Rights are a human invention. People in general have rights because they banded together and decided to create governments that have certain limits on their powers, among other things. Rights are not things that just magically exist, and that idea is superstitious nonsense.
Wording aside, the issue is whether people have rights in the absence of a government granting those rights or not. The US Constitution was written by people who felt strongly that people were born with rights, and it was just the role of the government to help protect those rights.
Now, if you feel otherwise that is nice, but you can't read the US Constitution through that lens - it just doesn't make any sense. Heck, the 9th amendment makes no sense if you consider human rights to be merely an emergent result of restrictions imposed on government.
My point isn't debating the philosophical origin of human rights - it is an interesting topic and I agree that there are a LOT of ways of looking at it. My point is just that the US constitution is written from a particular viewpoint and trying to look at it differently just results in a document that doesn't make sense, or which can be used to justify all kinds of abuse. The whole reason the 9th amendment was written was to prevent the absence of a clear prohibition on government power to be used as justification for the government having that power.
On the other hand, the first amendment says congress shall make no laws abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the government for a redress of grievances.
In this case, it is pretty clear what the intent was. Congress is not to make any laws. It doesn't say that you cannot contract your speech rights away with a third party or that the loss of free speech in any way is prohibited.
Yes, but in order to enforce that contract somebody would have to take you to court, and courts don't exist but for the laws that establish them. Congress isn't allowed to pass a law that creates a court that has the power to restrict somebody's freedom of speech, thus the courts that exist today aren't allowed to restrain people's speech. So, sure, you can put something in a contract restraining their speech, but no court is Constitutionally allowed to enforce such a clause.
Note, for the people who aren't local to the US - the sheriff is a county-level official, in no way associated with Congress. Hell, the sheriff isn't even associated with the State government.
State governments are independent from the Federal government, but local governments are subordinate to the state government. That is, the Federal government is, well, Federal. State governments are unitary. That means that the US government cannot remove a state governor from power, but a state government can remove a local mayor, town council, etc from power and even get rid of the legal town entity entirely.
Agree. Unless your router is made by a company associated with a government fairly adversarial to the US and only handled by couriers associated with similar governments, I'd assume that it is compromised if the NSA has any interest in compromising it. I'm sure the NSA can intercept routers made in the EU if it wants to with the cooperation of local governments (or the local government will compromise it and hand the NSA a feed, likely in exchange for the NSA sharing something else with them). Governments are pretty cozy about these sorts of things no matter what they say in public.
Now, if you buy a Chinese router using a Chinese courier and have it delivered in a country like China then you're fairly unlikely to be subject to NSA tampering. Of course, you're fairly likely to be subject to Chinese tampering instead.
You can't keep secrets from governments as a private individual - at least not unless you're allied with some other government that you can't keep secrets from.
Yup. I'd be shocked if most nations didn't do this sort of thing. You don't really get a choice in whether the hardware you buy is backdoored - you just get to pick whose backdoor it comes with.
Diplomatic credentials are granted by the country you travel to, not the country you came from. The US would not grant diplomatic credentials to somebody wanted for a crime.
If it worked as you suggest, than countries would send in spies over the border unlawfully all the time and if they got caught they'd just be sent home to try again. The reason spies are sent in NOC is so that the country they're visiting doesn't know they're there, which also means that they can be arrested as they have no diplomatic immunity. Diplomats with immunity have to walk up to immigration and declare themselves like anybody else (though they may very well get a special line). Every country works this way.
To be fair - an act of war is whatever somebody wants it to be.
If I'm a dictator of some country and you're the dictator of another country and you sneeze and I don't like it and I order my military to kill all your citizens, then you basically started a war with your sneeze. Or maybe I started it with my craziness. Either way there were some acts and a war, so figure it out however you like.
There are just some particular actions that people think of as crossing a line. The US never really went to war with Cuba and the Soviet Union, though the US did a naval blockade (sorry, quarantine) of Cuba which was clearly an act of war. Basically countries do stuff to each other all the time, and when one calculates that will make out better with bloodshed than diplomacy you have a war.
