Having followed the finest Slashdot tradition and only read TFA after posting, it appears that there was truth in my speculation. Fail0verflow, the group that found the keys, posted on twitter that "we only started looking at the ps3 after otheros was killed.". That means they did this in nine months.
I get the impression that the moderate openness of the PS3 at release was exactly what did preserve its uncracked status for so long. As soon as they locked out the 'Other OS' option, they pissed off the precise segment of the userbase who also have the skill to crack any subsequent security improvements.
Your initial post seemed to refer to all of Wikileaks' releases, but what you say above only applies to the diplomatic cables (a fairly off topic aside: why are we still using the early 20th century terminology?).
In my opinion the war documents were something that the taxpaying, voting public absolutely had the right to see as a record of the government's actions in their name, and they served to reveal a few outright lies told simply to save face. The release of private diplomatic communications is a far murkier issue, and although I still support Wikileaks in principle, I'm not sure how much I would wish to defend them on that decision.
It is about time people started taxing the people who buy Apple products.
Any real reason, or just being a grouchy bastard for the sake of it?
Seriously, I'm sure you can see that entirely arbitrary legislation like this just lines people up for problems in future. I don't care that much because I don't live in France, but I still think it's asinine to create a specific legal distinction between Windows and Android/iOS/Unix (but only if running on a tablet, not a netbook, apparently).
The same theory could actually work for a marketed product. Adverts (truthfully) saying that "In a study, 78% of participants felt better after taking our product" or some such, combined with the general population's underlying trust in the fact that "they wouldn't sell it if it didn't do something" could well be enough. As others mentioned, it seems to work for homoeopathy!
I was talking about total monthly transfer limits, not continuous bandwidth limits. I'm well aware that bandwidth is oversold to allow higher burst rates, and I fully support this - I was the one who made the GP post in favour of overselling.
I think that advertising unlimited transfer, unless you're damn sure that you can provide it to the proportion of customers who choose to use that full allowance, is misleading. That was the basis for my suggestion - work out the total monthly transfer capability of your uplink, divide by number of customers, and then multiply by a factor of two or three to account for the fact that some customers won't use their full allowance. That gives you a reasonable basis to say 300GB/month on your 10Mbps connection, or whatever, rather than saying "unlimited" and then using shady backhanded tactics to keep customers using a minimal fraction of their total theoretical allowance.
OK, last-mile fiber might be cheap enough to 'waste' by underutilising it (if the monopoly status was removed), but that just shifts the issue to the backbone connections (which is more what I was talking about anyway, although I admit that 'interfering with your neighbour's Skype call' isn't relevant in that case). Transcontinental or transoceanic fiber is, to my knowledge, still bloody expensive and thus an ISP's allocation is heavily oversold against end-user connections. Then again, I suppose that with increased last-mile competition, local caching at the ISP could be used to reduce backbone load somewhat.
Interesting post. I'm no expert on network design, so I'm quite happy to defer to your judgement on congestion. Out of interest, what kind of overselling ratios are reasonable? I've been told that UK ISPs operate anywhere between 30:1 and 100:1.
I quite agree with you that the telcos are just out for all they can squeeze from us, but do you really think per-GB charges are a bad thing in themselves, or is it more the potential for abuse that's the issue? I'm of the opinion that a published limit based simply on (total monthly backbone transfer capacity)/(number of users), perhaps with an extra multiplier to account for those users who don't consume their total allowance, would be more sensible and more user friendly than the current method of selling "unlimited" and complaining when the users go over 10% of that.
Overselling doesn't mean they won't give you what's in the contract, nor that they don't have the capacity to do so, it just means that they don't have the capacity to give everyone that. It's a gamble, but usually a worthwhile one. As the article I linked above put it:
Imagine we didn’t “oversell” at all. We still offer 20GB of disk space and 1TB of bandwidth on our $7.95/month plan because that’s what the competition has forced us to offer. 1TB of bandwidth is about an average of 3Mbs. 3Mbs for a month costs us about $90/month. The 20GB of disk space actually costs us about $200 (BELIEVE IT OR NOT!), because of the level of availability and backups we provide. So, we’d be losing about $200 up front and $82 / month on each and every customer!
