My point is they've already done the paying years ago - they don't pay per byte charges on the fibre they own.
As for 'pay to connect', their network is sufficiently large as to merit peering at IXs - IE minimal cost (routing hardware, maintenance, etc.). They can trade their unused fibre bandwidth for the ISP last mile bandwidth.
But they don't pay for all of the pipes... Remember all that Dark Fibre they bought up in 2007?
I remember thinking they're preparing for this sort of thing (in one form or another) - they're pretty good at anticipating trends.
If they've got the backbone bandwidth to trade for last mile bandwidth they'll be able to operate at substantially lower cost than other
high bandwidth users (read:Apple, Microsoft, Amazon, Facebook, Twitter - prime competetors all).
While I wholeheartedly welcome the opportunity to improve some of the frankly stupid laws floating around at the moment, the pessimist in me wonders how this will be twisted by lobbying into some ridiculous new round of laws. I'm going to wait six months before I celebrate this.
So when the virtual particle pair is created at the event horizon, one is trapped stationary beyond the horizon, and the other escapes (becoming real).
In this experiment obviously the event horizon doesn't persist indefinitely, so when the horizon collapses, do the 'trapped' photons escape? and hence is there a time delayed double emission of the hawking radiation? Would this provide a testable signature?
Hope you don't mind me requoting some of my earlier comments:
I've had plenty of hardware failures, it's just that I've never found a restore disc to be that useful. The only real benefit of a restore CD (for restoring, rather than OEM economic purposes) is the drivers, and I find they get out of date pretty fast. When I restore (or repair) a Win install, I'll use a Win disk and download the new drivers directly.
Windows computers I look after either have remote deployment or I already have appropriate physical media (I've got drawers stuffed full of the shiny holographic XPsp2 CDs.)
As for Linux machines, the distros are available everywhere - even if I didn't have a live flashdrive.
I'm not saying unbundling the CD is right for everyone, but it is for me.
Having a restore disc is really not "optional". It's just as important as having a backup of your files
I can see your point, but I disagree. I've had plenty of hardware failures, it's just that I've never found a restore disc to be that useful. The only real benefit of a restore CD (for restoring, rather than OEM economic purposes) is the drivers, and I find they get out of date pretty fast. When I restore (or repair) a Win install, I'll use a Win disk and download the new drivers directly.
That said, as I've said in reply to a comment above, many of my machines are remote deployment, or Linux (I use Linux personally).
I'm not saying unbundling the CD is right for everyone, but it is for me.
Windows computers I look after either have remote deployment or I already have appropriate physical media (I've got drawers stuffed full of the shiny holographic XPsp2 CDs.) (Although I admit that route loses the install time OEM tailored drivers)
As for Linux machines, the distros are available everywhere - even if I didn't have a live flashdrive.
Personally, I have never used a recovery CD. When I buy a PC - I do not need or want the recovery CD (It just fills up cabinet space). If this cost is unbundled (and I'm not saying it is) - I'd prefer to pay a little less and not receive the physical media.
It's (obviously) MD5 length. The results of a quick reverese MD5 lookup are as follows:
USCYBERCOM plans, coordinates, integrates, synchronizes and conducts activities to: direct the operations and defense of specified Department of Defense information networks and; prepare to, and when directed, conduct full spectrum military cyberspace operations in order to enable actions in all domains, ensure US/Allied freedom of action in cyberspace and deny the same to our adversaries.
However, as we all know, MD5 isn't 1-1. It could well just be a conincidence, or something completely different.
Google handled approx 88 billion searches in Dec-2009.
(88b/31)*2=5.67billion searches in two days.
If (conservativly) one tenth of those are work related, that's 567m.
If one in ten work related users plays this once for 60 seconds, that's 3.4 billion seconds.
3.4 billion seconds is approximatly 108 person-years worth of productivity.
(Which at US federal minimum wage is about 1.6 million dollars).
That's a low figure as those who need google to work probably don't earn minimum wage.
Now that's power!
I personally played for more than 60s....
As an Englishman who lives in France from time to time, I can attest to this. In fact it is sometimes hard to get people not to speak English. My early attempts at French were immediately met with English making practice a bit difficult.
Incidentally, you should probably go with 'vous' instead of 'tu'. The 'tu' form is for people you know pretty well (family, good friends, etc...).
Continuing your analogy and at the risk of starting an argument; It's more like he's asking for advice on increasing the performance of his car because he wants to get A to B in under 2 hours. If someone were to point out that the train only takes 1 hour, it would be a point worthy of consideration.
Most modern O/S support suspend to disk which can give you a usable desktop in under 20s. Per your example both XP and Ubuntu can do it in that time. And that's ignoring the even faster suspend to ram which almost all laptops feature these days (granted that for that there is a power requirement).