.. what will the Bahama government/people do - will they sue the US for the presumable crime of breaking into their phone system?
In what court would they do this? You can't sue the US government in a US court without the permission of the US government, and the US will just ignore the ruling of just about any other court.
And yes, many (most?) other countries work the same way...
No argument there. I work for a big company and I'm amazed at the amount of waste that basically amounts to externalizes and other accounting incentives.
Honestly, most of your post boils down to this. Obviously mmapping a file on an OS that apparently has a poor virtual memory management system isn't going to work well. I wasn't aware that you were thinking mainly about Windows.
It will page to disk after time.
That actually isn't dumb - if something isn't being used, then the RAM is probably better used for something like cache. Linux swaps stuff out all the time even when memory is free - this behavior can be tuned. It doesn't always get it right.
Also, when you mmap a file the whole thing starts out as swapped out. Then when you "swap it out" you really just delete the page from memory - you don't have to write a mmapped file to the swap file, because its contents are already stored in the file it is mmapped from. The backing store for mmap isn't the swap file. Ditto for swapping out executing programs - they're already present on disk, so there is no need to write them to a swap file.
I think we're talking past each other or something.
I never suggested that somebody writing software for DOS should just mmap a file to read it - DOS obviously has no mmap system call.
I just got the impression that you thought that we shouldn't be using mmap today - on operating systems that fully support it and on systems that have gobs of virtual memory address space. Back in the day it was done the way it was done for a reason, but those reasons are largely behind us now. I probably wouldn't use mmap to deal with exabyte-sized files today, but for just about any ordinary application it is usually the right tool for the job.
Yup. I could see a new CEO coming in to a screwed-up company and bringing in a few consultants to help him straighten it out, but the goal would be to fire the folks who should be doing the job the consultants had to do, and replace them with somebody competent so that you don't need the consultants.
Big companies shouldn't have huge gaps in skills in anything they do. Sure, I buy the core competencies thing to an extent, but I've seen gross mismanagement basically go ignored. I don't care if records storage isn't your profit-making center - if you have more than 10k employees you can afford to at least have one person in charge of your records storage dept that knows how to do their job. The only thing big companies have going for them is their economy of scale, and yet so often they don't want to embrace that.
I think the reason for half of the outsourcing arrangements out there is that a company hopes that by outsourcing something and having real charge-backs that people will just stop using the service entirely. They don't want to deal with the culture problems that create all the waste in the first place, so they just add more waste on top of it. I'm working on a storage-related project at work and it is almost impossible to get people to realize that if they never actually use something they don't need to store it (no legal requirements here - they just store stuff because they spent a lot of money making it in the first place, and I'm not talking about data that costs $100/TB/year - I'm talking about stuff that has mass and takes up space and often needs special handling).
That's what consultants do. Calling someone who provides outsourcing and asking them if you should outsource will get the same answer every time.
No argument there. I think companies are WAY too dependent on consulting services. I can see why a small business might need to engage them, but a large business that needs to hire a lot of consultants is probably mismanaged.
I'm not talking about bringing in somebody to assist with a project. I'm talking about bringing in people who you'll basically depend on to figure out what the heck you ought to be doing.
Creating virtual memory and swap doesn't change the fact that it gets loaded into "memory" it just lets pedants argue about whether virtualized RAM, stored on a disk is "memory" which is a completely different and unrelated argument to whether modern programs on a modern OS generally "load" all of a file opened, rather than bits and pieces at a time.
You previously said:
There's no reason you need 640k of RAM to read a 2M file. You just can't have all of it in RAM at the same time. That's how it used to be. The idea of ramdriving every program by loading 100% of every program you are running and 100% of every file used by every one of those programs is silly, but it's the new norm.
So, what exactly is wrong with "loading" all of a file when it is opened at once? Your argument seemed to be that it wastes RAM, but it doesn't actually waste RAM as you acknowledge since the OS manages what actually ends up in RAM.
