildenhall.com might prevent some legitimate use of a tourist site, perhaps for military with families visiting the area.
Additionally, that wouldn't solve the greater problem which could easily crop up again with randolph.com, eglin.com, edwards.com or any number of similarly named commercial sites.
Education has its limits, and even experienced users will type the wrong.tld occasionally in the heat of the moment.
Certainly nobody should send sensitive information unencrypted over non-secure channels, but it sounds like the biggest problem here was the volume of the traffic.
Naah, I just forgot my [tongue-in-cheek][/tongue-in-cheek] tags.;) Actually, I think the Orwellian accusations thrown at both Apple and Microsoft can be pretty funny sometimes.
I was impressed at how quickly and seamlessly they fixed the video, but if Apple's marketing team can't pull that off then I don't know who can.
Slashdot Mirror
Source (link) please.
It's in the WWDC 2010 Keynote.
A summary can be found in this article.
More like:
Mac OS X 10.7 Liger
"It's pretty much my favorite animal."
- Steve Jobs
You mean like this?
"No bucks, no Buck Rogers"
Exactly.
It's totally unscientific, but I just ran Xbench on my 2009 Mac mini and got around 3 GigaFLOPS. It's not very accurate, but probably good to within an order of magnitude.
According to Wikipedia's supercomputer article, that compares roughly with a 1985 Cray-2, which cost about $25 million at the time and was the size of a large closet.
All we have to do is wait about 25 years.
Or that it will never allow itself to fall into the hands of the likes of you and me.
The Singularity Sky series by Charles Stross shows what one of the more benign versions of that future might look like.
Or for something less benign you could always watch Terminator...
Or, if you believe we're all about to have our personalities uploaded to the great singularity in the sky like Ray Kurzweil, you could have an instance of you uploaded to a tiny computer-starship, and live in a virtual environment for the entire journey.
For an interesting and entertaining take on this concept (and other singularity-related ideas) check out the novel Accelerando by Charles Stross.
It's a great book by a fellow Slashdot user, and you can download it free!
(Then go buy some of his other fine works)
It wouldn't fit in the bay:
B61 - 11.8 feet (141.6 inches) long
Predator C's weapons bay - 10 ft long
No pilot. Less payload than a Viper. Lame.
Not these, but something close: Scan Eagle
There's going to be a whole lot of pissed off Navy pilots if they make a UAV that can land on a carrier deck at night in crap weather. Their main reason for superiority over all other pilots will be shot to hell.
I'm the senior Landing Signal Officer for the US Navy's Atlantic Fleet, and we've actually had fully automated landing systems on carrier aircraft for a long while. The first test of any Automatic Carrier Landing System (ACLS) was in August 1957, and after extensive development the system was regularly used in Vietnam. The current AN/SPN-46 is the latest iteration, but essentially it's just a glorified missile tracking radar that feeds into the airplane's autopilot via a simple UHF datalink. It's all old tech.
While not all aircraft since Vietnam have done it well (my old F-14B Tomcat was actually worse at "Mode I" (fully coupled) ACLS approaches than the F-4 Phantom it replaced) today's Hornets and Super Hornets are very smooth when coupled up -- much smoother than the typical manual landing.
The problem comes when the system fails (something that can happen in any large automated system - in the air or on the ground). Pilots regularly practice landing by hand, because they never know when the ACLS might not be there for them. They could perform coupled approaches every pass, but they wouldn't have the skills to confidently get aboard if the system ever went away. Those skills require lots of practice to stay sharp, and landing at sea is really hard. I've been doing it for ten years, and it's still just as challenging as ever.
Sometime in the next decade the N-UCAS is supposed to demonstrate truly autonomous UAV operations in a carrier environment. It will rely on a draft version of our next-generation GPS-based replacement for the SPN-46: JPALS. It's stated goal is to fully integrate with our normal manned carrier air traffic procedures. Having seen highly trained aviators struggle with the challenges of operating around the boat, I'll be impressed if it lives up to its goals.
http://xkcd.com/386/
Examples of prohibited uses include, without limitation, the following:
...
(ii) as a substitute or backup for private lines, landlines or full-time or dedicated data connections;
Someone correct me if I'm wrong, but doesn't this seem to imply that you *must also* have dedicated home telephone *and* data service or you're violating their TOS?
See? THIS is why you can't have nice things.
A big name like HP offers a Linux laptop that you can buy on Amazon.com and in BestBuy, and then suddenly the public will see Linux in a very different light.
As a matter of fact, they already do.
I wonder if this will successfully force the sales numbers to reflect Ubuntu's popularity?
Why is this modded -1?
The poster asked a very legitimate question that points out the real issue here, namely that people are addressing their mildenhall.af.gov mail to mildenhall.com.
Thanks for your insightful post, and if I hadn't already commented on this thread I'd mod you up.
The challenge is the part you mention here:
Unfortunately, it's the other way around. The system connected to the outside world is available to nearly everyone, since nearly everyone has legitimate needs in that arena. Not surprisingly, these are also your less trained, less experienced, less trusted users. As another user comments below, PEBKAC.
Well said.
.tld when they address their e-mail.
I asked a similar question a few comments above. The problem is senders typing the wrong
I don't know of any easy solutions.
Thanks for the response and I agree completely. Sensitive information has to stay on the high-side.
But what about the guy who sends a "Hey John, what's up?" e-mail to his bud, Airman John Doe at john.doe@mildenhall.com. What happens when that happens a few thousand times?
There was obviously a traffic problem in addition to a sensitivity issue with mildenhall.com. That much well intentioned but incorrectly addressed e-mail could easily overwhelm a small site.
And what do you do as a network administrator of a site with a customer at widgets.net or widgets.us or widgets.info when a bunch of your users keep inadvertently sending e-mail to your (or his) competitor at widgets.com?
It's easy to poke fun at the Air Force, but this is a serious IT question.
.tld?
.com mail, or rerouting that mail to .mil addresses, since they certainly have users with legitimate e-mail needs that send mail to .com accounts. Even blocking mildenhall.com might prevent some legitimate use of a tourist site, perhaps for military with families visiting the area.
.tld occasionally in the heat of the moment.
How do you keep (sometimes stupid) users from sending proprietary (or even run of the mill) e-mail to addresses with the wrong
It's not as easy as blocking all
Additionally, that wouldn't solve the greater problem which could easily crop up again with randolph.com, eglin.com, edwards.com or any number of similarly named commercial sites.
Education has its limits, and even experienced users will type the wrong
Certainly nobody should send sensitive information unencrypted over non-secure channels, but it sounds like the biggest problem here was the volume of the traffic.
Does anyone have a good solution to this problem?
Naah, I just forgot my [tongue-in-cheek][/tongue-in-cheek] tags. ;) Actually, I think the Orwellian accusations thrown at both Apple and Microsoft can be pretty funny sometimes.
I was impressed at how quickly and seamlessly they fixed the video, but if Apple's marketing team can't pull that off then I don't know who can.
Cheers!
In true revisionist fashion, the video on Apple's website has now been fixed.
Mistakes
"It could be that the purpose of your life is only to serve as a warning to others."