It's not in the 'spirit' of your question, but perhaps it's a better solution to your problem?
To be capable of infecting machines it would either have to be:
Directly executable, in which case (window) users would have to rename the file.
or
Exploit an unpatched flaw in a media player, which dramatically reduces your targets.
Neither scenario is conducive to rapid, widespread infection. I doubt anyone would go to the trouble of specifically crafting a file that passed signature checks (esp if the signatures checks are chunked) with such a small target audience. Also bear in mind that any virus that did match a know signature would require people to be after that specific track, further limiting your targets.
I am not a biologist, so I'll just ask; "Is it reasonable to assume that there even is a fear scent in humans?". Is there any evidence of it's exsitence in closely related species?
Sorry, wasn't quite clear there; I meant "character.weapon.mod1" to represent inheritance, not as an identifier; which as you quite correctly point out does not take into consideration asset abstraction, which I suppose complicates things, taking my estimates out of context.
Having played mass effect and coded games, would you agree with me that it really doesn't constitue a significant portion of dev?
I'm not talking about inventory systems in general. I'm talking about the Mass Effect inventory system. ALL items of any type have 3 stats, there is a hard limit of 150 items (of any type). There is no weight or size. There are no bags, you can just have 0-150 items in your inventory. That's it.
I should learn to keep quiet about these things, but I'm not trying to be arrogant. I work in C++ OpenGL interfaces every day, and based on my experience of the complexity of the inventory system, this is my genuine opinion about the length time it would take to implement. I'm not talking about the rest of the game - that would take a long time, but the inventory system along - Given the design spec and drawings, really isn't that complex imo.
You're absolutely right that people will insist on changes, however I'm just talking from a strict number of hours to write POV. As for feedback and use cases, the inventory system in mass effect is incredibly simple. [On the assumption you've not played it, theres a handful of slots where you can place items from your inventory, a 150 item limit on the inventory].
For modifications, that's just an item id in something like '(char)character.weapon.mod1'. Consumable, you can press a button (I think it's Y) do remove the item from inventory, and add id to another stat. (Again - really simple).
Slashdot Mirror
For reference, they already peer vast quantities of traffic at IXs anyway : Google Peering Info
My point is they've already done the paying years ago - they don't pay per byte charges on the fibre they own.
As for 'pay to connect', their network is sufficiently large as to merit peering at IXs - IE minimal cost (routing hardware, maintenance, etc.). They can trade their unused fibre bandwidth for the ISP last mile bandwidth.
But they don't pay for all of the pipes... Remember all that Dark Fibre they bought up in 2007?
I remember thinking they're preparing for this sort of thing (in one form or another) - they're pretty good at anticipating trends. If they've got the backbone bandwidth to trade for last mile bandwidth they'll be able to operate at substantially lower cost than other high bandwidth users (read:Apple, Microsoft, Amazon, Facebook, Twitter - prime competetors all).
While I wholeheartedly welcome the opportunity to improve some of the frankly stupid laws floating around at the moment, the pessimist in me wonders how this will be twisted by lobbying into some ridiculous new round of laws. I'm going to wait six months before I celebrate this.
Great - so when my wireless internet enabled skin display gets infected with spam, I'll have adverts for viagra appearing on my chest. Brilliant
So when the virtual particle pair is created at the event horizon, one is trapped stationary beyond the horizon, and the other escapes (becoming real).
In this experiment obviously the event horizon doesn't persist indefinitely, so when the horizon collapses, do the 'trapped' photons escape? and hence is there a time delayed double emission of the hawking radiation? Would this provide a testable signature?
Any physicists know?
I've had plenty of hardware failures, it's just that I've never found a restore disc to be that useful. The only real benefit of a restore CD (for restoring, rather than OEM economic purposes) is the drivers, and I find they get out of date pretty fast. When I restore (or repair) a Win install, I'll use a Win disk and download the new drivers directly.
Windows computers I look after either have remote deployment or I already have appropriate physical media (I've got drawers stuffed full of the shiny holographic XPsp2 CDs.)
As for Linux machines, the distros are available everywhere - even if I didn't have a live flashdrive.
I'm not saying unbundling the CD is right for everyone, but it is for me.
Having a restore disc is really not "optional". It's just as important as having a backup of your files
I can see your point, but I disagree. I've had plenty of hardware failures, it's just that I've never found a restore disc to be that useful. The only real benefit of a restore CD (for restoring, rather than OEM economic purposes) is the drivers, and I find they get out of date pretty fast. When I restore (or repair) a Win install, I'll use a Win disk and download the new drivers directly.
That said, as I've said in reply to a comment above, many of my machines are remote deployment, or Linux (I use Linux personally).
I'm not saying unbundling the CD is right for everyone, but it is for me.
Nope.