I think the modern approach is much cleaner. You can either implement in every word processor the logic needed to predict what parts of a document need to be in RAM and implement your own memory manager, or you can just mmap the file and let the OS deal with the details of where it gets stored. Then, when somebody comes along and improves the logic in the OS, your 10-year-old-never-touched-since word processor works a little better, instead of being constrained by whatever logic the programmer had time to implement.
I'd think that the time of a word processor programmer would be better spent figuring out how to better handle background re-pagination or adding better text formatting logic, than designing their own memory management system. Should we go the route of WordPerfect back in the day and put the print drivers in the word processor as well?
When the AMD extensions became mainstream, the defacto standard, Intel licensed it. But the extension is still tied to the Intel architecture and Intel is still in control.
What does it even mean to be "in control?" AMD could apparently fork the last architecture/instruction-set/whatever-you-want-to-call-it (AIW for short), and create their own AIW which everybody adopted, just as Intel routinely adds new instructions to newer chips.
I don't question that Intel is usually in the lead with advancing the x86 world. However, it is fair to say that by introducing amd64/x86_64/whatever-you-want-to-call-it (AXW for short) AMD brought 64-bit to the masses. They certainly didn't invent the first 64-bit chip, but they created the first one that anybody was willing to use.
I suspect that today moving to ARM or some other architecture would be easier than it would have been 10 years ago. Back when the Athlon 64 came out virtually all workstations ran Windows, with all software being win32. On the server front Linux was popular, but even in that world there was a lot of non-portable software around. Using AXW made it a lot easier to migrate, and at least in the Linux world I'd like to think that people have learned to avoid reading a 32-bit record using an int data type. In the non-linux world things are much more heterogeneous than they were a decade ago as well. That means that changing instruction sets entirely is more practical than it used to be.
Yup, and the BS about them being first to 64-bit...maybe in the consumer sector, but Intel, IBM and DEC all had 64-bit chips before the Athlon was even designed let alone shipped.
They invented the architecture that you probably typed your post on. That was the point. Heck, on my linux distro it is still called amd64...
Agree. Fuel cells actually make a lot of sense in any kind of transportation situation. They're far more effective than batteries at storing power. They're just not all that practical yet. As far as I'm aware the only fuel cell that REALLY works reliably is the hydrogen-oxygen fuel cell.
True- and large cryogenic tanks and associated filling equipment are probably more realistic on a large aircraft then on public automobiles.
I wasn't even thinking about cryogenic storage. That might make sense for large aircraft. The energy per unit mass of hydrogen is pretty high I think. If kept at atmospheric pressure the storage tanks don't have to be heavy. You'd have constant boil-off, but that is only an issue if you leave aircraft parked with fuel in them, which isn't really the case for large aircraft. For a car having to burn all your fuel within 8 hours of fueling is a big problem, but aircraft fill up before just about every flight.
How is moving everything to the cloud working out for those users?
About as well as things work out for my relatives when they have a hard drive fail without backups. Well, except that the users of the cloud get their data back in a few hours 99% of the time.
I can't say that applications at my workplace which has a professionally-managed datacenter are any more reliable than the typical cloud service. You just don't hear about it on the news when things go down there.
Sure, you and I know how to backup our data, but we're the exception. And for my data which is fairly critical, where do I back it up? Well, that would be on Amazon S3. This way I know that if my house burns down the last backup is always only a few hours old, and not from the last time I grabbed my disks and moved them offsite. And where do I back up my data on Google Drive or Amazon or Dropbox? Well, that would be my PC. If Amazon and my house both have a major catastrophe at the same time I'll be screwed, but chances are in that case I'll have to live in my basement until the fallout cools down.
And since I shouldn't store the keys in the cloud, for every instance that uses some encrypted disk, I need to manually login then provide the key and passphrase.