And, all in the name of not “overselling”, our disk arrays would sit 98% empty and our network pipes 1% full! [...] But with us, you really CAN use all the stuff we’re offering. You won’t be disabled for it. You won’t have to wait. Your performance won’t suffer. It’s just a good thing for us there’s a difference between being able to use something and actually using it!
“But what if, let’s just say, everybody DID use it one day? Just WHAT IF? Then you’re screwed, eh?! Then the house of cards all comes crashing down around this charade of a pyramid scheme scam!!”
That’s true. I guess it’s a good thing we live in this universe, where we have the law of large numbers, and not in your universe where ANYTHING THAT CAN HAPPEN, DOES!!!
The prices are probably a bit dated, but the point stands.
That's only an issue if the limits are unreasonably small. Per-GB pricing doesn't need to mean that you only get a few GB, after all! An unlimited 10Mbps connection equates to about 3.25TB of transfer per month - if the 'base package' gave you only 5% of that transfer you'd still have 160GB/month to play with. Admittedly in the current US market I'd expect the ISPs to use transfer-limited pricing to screw you for every penny you're worth, but that's a problem stemming from lack of competition.
If transfer limits were being used as an actual form of traffic regulation rather than an attempt to squeeze out extra profits, here's what you might see for a fairly cheap connection (back of the envelope numbers, but it'll do for an approximation): 100Mbps of backbone bandwidth, sold to 500 customers as 20Mbps connections. The actual bandwidth is thus oversold 100:1 - not great, but acceptable given normal usage patterns. Call it $5000/month for the 100Mbps backbone and you've got a base cost of $10/customer/month for a total monthly transfer capability of 66GB/customer burstable up to 20Mbps. Since some customers won't even use that 66GB every month, you can probably get away with a bit of excess on your transfer limits and still stay within your means - not the 100:1 ratio you can manage on bandwidth, but maybe a 150-200GB/customer/month allowance without oversaturating your backbone.
In theory, sure. In practice, nah, not really. They don't answer to anyone because there is not a significant enough chunk of folk that give a damn anymore.
I think it's perhaps more the case that not enough people give a damn about each issue. A few of us might change our votes based on a politician's policy on net neutrality, but that one thing probably won't be enough to really damage their campaign. We might even still vote for them next time if we find the other candidate objectionable enough.
Direct democracy is probably too inefficient to work on a large scale, even with the internet to help, but giving it a try would be the only plausible way of overcoming this problem.
You seem to be conflating two related (but separate) issues; admittedly the fact that we seem to use 'bandwidth' interchangeably with 'monthly data transfer' doesn't help matters. Point is, paying by the megabyte would do nothing to ease congestion at peak time unless they also stopped overselling bandwidth - if everybody pays for 1GB of transfer and tries to use it simultaneously, your streaming video will still interfere with your neighbour's Skype call.
The overselling of bandwidth, however, is necessary to make remotely efficient use of the infrastructure. The vast majority of net use is burst-based, so it's a huge waste to pay for the full 10Mbps of capacity for every user - just look at the prices on a leased line if you want to see how much that'd actually cost. Even at peak time, nowhere near 100% of customers are using their lines at 100% capacity. Here's a fairly good post on the subject (admittedly biased, but accurate nonetheless) from Dreamhost.
The problem arises when that oversold bandwidth is coupled with so-called unlimited transfer. If there is no limit on your data transfer, you have the right to use your line at full bandwidth capacity 24/7 - if too many people start doing that (and if the bandwidth has been oversold too heavily) then problems start to arise.