Windows computers I look after either have remote deployment or I already have appropriate physical media (I've got drawers stuffed full of the shiny holographic XPsp2 CDs.) (Although I admit that route loses the install time OEM tailored drivers)
As for Linux machines, the distros are available everywhere - even if I didn't have a live flashdrive.
Personally, I have never used a recovery CD. When I buy a PC - I do not need or want the recovery CD (It just fills up cabinet space). If this cost is unbundled (and I'm not saying it is) - I'd prefer to pay a little less and not receive the physical media.
Am I the only one who noticed that the colony pictured in the article is more likely a Standford Torus, or am I just being picky?
(The US Cyber Command mission statement.)
....as stated by everyone else here, replies to the original article, wikipedia etc....
Yet another reminder to me to use Google first, then do things like reverse MD5 lookups.....
It's (obviously) MD5 length. The results of a quick reverese MD5 lookup are as follows :
USCYBERCOM plans, coordinates, integrates, synchronizes and conducts activities to: direct the operations and defense of specified Department of Defense information networks and; prepare to, and when directed, conduct full spectrum military cyberspace operations in order to enable actions in all domains, ensure US/Allied freedom of action in cyberspace and deny the same to our adversaries.
However, as we all know, MD5 isn't 1-1. It could well just be a conincidence, or something completely different.
Google handled approx 88 billion searches in Dec-2009. (88b/31)*2=5.67billion searches in two days. If (conservativly) one tenth of those are work related, that's 567m. If one in ten work related users plays this once for 60 seconds, that's 3.4 billion seconds. 3.4 billion seconds is approximatly 108 person-years worth of productivity. (Which at US federal minimum wage is about 1.6 million dollars). That's a low figure as those who need google to work probably don't earn minimum wage. Now that's power! I personally played for more than 60s....
As an Englishman who lives in France from time to time, I can attest to this. In fact it is sometimes hard to get people not to speak English. My early attempts at French were immediately met with English making practice a bit difficult. Incidentally, you should probably go with 'vous' instead of 'tu'. The 'tu' form is for people you know pretty well (family, good friends, etc...).
Continuing your analogy and at the risk of starting an argument; It's more like he's asking for advice on increasing the performance of his car because he wants to get A to B in under 2 hours. If someone were to point out that the train only takes 1 hour, it would be a point worthy of consideration.
Most modern O/S support suspend to disk which can give you a usable desktop in under 20s. Per your example both XP and Ubuntu can do it in that time. And that's ignoring the even faster suspend to ram which almost all laptops feature these days (granted that for that there is a power requirement).
It's not in the 'spirit' of your question, but perhaps it's a better solution to your problem?
- Directly executable, in which case (window) users would have to rename the file.
or- Exploit an unpatched flaw in a media player, which dramatically reduces your targets.
Neither scenario is conducive to rapid, widespread infection. I doubt anyone would go to the trouble of specifically crafting a file that passed signature checks (esp if the signatures checks are chunked) with such a small target audience. Also bear in mind that any virus that did match a know signature would require people to be after that specific track, further limiting your targets.There's no need to be rude, I was just stating my opinion. Ad hominem arguments are fallacious.
I am not a biologist, so I'll just ask; "Is it reasonable to assume that there even is a fear scent in humans?". Is there any evidence of it's exsitence in closely related species?
Sorry, wasn't quite clear there; I meant "character.weapon.mod1" to represent inheritance, not as an identifier; which as you quite correctly point out does not take into consideration asset abstraction, which I suppose complicates things, taking my estimates out of context.
Having played mass effect and coded games, would you agree with me that it really doesn't constitue a significant portion of dev?
I'm not talking about inventory systems in general. I'm talking about the Mass Effect inventory system. ALL items of any type have 3 stats, there is a hard limit of 150 items (of any type). There is no weight or size. There are no bags, you can just have 0-150 items in your inventory. That's it.
I should learn to keep quiet about these things, but I'm not trying to be arrogant. I work in C++ OpenGL interfaces every day, and based on my experience of the complexity of the inventory system, this is my genuine opinion about the length time it would take to implement. I'm not talking about the rest of the game - that would take a long time, but the inventory system along - Given the design spec and drawings, really isn't that complex imo.
If it's big enough to be visible on google earth, it's a pity they don't give the coordinates so we could all have a look.
You're absolutely right that people will insist on changes, however I'm just talking from a strict number of hours to write POV. As for feedback and use cases, the inventory system in mass effect is incredibly simple. [On the assumption you've not played it, theres a handful of slots where you can place items from your inventory, a 150 item limit on the inventory].
For modifications, that's just an item id in something like '(char)character.weapon.mod1'. Consumable, you can press a button (I think it's Y) do remove the item from inventory, and add id to another stat. (Again - really simple).