Encrypting the storage only works if the processing is all done outside of the cloud. The instant you type your passphrase into the cloud server the cloud provider knows what it is if they want to know it. They aren't limited to reading data stored on disk.
If someone claiming to be a networking expert said he was an expert in TPC/IPX (in a context obviously meaning TCP/IP), you wouldn't take that as a good sign.
I know what TCP is, but I'm sure I've posted enough stuff online that you will find a case of me writing TPC to quote me on. Call me when the guy starts going on about "tubes."
Compliance costs money no matter how you do it - having the data in-house often just means that it is easier to pretend that you're doing things right. One of the advantages of outsourcing is that you can actually get contracts with the party managing your data. When you do things in-house you can't make a contract with the IT department, because it is the same legal entity as yourself. Big companies often fall victim to their own internal cost-cutting.
Outsourcing also gives you hard costs on activities. When the work is done in-house the costs are always fuzzy - do you include the cost of the janitor that cleans the toilet the sysadmins sit on? When you outsource you pay by the whatever and your costs come in on an invoice and everybody is accountable for exactly how much they're utilizing.
At work I'm working on outsourcing of non-data-storage. In theory it isn't anything we couldn't do more efficiently in-house. The problem is that due to company culture/etc it is inevitable that if we continue to do things in-house we'll continue to do them poorly, so by outsourcing we can let somebody manage the work who will actually do it properly. I'd rather pay a bit more and ensure that things are being done right than pay less and have to deal with the issues when it turns out that they aren't.
Does rain actually wash off skid marks? Asphalt is dark for the most part, so you can't really see light rubber deposits, but if they are there then they would probably interfere with solar panels underneath.
Everybody gets new tires every few years. All that rubber has to go somewhere, and I suspect that quite a bit of it ends up being absorbed into/onto the streets.
Slashdot Mirror
And here in Europe we _are_ protected in our interactions with private companies just as we are with the government.
Well, unless you live in the UK and try to criticize McDonalds.
Don't get me wrong - I admire a lot of things about most European governments, but it isn't without its flaws.
Rights are a human invention. People in general have rights because they banded together and decided to create governments that have certain limits on their powers, among other things. Rights are not things that just magically exist, and that idea is superstitious nonsense.
Wording aside, the issue is whether people have rights in the absence of a government granting those rights or not. The US Constitution was written by people who felt strongly that people were born with rights, and it was just the role of the government to help protect those rights.
Now, if you feel otherwise that is nice, but you can't read the US Constitution through that lens - it just doesn't make any sense. Heck, the 9th amendment makes no sense if you consider human rights to be merely an emergent result of restrictions imposed on government.
My point isn't debating the philosophical origin of human rights - it is an interesting topic and I agree that there are a LOT of ways of looking at it. My point is just that the US constitution is written from a particular viewpoint and trying to look at it differently just results in a document that doesn't make sense, or which can be used to justify all kinds of abuse. The whole reason the 9th amendment was written was to prevent the absence of a clear prohibition on government power to be used as justification for the government having that power.
On the other hand, the first amendment says congress shall make no laws abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the government for a redress of grievances.
In this case, it is pretty clear what the intent was. Congress is not to make any laws. It doesn't say that you cannot contract your speech rights away with a third party or that the loss of free speech in any way is prohibited.
Yes, but in order to enforce that contract somebody would have to take you to court, and courts don't exist but for the laws that establish them. Congress isn't allowed to pass a law that creates a court that has the power to restrict somebody's freedom of speech, thus the courts that exist today aren't allowed to restrain people's speech. So, sure, you can put something in a contract restraining their speech, but no court is Constitutionally allowed to enforce such a clause.
Note, for the people who aren't local to the US - the sheriff is a county-level official, in no way associated with Congress. Hell, the sheriff isn't even associated with the State government.
State governments are independent from the Federal government, but local governments are subordinate to the state government. That is, the Federal government is, well, Federal. State governments are unitary. That means that the US government cannot remove a state governor from power, but a state government can remove a local mayor, town council, etc from power and even get rid of the legal town entity entirely.