Per-GB pricing on data transfer doesn't actually do anything to prevent congestion in itself - as I mentioned above, you'll still hit problems if you all try to use your 1GB allowance at the same time - but it helps to even out usage, and you can at least reach the point where everyone can theoretically use their entire transfer allowance (even with oversold bandwidth) over the course of the month.
Unsecured doesn't imply incompetent - there are people who happily leave a public WiFi connection to the net which is securely isolated from their internal network. Hell, there are businesses built on exactly that premise.
The article is rather sparse on details, but what interests me is that Ardolf didn't succeed in his "this'll get the dude in trouble" plan; what led the police to believe that the access point had been 'hacked'? What security was used, for that matter? Were there logs?
The guilty plea certainly makes it seem like this is a case where computer fraud was handled correctly by the system, and since the courts often seem to make the mistake that 'IP address == person' it'd be good to see how they went about distinguishing the actual criminal from the victim here.
I agree with you to an extent, and I'm sure there are some people who's only hope is a calculator, but (as I mentioned above) I'd say that withholding them in lower level courses does force the development of at least a basic level of mathematical competence in the middle ability range. People might be bad at fast and accurate calculation, but in my experience we're pretty good at ballpark figures and sanity checking - if anything those skills are more important as we rely more on computers; when there's a 'black box' doing the work for you, it's all the more valuable to be able to spot errors. The alternative is, as we have seen, customer service reps insisting that 0.002cents = $0.002 simply because the computer says so.
So surely one solves the problem by only allowing calculators in the more advanced courses? I know that's how it was done throughout my (relatively recent) education - no calculators in primary school, some exams that allowed them and some that didn't in high school, and pretty much no question about using them in university exams (with the caveat that most answers aren't numerical, and they won't do you any good if you can't handle the underlying mathematics on paper anyway).
I'm assuming it was done simply for price - the case with the light uses the hooks for power, so rather than having two entirely separate designs they just left out the light and painted over the conductive parts (and apparently didn't do a very good job of it...). Probably cheaper to mould 10,000 pairs of hooks out of metal and then paint over half than to make 5,000 out of metal and 5,000 out of plastic.
It's a problem with the case design, not the Kindle, as far as I can see.
Even if we accept that point of view (and I'm really not sure that I agree), the next step is to examine the reasoning she gave to the court: "I was overwhelmed with anxiety that I might be the target of a sex crime". That makes no fucking sense. None whatsoever. She thinks that if someone sees a picture of bra on a washing line (which they could've seen while walking down the street), they're going to find and assault her?
The only way that it makes any sense is in the context (as given by the article) of her mental illness: "The suit claims her existing obsessive-compulsive disorder was worsened by the anxiety brought on by the photo, as she feared that everything she was doing throughout the day was being secretly recorded.". Taking that into account, I do sympathise with her problems, but Google can't reasonably be held responsible for them.
You undermine your (quite valid) argument by letting your bias impair your accuracy. I quite agree with you that 'hate speech' laws are a terrible idea; even in the title you can see that their intent is to limit freedom of speech, but we both know that "saying anything negative about a minority group, regardless of the legitimacy of the criticism" is not actually within their mandate. Proponents of the legislation just need to point to criticisms levelled at a certain group and allowed by law, and your argument falls apart - this is a problem from my perspective because allowing the proponents to win the argument only reinforces their views.
It's been quite a while since I looked at the Freenet project, but perhaps stripping away most of the anonymity and obfuscation would actually make it more useful in day-to-day circumstances? Obviously a fork of that nature would be unusable for the original purpose, communicating within oppressive regimes and the like, but with the enhancements in speed and ease of use it could be a good compromise for general content distribution.
Basically, I'm thinking of a P2P web-server 'cloud' (as much as the term is overused, I think it's appropriate here). You upload a document, it's split into chunks and (redundantly) uploaded to everyone running the software, and some kind of key is added to an index (also redundantly distributed) for later retrieval of the file. Retaining SSL communication to avoid DPI would probably be sensible, and perhaps keeping the local cache of file fragments encrypted would help to skirt around liability issues (although ideally the simple fact that the files were incomplete and placed there automatically would be enough) would be a good idea, but the software wouldn't be that difficult to put together - it's very similar to current P2P offerings, with the key difference that you don't rely on single peers seeding entire files - achieving a decent userbase would be the issue.
