Easier? sure. Just throw a bunch of people with root access to your servers and datastores at the problem. Of course you'll spend huge sums of money and expose yourself to some other risks along the way...
I was trying to avoid some typing with the DOD reference, but that obviously didn't work, so let's try a couple of examples. The basic point is that easy-to-access and easy-to-delete are not same thing; data is just too easy to copy.
Example 1: I send you an email with a pdf attachment. You read it, and then we decide to delete all copies of the pdf. Where do we need to look?
Directory where pdf was created.
Any cache directories used by PDF tools.
Within my mail tool.
At my ISP.
At any ISP along the way.
At your ISP.
In your mail tool.
In a browser cache, if was viewed via webmail.
Recycle bin/trash can of any system.
Any backup made on any system while the pdf was there.
There are probably more...
Example 2 (a bit more like the system we're talking about):
User uploads document to server1.
User comes in the next day, and retrieves document via server2. The url told the system how to find it on server1, but a copy is loaded on server2 to support multiple retrievals.
User comes back a week later, deletes document via server1.
How do we know which servers to notify for deletion? Do we maintain a list somewhere? Do we tell all of our servers to delete it? What about backups? What do we do if the server is not available when we send the notification?
> And as far as deleting backups on redundant servers, it sounds like it could be done with a few lines of code.
You have obviously never done anything at this scale. Deleting all copies of information on a significant system is a very hard problem to solve. Demonstrating to an auditor that you've deleted everything makes it even harder.
There's actually an entire Defense Department specification/procedure that attempts to describe how to do it: http://www.google.com/?q=DOD+5015
Wow - an entire post of contradictions. Apparently money is not fungible, and neither are customers with money.
There's an argument to be made about a currency bubble, but you're not making it.
For my prediction, I'm going to predict that people will die in automobile accidents in the US tomorrow. About 90 people, in fact. Accurate? yes. Useful in preventing a specific accident? no. More useful than your prediction? probably.
Go ahead, you can have the last post in this thread now...
Hmmm... so by your logic, if government reduces interest rates, and this leads to investment in production, that's bad. But if people get capital from their facebook stock and invest it in production, that's good. The presence of customers is apparently immaterial.
Yep, simplistic is the right word to describe your model - as another poster noted, the model is 'government == bad'. It is interesting to see the back flips you have to go through to keep that one part going (i.e., things like making money non-fungible).
The only real 'trickle down' is in production, not in consumption.
Yeah - that's while all the.com companies are doing so well, they produced something. The fact that there were not any customers was just a side note (the preceding lines are sarcasm, in case that isn't obvious).
That's an insanely simplistic model you're spewing there. Reality is more of a balancing act, with investment meeting demand. At the moment, there is no shortage of investment capital, but high unemployment and other factors have really put a crimp on the demand side of the equation.
Henry Ford understood this balance - that's why he raised his worker's salaries. These days, we get a lot of pressure to keep capital gains and other taxes on the wealthy low, mostly from folks who do not understand this balance.
Don't know if it was "clearly an act of self defense", since the only account comes from the BART police, and they have a pretty poor record in past shootings. It's interesting that the video tape has not been released.
Slashdot Mirror
Easier? sure. Just throw a bunch of people with root access to your servers and datastores at the problem. Of course you'll spend huge sums of money and expose yourself to some other risks along the way ...
I was trying to avoid some typing with the DOD reference, but that obviously didn't work, so let's try a couple of examples. The basic point is that easy-to-access and easy-to-delete are not same thing; data is just too easy to copy.
Example 1: I send you an email with a pdf attachment. You read it, and then we decide to delete all copies of the pdf. Where do we need to look?
Example 2 (a bit more like the system we're talking about):
How do we know which servers to notify for deletion? Do we maintain a list somewhere? Do we tell all of our servers to delete it? What about backups? What do we do if the server is not available when we send the notification?
It is not easy ...
> And as far as deleting backups on redundant servers, it sounds like it could be done with a few lines of code.
You have obviously never done anything at this scale. Deleting all copies of information on a significant system is a very hard problem to solve. Demonstrating to an auditor that you've deleted everything makes it even harder.
There's actually an entire Defense Department specification/procedure that attempts to describe how to do it: http://www.google.com/?q=DOD+5015
Wow - an entire post of contradictions. Apparently money is not fungible, and neither are customers with money.
There's an argument to be made about a currency bubble, but you're not making it.
For my prediction, I'm going to predict that people will die in automobile accidents in the US tomorrow. About 90 people, in fact. Accurate? yes. Useful in preventing a specific accident? no. More useful than your prediction? probably.
Go ahead, you can have the last post in this thread now ...
Hmmm ... so by your logic, if government reduces interest rates, and this leads to investment in production, that's bad. But if people get capital from their facebook stock and invest it in production, that's good. The presence of customers is apparently immaterial.
Yep, simplistic is the right word to describe your model - as another poster noted, the model is 'government == bad'. It is interesting to see the back flips you have to go through to keep that one part going (i.e., things like making money non-fungible).
The only real 'trickle down' is in production, not in consumption.
Yeah - that's while all the .com companies are doing so well, they produced something. The fact that there were not any customers was just a side note (the preceding lines are sarcasm, in case that isn't obvious).
That's an insanely simplistic model you're spewing there. Reality is more of a balancing act, with investment meeting demand. At the moment, there is no shortage of investment capital, but high unemployment and other factors have really put a crimp on the demand side of the equation.
Henry Ford understood this balance - that's why he raised his worker's salaries. These days, we get a lot of pressure to keep capital gains and other taxes on the wealthy low, mostly from folks who do not understand this balance.
linkage: http://www.sfbg.com/bruce/2011/07/11/editorial-end-bart-coverup
No human is "absolutely, irrevocably correct." Therefore, I question whether you are human.