In the International Court of Justice.
Hence my point about "the US will just ignore the ruling of just about any other court..."
Agree. Unless your router is made by a company associated with a government fairly adversarial to the US and only handled by couriers associated with similar governments, I'd assume that it is compromised if the NSA has any interest in compromising it. I'm sure the NSA can intercept routers made in the EU if it wants to with the cooperation of local governments (or the local government will compromise it and hand the NSA a feed, likely in exchange for the NSA sharing something else with them). Governments are pretty cozy about these sorts of things no matter what they say in public.
Now, if you buy a Chinese router using a Chinese courier and have it delivered in a country like China then you're fairly unlikely to be subject to NSA tampering. Of course, you're fairly likely to be subject to Chinese tampering instead.
You can't keep secrets from governments as a private individual - at least not unless you're allied with some other government that you can't keep secrets from.
Yup. I'd be shocked if most nations didn't do this sort of thing. You don't really get a choice in whether the hardware you buy is backdoored - you just get to pick whose backdoor it comes with.
Meh, the US would just phone up the PM and something would get worked out - probably without firing a shot.
Diplomatic credentials are granted by the country you travel to, not the country you came from. The US would not grant diplomatic credentials to somebody wanted for a crime.
If it worked as you suggest, than countries would send in spies over the border unlawfully all the time and if they got caught they'd just be sent home to try again. The reason spies are sent in NOC is so that the country they're visiting doesn't know they're there, which also means that they can be arrested as they have no diplomatic immunity. Diplomats with immunity have to walk up to immigration and declare themselves like anybody else (though they may very well get a special line). Every country works this way.
To be fair - an act of war is whatever somebody wants it to be.
If I'm a dictator of some country and you're the dictator of another country and you sneeze and I don't like it and I order my military to kill all your citizens, then you basically started a war with your sneeze. Or maybe I started it with my craziness. Either way there were some acts and a war, so figure it out however you like.
There are just some particular actions that people think of as crossing a line. The US never really went to war with Cuba and the Soviet Union, though the US did a naval blockade (sorry, quarantine) of Cuba which was clearly an act of war. Basically countries do stuff to each other all the time, and when one calculates that will make out better with bloodshed than diplomacy you have a war.
.. what will the Bahama government/people do - will they sue the US for the presumable crime of breaking into their phone system?
In what court would they do this? You can't sue the US government in a US court without the permission of the US government, and the US will just ignore the ruling of just about any other court.
And yes, many (most?) other countries work the same way...
No argument there. I work for a big company and I'm amazed at the amount of waste that basically amounts to externalizes and other accounting incentives.
Windows is dumb.
Honestly, most of your post boils down to this. Obviously mmapping a file on an OS that apparently has a poor virtual memory management system isn't going to work well. I wasn't aware that you were thinking mainly about Windows.
It will page to disk after time.
That actually isn't dumb - if something isn't being used, then the RAM is probably better used for something like cache. Linux swaps stuff out all the time even when memory is free - this behavior can be tuned. It doesn't always get it right.
Also, when you mmap a file the whole thing starts out as swapped out. Then when you "swap it out" you really just delete the page from memory - you don't have to write a mmapped file to the swap file, because its contents are already stored in the file it is mmapped from. The backing store for mmap isn't the swap file. Ditto for swapping out executing programs - they're already present on disk, so there is no need to write them to a swap file.
I think we're talking past each other or something.
I never suggested that somebody writing software for DOS should just mmap a file to read it - DOS obviously has no mmap system call.
I just got the impression that you thought that we shouldn't be using mmap today - on operating systems that fully support it and on systems that have gobs of virtual memory address space. Back in the day it was done the way it was done for a reason, but those reasons are largely behind us now. I probably wouldn't use mmap to deal with exabyte-sized files today, but for just about any ordinary application it is usually the right tool for the job.