They are a giant precisely because they are the force behind a fairly diverse range of sites, all of which are big names in their respective fields. You may not have heard the name 'Gawker Media', and I don't expect valleywag or Jezebel to come up on most Slashdotters' daily rotation, but Gizmodo gets linked here (either in stories or comments) fairly regularly.
I agree pretty much entirely, but that last bit "anti-conformity for the sake of anti-conformity gets old" doesn't seem especially applicable; sure, turning up to a bank job wearing a T-shirt that says "fuck capitalism" would be doing it for the sake of it, but wanting to wear jeans and a T-shirt because they're simple, comfortable and (if worn right) can look damn good isn't contrariness, it's a perfectly logical position to take. If nobody complains, illogical rules will never get changed!
I'd also add that while I fully understand that a business wants to control its customer-facing image, for better or worse, I reserve a special level of disdain for those who: enforce dress code regulations on non-customer-facing staff, enforce entirely arbitrary restrictions that do not affect the overall appearance of the staff, or (most frequently) both.
That much I understood; my immediate reaction was that if a 'nonexistent' $1000 can be transferred on in that manner, then surely it could easily be used to defraud the banks? If, rather than sending the money on to the scammer, the 'victim' withdrew the money in a cashiers cheque (or even a suitcase full of notes, I guess) and disappeared, surely they can keep the money as long as they can hide (or avoid extradition) and the bank is screwed?
Obviously not worth it for your $1000 example, but with a few million the payoff could be worth it to many, surely?
Slashdot Mirror
Having followed the finest Slashdot tradition and only read TFA after posting, it appears that there was truth in my speculation. Fail0verflow, the group that found the keys, posted on twitter that "we only started looking at the ps3 after otheros was killed.". That means they did this in nine months.
I get the impression that the moderate openness of the PS3 at release was exactly what did preserve its uncracked status for so long. As soon as they locked out the 'Other OS' option, they pissed off the precise segment of the userbase who also have the skill to crack any subsequent security improvements.
I thought that the Pentagon investigation of the war docs concluded that the release as performed by Wikileaks did not put anyone in danger?
Your initial post seemed to refer to all of Wikileaks' releases, but what you say above only applies to the diplomatic cables (a fairly off topic aside: why are we still using the early 20th century terminology?).
In my opinion the war documents were something that the taxpaying, voting public absolutely had the right to see as a record of the government's actions in their name, and they served to reveal a few outright lies told simply to save face. The release of private diplomatic communications is a far murkier issue, and although I still support Wikileaks in principle, I'm not sure how much I would wish to defend them on that decision.
It is about time people started taxing the people who buy Apple products.
Any real reason, or just being a grouchy bastard for the sake of it?
Seriously, I'm sure you can see that entirely arbitrary legislation like this just lines people up for problems in future. I don't care that much because I don't live in France, but I still think it's asinine to create a specific legal distinction between Windows and Android/iOS/Unix (but only if running on a tablet, not a netbook, apparently).
The same theory could actually work for a marketed product. Adverts (truthfully) saying that "In a study, 78% of participants felt better after taking our product" or some such, combined with the general population's underlying trust in the fact that "they wouldn't sell it if it didn't do something" could well be enough. As others mentioned, it seems to work for homoeopathy!
"They dilute a compound until they're actually giving somebody water..."
A minor distinction, perhaps, but one worth making. The majority of homoeopathic 'medicines' contain literally zero active ingredient.
I was talking about total monthly transfer limits, not continuous bandwidth limits. I'm well aware that bandwidth is oversold to allow higher burst rates, and I fully support this - I was the one who made the GP post in favour of overselling.
I think that advertising unlimited transfer, unless you're damn sure that you can provide it to the proportion of customers who choose to use that full allowance, is misleading. That was the basis for my suggestion - work out the total monthly transfer capability of your uplink, divide by number of customers, and then multiply by a factor of two or three to account for the fact that some customers won't use their full allowance. That gives you a reasonable basis to say 300GB/month on your 10Mbps connection, or whatever, rather than saying "unlimited" and then using shady backhanded tactics to keep customers using a minimal fraction of their total theoretical allowance.
OK, last-mile fiber might be cheap enough to 'waste' by underutilising it (if the monopoly status was removed), but that just shifts the issue to the backbone connections (which is more what I was talking about anyway, although I admit that 'interfering with your neighbour's Skype call' isn't relevant in that case). Transcontinental or transoceanic fiber is, to my knowledge, still bloody expensive and thus an ISP's allocation is heavily oversold against end-user connections. Then again, I suppose that with increased last-mile competition, local caching at the ISP could be used to reduce backbone load somewhat.
Interesting post. I'm no expert on network design, so I'm quite happy to defer to your judgement on congestion. Out of interest, what kind of overselling ratios are reasonable? I've been told that UK ISPs operate anywhere between 30:1 and 100:1.
I quite agree with you that the telcos are just out for all they can squeeze from us, but do you really think per-GB charges are a bad thing in themselves, or is it more the potential for abuse that's the issue? I'm of the opinion that a published limit based simply on (total monthly backbone transfer capacity)/(number of users), perhaps with an extra multiplier to account for those users who don't consume their total allowance, would be more sensible and more user friendly than the current method of selling "unlimited" and complaining when the users go over 10% of that.
Overselling doesn't mean they won't give you what's in the contract, nor that they don't have the capacity to do so, it just means that they don't have the capacity to give everyone that. It's a gamble, but usually a worthwhile one. As the article I linked above put it:
Imagine we didn’t “oversell” at all. We still offer 20GB of disk space and 1TB of bandwidth on our $7.95/month plan because that’s what the competition has forced us to offer. 1TB of bandwidth is about an average of 3Mbs. 3Mbs for a month costs us about $90/month. The 20GB of disk space actually costs us about $200 (BELIEVE IT OR NOT!), because of the level of availability and backups we provide. So, we’d be losing about $200 up front and $82 / month on each and every customer!
And, all in the name of not “overselling”, our disk arrays would sit 98% empty and our network pipes 1% full! [...] But with us, you really CAN use all the stuff we’re offering. You won’t be disabled for it. You won’t have to wait. Your performance won’t suffer. It’s just a good thing for us there’s a difference between being able to use something and actually using it!
“But what if, let’s just say, everybody DID use it one day? Just WHAT IF? Then you’re screwed, eh?! Then the house of cards all comes crashing down around this charade of a pyramid scheme scam!!”
That’s true. I guess it’s a good thing we live in this universe, where we have the law of large numbers, and not in your universe where ANYTHING THAT CAN HAPPEN, DOES!!!
The prices are probably a bit dated, but the point stands.
That's only an issue if the limits are unreasonably small. Per-GB pricing doesn't need to mean that you only get a few GB, after all! An unlimited 10Mbps connection equates to about 3.25TB of transfer per month - if the 'base package' gave you only 5% of that transfer you'd still have 160GB/month to play with. Admittedly in the current US market I'd expect the ISPs to use transfer-limited pricing to screw you for every penny you're worth, but that's a problem stemming from lack of competition.