Yup. I could see a new CEO coming in to a screwed-up company and bringing in a few consultants to help him straighten it out, but the goal would be to fire the folks who should be doing the job the consultants had to do, and replace them with somebody competent so that you don't need the consultants.
Big companies shouldn't have huge gaps in skills in anything they do. Sure, I buy the core competencies thing to an extent, but I've seen gross mismanagement basically go ignored. I don't care if records storage isn't your profit-making center - if you have more than 10k employees you can afford to at least have one person in charge of your records storage dept that knows how to do their job. The only thing big companies have going for them is their economy of scale, and yet so often they don't want to embrace that.
I think the reason for half of the outsourcing arrangements out there is that a company hopes that by outsourcing something and having real charge-backs that people will just stop using the service entirely. They don't want to deal with the culture problems that create all the waste in the first place, so they just add more waste on top of it. I'm working on a storage-related project at work and it is almost impossible to get people to realize that if they never actually use something they don't need to store it (no legal requirements here - they just store stuff because they spent a lot of money making it in the first place, and I'm not talking about data that costs $100/TB/year - I'm talking about stuff that has mass and takes up space and often needs special handling).
That's what consultants do. Calling someone who provides outsourcing and asking them if you should outsource will get the same answer every time.
No argument there. I think companies are WAY too dependent on consulting services. I can see why a small business might need to engage them, but a large business that needs to hire a lot of consultants is probably mismanaged.
I'm not talking about bringing in somebody to assist with a project. I'm talking about bringing in people who you'll basically depend on to figure out what the heck you ought to be doing.
Creating virtual memory and swap doesn't change the fact that it gets loaded into "memory" it just lets pedants argue about whether virtualized RAM, stored on a disk is "memory" which is a completely different and unrelated argument to whether modern programs on a modern OS generally "load" all of a file opened, rather than bits and pieces at a time.
You previously said:
There's no reason you need 640k of RAM to read a 2M file. You just can't have all of it in RAM at the same time. That's how it used to be. The idea of ramdriving every program by loading 100% of every program you are running and 100% of every file used by every one of those programs is silly, but it's the new norm.
So, what exactly is wrong with "loading" all of a file when it is opened at once? Your argument seemed to be that it wastes RAM, but it doesn't actually waste RAM as you acknowledge since the OS manages what actually ends up in RAM.
I think the modern approach is much cleaner. You can either implement in every word processor the logic needed to predict what parts of a document need to be in RAM and implement your own memory manager, or you can just mmap the file and let the OS deal with the details of where it gets stored. Then, when somebody comes along and improves the logic in the OS, your 10-year-old-never-touched-since word processor works a little better, instead of being constrained by whatever logic the programmer had time to implement.
I'd think that the time of a word processor programmer would be better spent figuring out how to better handle background re-pagination or adding better text formatting logic, than designing their own memory management system. Should we go the route of WordPerfect back in the day and put the print drivers in the word processor as well?
When the AMD extensions became mainstream, the defacto standard, Intel licensed it. But the extension is still tied to the Intel architecture and Intel is still in control.
What does it even mean to be "in control?" AMD could apparently fork the last architecture/instruction-set/whatever-you-want-to-call-it (AIW for short), and create their own AIW which everybody adopted, just as Intel routinely adds new instructions to newer chips.
I don't question that Intel is usually in the lead with advancing the x86 world. However, it is fair to say that by introducing amd64/x86_64/whatever-you-want-to-call-it (AXW for short) AMD brought 64-bit to the masses. They certainly didn't invent the first 64-bit chip, but they created the first one that anybody was willing to use.
I suspect that today moving to ARM or some other architecture would be easier than it would have been 10 years ago. Back when the Athlon 64 came out virtually all workstations ran Windows, with all software being win32. On the server front Linux was popular, but even in that world there was a lot of non-portable software around. Using AXW made it a lot easier to migrate, and at least in the Linux world I'd like to think that people have learned to avoid reading a 32-bit record using an int data type. In the non-linux world things are much more heterogeneous than they were a decade ago as well. That means that changing instruction sets entirely is more practical than it used to be.