If transfer limits were being used as an actual form of traffic regulation rather than an attempt to squeeze out extra profits, here's what you might see for a fairly cheap connection (back of the envelope numbers, but it'll do for an approximation): 100Mbps of backbone bandwidth, sold to 500 customers as 20Mbps connections. The actual bandwidth is thus oversold 100:1 - not great, but acceptable given normal usage patterns. Call it $5000/month for the 100Mbps backbone and you've got a base cost of $10/customer/month for a total monthly transfer capability of 66GB/customer burstable up to 20Mbps. Since some customers won't even use that 66GB every month, you can probably get away with a bit of excess on your transfer limits and still stay within your means - not the 100:1 ratio you can manage on bandwidth, but maybe a 150-200GB/customer/month allowance without oversaturating your backbone.
In theory, sure. In practice, nah, not really. They don't answer to anyone because there is not a significant enough chunk of folk that give a damn anymore.
I think it's perhaps more the case that not enough people give a damn about each issue. A few of us might change our votes based on a politician's policy on net neutrality, but that one thing probably won't be enough to really damage their campaign. We might even still vote for them next time if we find the other candidate objectionable enough.
Direct democracy is probably too inefficient to work on a large scale, even with the internet to help, but giving it a try would be the only plausible way of overcoming this problem.
You seem to be conflating two related (but separate) issues; admittedly the fact that we seem to use 'bandwidth' interchangeably with 'monthly data transfer' doesn't help matters. Point is, paying by the megabyte would do nothing to ease congestion at peak time unless they also stopped overselling bandwidth - if everybody pays for 1GB of transfer and tries to use it simultaneously, your streaming video will still interfere with your neighbour's Skype call.
The overselling of bandwidth, however, is necessary to make remotely efficient use of the infrastructure. The vast majority of net use is burst-based, so it's a huge waste to pay for the full 10Mbps of capacity for every user - just look at the prices on a leased line if you want to see how much that'd actually cost. Even at peak time, nowhere near 100% of customers are using their lines at 100% capacity. Here's a fairly good post on the subject (admittedly biased, but accurate nonetheless) from Dreamhost.
The problem arises when that oversold bandwidth is coupled with so-called unlimited transfer. If there is no limit on your data transfer, you have the right to use your line at full bandwidth capacity 24/7 - if too many people start doing that (and if the bandwidth has been oversold too heavily) then problems start to arise.
Per-GB pricing on data transfer doesn't actually do anything to prevent congestion in itself - as I mentioned above, you'll still hit problems if you all try to use your 1GB allowance at the same time - but it helps to even out usage, and you can at least reach the point where everyone can theoretically use their entire transfer allowance (even with oversold bandwidth) over the course of the month.
Unsecured doesn't imply incompetent - there are people who happily leave a public WiFi connection to the net which is securely isolated from their internal network. Hell, there are businesses built on exactly that premise.
The article is rather sparse on details, but what interests me is that Ardolf didn't succeed in his "this'll get the dude in trouble" plan; what led the police to believe that the access point had been 'hacked'? What security was used, for that matter? Were there logs?
The guilty plea certainly makes it seem like this is a case where computer fraud was handled correctly by the system, and since the courts often seem to make the mistake that 'IP address == person' it'd be good to see how they went about distinguishing the actual criminal from the victim here.
I agree with you to an extent, and I'm sure there are some people who's only hope is a calculator, but (as I mentioned above) I'd say that withholding them in lower level courses does force the development of at least a basic level of mathematical competence in the middle ability range. People might be bad at fast and accurate calculation, but in my experience we're pretty good at ballpark figures and sanity checking - if anything those skills are more important as we rely more on computers; when there's a 'black box' doing the work for you, it's all the more valuable to be able to spot errors. The alternative is, as we have seen, customer service reps insisting that 0.002cents = $0.002 simply because the computer says so.
So surely one solves the problem by only allowing calculators in the more advanced courses? I know that's how it was done throughout my (relatively recent) education - no calculators in primary school, some exams that allowed them and some that didn't in high school, and pretty much no question about using them in university exams (with the caveat that most answers aren't numerical, and they won't do you any good if you can't handle the underlying mathematics on paper anyway).
I'm assuming it was done simply for price - the case with the light uses the hooks for power, so rather than having two entirely separate designs they just left out the light and painted over the conductive parts (and apparently didn't do a very good job of it...). Probably cheaper to mould 10,000 pairs of hooks out of metal and then paint over half than to make 5,000 out of metal and 5,000 out of plastic.
It's a problem with the case design, not the Kindle, as far as I can see.
Even if we accept that point of view (and I'm really not sure that I agree), the next step is to examine the reasoning she gave to the court: "I was overwhelmed with anxiety that I might be the target of a sex crime". That makes no fucking sense. None whatsoever. She thinks that if someone sees a picture of bra on a washing line (which they could've seen while walking down the street), they're going to find and assault her?
The only way that it makes any sense is in the context (as given by the article) of her mental illness: "The suit claims her existing obsessive-compulsive disorder was worsened by the anxiety brought on by the photo, as she feared that everything she was doing throughout the day was being secretly recorded.". Taking that into account, I do sympathise with her problems, but Google can't reasonably be held responsible for them.
You undermine your (quite valid) argument by letting your bias impair your accuracy. I quite agree with you that 'hate speech' laws are a terrible idea; even in the title you can see that their intent is to limit freedom of speech, but we both know that "saying anything negative about a minority group, regardless of the legitimacy of the criticism" is not actually within their mandate. Proponents of the legislation just need to point to criticisms levelled at a certain group and allowed by law, and your argument falls apart - this is a problem from my perspective because allowing the proponents to win the argument only reinforces their views.
It's been quite a while since I looked at the Freenet project, but perhaps stripping away most of the anonymity and obfuscation would actually make it more useful in day-to-day circumstances? Obviously a fork of that nature would be unusable for the original purpose, communicating within oppressive regimes and the like, but with the enhancements in speed and ease of use it could be a good compromise for general content distribution.
Basically, I'm thinking of a P2P web-server 'cloud' (as much as the term is overused, I think it's appropriate here). You upload a document, it's split into chunks and (redundantly) uploaded to everyone running the software, and some kind of key is added to an index (also redundantly distributed) for later retrieval of the file. Retaining SSL communication to avoid DPI would probably be sensible, and perhaps keeping the local cache of file fragments encrypted would help to skirt around liability issues (although ideally the simple fact that the files were incomplete and placed there automatically would be enough) would be a good idea, but the software wouldn't be that difficult to put together - it's very similar to current P2P offerings, with the key difference that you don't rely on single peers seeding entire files - achieving a decent userbase would be the issue.
They are a giant precisely because they are the force behind a fairly diverse range of sites, all of which are big names in their respective fields. You may not have heard the name 'Gawker Media', and I don't expect valleywag or Jezebel to come up on most Slashdotters' daily rotation, but Gizmodo gets linked here (either in stories or comments) fairly regularly.
I agree pretty much entirely, but that last bit "anti-conformity for the sake of anti-conformity gets old" doesn't seem especially applicable; sure, turning up to a bank job wearing a T-shirt that says "fuck capitalism" would be doing it for the sake of it, but wanting to wear jeans and a T-shirt because they're simple, comfortable and (if worn right) can look damn good isn't contrariness, it's a perfectly logical position to take. If nobody complains, illogical rules will never get changed!
I'd also add that while I fully understand that a business wants to control its customer-facing image, for better or worse, I reserve a special level of disdain for those who: enforce dress code regulations on non-customer-facing staff, enforce entirely arbitrary restrictions that do not affect the overall appearance of the staff, or (most frequently) both.
That much I understood; my immediate reaction was that if a 'nonexistent' $1000 can be transferred on in that manner, then surely it could easily be used to defraud the banks? If, rather than sending the money on to the scammer, the 'victim' withdrew the money in a cashiers cheque (or even a suitcase full of notes, I guess) and disappeared, surely they can keep the money as long as they can hide (or avoid extradition) and the bank is screwed?
Obviously not worth it for your $1000 example, but with a few million the payoff could be worth it to many, surely?