Yup, and the BS about them being first to 64-bit...maybe in the consumer sector, but Intel, IBM and DEC all had 64-bit chips before the Athlon was even designed let alone shipped.
They invented the architecture that you probably typed your post on. That was the point. Heck, on my linux distro it is still called amd64...
Agree. Fuel cells actually make a lot of sense in any kind of transportation situation. They're far more effective than batteries at storing power. They're just not all that practical yet. As far as I'm aware the only fuel cell that REALLY works reliably is the hydrogen-oxygen fuel cell.
True- and large cryogenic tanks and associated filling equipment are probably more realistic on a large aircraft then on public automobiles.
I wasn't even thinking about cryogenic storage. That might make sense for large aircraft. The energy per unit mass of hydrogen is pretty high I think. If kept at atmospheric pressure the storage tanks don't have to be heavy. You'd have constant boil-off, but that is only an issue if you leave aircraft parked with fuel in them, which isn't really the case for large aircraft. For a car having to burn all your fuel within 8 hours of fueling is a big problem, but aircraft fill up before just about every flight.
How is moving everything to the cloud working out for those users?
About as well as things work out for my relatives when they have a hard drive fail without backups. Well, except that the users of the cloud get their data back in a few hours 99% of the time.
I can't say that applications at my workplace which has a professionally-managed datacenter are any more reliable than the typical cloud service. You just don't hear about it on the news when things go down there.
Sure, you and I know how to backup our data, but we're the exception. And for my data which is fairly critical, where do I back it up? Well, that would be on Amazon S3. This way I know that if my house burns down the last backup is always only a few hours old, and not from the last time I grabbed my disks and moved them offsite. And where do I back up my data on Google Drive or Amazon or Dropbox? Well, that would be my PC. If Amazon and my house both have a major catastrophe at the same time I'll be screwed, but chances are in that case I'll have to live in my basement until the fallout cools down.
And since I shouldn't store the keys in the cloud, for every instance that uses some encrypted disk, I need to manually login then provide the key and passphrase.
Encrypting the storage only works if the processing is all done outside of the cloud. The instant you type your passphrase into the cloud server the cloud provider knows what it is if they want to know it. They aren't limited to reading data stored on disk.
If someone claiming to be a networking expert said he was an expert in TPC/IPX (in a context obviously meaning TCP/IP), you wouldn't take that as a good sign.
I know what TCP is, but I'm sure I've posted enough stuff online that you will find a case of me writing TPC to quote me on. Call me when the guy starts going on about "tubes."
Compliance costs money no matter how you do it - having the data in-house often just means that it is easier to pretend that you're doing things right. One of the advantages of outsourcing is that you can actually get contracts with the party managing your data. When you do things in-house you can't make a contract with the IT department, because it is the same legal entity as yourself. Big companies often fall victim to their own internal cost-cutting.
Outsourcing also gives you hard costs on activities. When the work is done in-house the costs are always fuzzy - do you include the cost of the janitor that cleans the toilet the sysadmins sit on? When you outsource you pay by the whatever and your costs come in on an invoice and everybody is accountable for exactly how much they're utilizing.
At work I'm working on outsourcing of non-data-storage. In theory it isn't anything we couldn't do more efficiently in-house. The problem is that due to company culture/etc it is inevitable that if we continue to do things in-house we'll continue to do them poorly, so by outsourcing we can let somebody manage the work who will actually do it properly. I'd rather pay a bit more and ensure that things are being done right than pay less and have to deal with the issues when it turns out that they aren't.
Rain.
Does rain actually wash off skid marks? Asphalt is dark for the most part, so you can't really see light rubber deposits, but if they are there then they would probably interfere with solar panels underneath.
Everybody gets new tires every few years. All that rubber has to go somewhere, and I suspect that quite a bit of it ends up being absorbed into/onto the streets.