I don't think it's really a peering issue, much as they want folks to think it is.
In normal peering operations, traffic flows both ways in approximately equal amounts. Netflix goes to residential accounts; residential accounts can't run servers. Given a lack of servers, anyone expecting a residential account individually or a collection of residential accounts in aggregate to have anything resembling equal upload and download is being stupid.
In (most) normal peering operations, traffic can be on the network whose endpoints are not on the network. Virtually all Netflix traffic on ATTs network has an endpoint in ATTs network - the subscriber.
In (most) normal peering operations, traffic can be initiated from either side. Netflix isn't going to send me video streams without me requesting it.
No, this isn't a typical peering situation. They shouldn't be trying to ding Netflix for sending more than they receive, they should be trying to ding me for requesting more data. But they don't want to do that because they want to keep pretending they charge less for data than the other guy. So instead they try to charge me _through_ Netflix (they know durn well that the only person Netflix can charge to pay this fee is me, so it's coming out of my wallet either way) in the hopes that I'll get mad at Netflix instead.
They should be charging the person who ordered the sofa (me), just like they charge me for everything else that comes down that road. The only reason that truck is there is me, so just go ahead and bill me. What they're trying to do is bill UPS and make UPS bill me, which will hopefully make me mad at UPS instead of making me mad(der) at how much they're overpricing the road.
I think the key is to limit the loan duration and add a hardship clause. With a 5 year loan, you're only stuck for at most 5 years (and how stuck you are is decreasing over time), and the more stuck they have you for those 5 years the more they're going to be paying you afterwards. (Unless they just keep you for 5 and then drop you, which seems a bit shortsighted of them.) But yeah, it's not a perfect solution.
I'll freely agree that too much regulation is a problem. But too little is also a problem in a non-free market, and telecom in general is almost as non-free as it gets (bettered only by electric, water, sewer, etc.) Since there's not enough competition to force a given broadband provider to not gouge their customers and partners, we (as a society) either have to use regulation or settle for getting gouged.
I say as a society because an individual does have the option (however unpalatable it is) to simply do without internet; "take it or leave it". But that's becoming less and less of a viable solution as more and more of our day to day interactions with each other, with companies, and with government move to the internet. If improvement only happens after substantial numbers quit and quitting is infeasible, then nothing will ever improve.
Of course, I believe it is possible to have an amount of regulation which is neither too much nor too little, though it's harder to maintain that balance as the bureaucratic empires grow and harden. If you don't feel that such a sweet spot is possible, then I can see how less regulation would be preferable to more. But in that case we need to drop this districted monopoly system too, so we can actually hope for some competitors.
Perhaps they need a contract stating "Training will cost you $x. We will loan you this amount and extract loan payments from your paycheck for 5 years. If you quit before the loan is paid, you owe us the balance immediately. If we fire you or lay you off, the loan is cancelled. Once the loan is paid off, your takehome pay will go up by the amount of the loan payment."
I don't see what you mean. Assuming revenue of 100 and 40% expenses (taxes, sales channel), 5% of gross is 5 bucks, but 5% of net is only 3 bucks. 2 bucks difference on the bottom line may not sound like a lot, but scale it up and it starts to matter more.
Since it seems I'm misinterpreting what you said, could you please clarify?
Of course they do. It's not their only data source, certainly, but why would they not make use of a publicly available database? They probably use the Yellow Pages (tm)(R)(c)(pat pending) too. It gives them a small shred of deniability, even.
This page has their three plans. One is free, but it assumes your backup locations are not Crashplan's systems (friends, family, other computers you control). That's probably the catch you're looking for; you have to come up with the backup location yourself. Being able to back up one system to CP's servers (unlimited quantity) is on the order of 5 a month; being able to back up 2-10 systems is on the order of 15.
You may also consider the encryption options to be a catch; as you note, the free plan has less encryption. On the other hand, the data is being stored somewhere you presumably trust. The plans that save to CP's systems have a couple of key management options, and in one they don't get the key (barring sneakiness, but if you're that worried about it you're likely going to rig it so what they're backing up is encrypted already).
Personally, I back up one system (my file server) to Crashplan, and most of my other ones to the file server. NOTE! This does not mean my other systems are getting backed up to Crashplan indirectly! If I want it backed up to Crashplan I have to copy it to the fileserver as a regular file. It just means that stuff on my laptop that I don't think is important enough to go offsite still gets automatically backed up to a different disk.
I've thought about going tape for local backup, but I never felt comfortable with it, because I have the impression that the lifespan of tape is poor. What's your experience in tape lifespan (which I suppose also involves how often/whether you rewrite instead of getting a new tape)?
(Currenly local backup for most of my stuff is "put it on the file server", where it's mirrored, which is of course not perfect. But the file server is also backed up offsite through CrashPlan.)
from what I've read, the goal of dumping this water was to keep any smelt from getting caught in the pumps that would have taken the water to the farming area. The same article/blogpost (I don't remember which) asserted that the loss of smelt in pumps was lower than the loss to federal monitoring practices, which at least _sounds_ like the population could probably stand a _little_ more peril.
I think that perhaps it's not as firmly equivalent as you imply; a concept in a book cannot be used in the same ways as a concept in human memory without being copied to human memory. At which point it's not the concept in the book getting used any more.
I agree with you. My statement was not about what they should regulate but what they are allowed to regulate, and currently they regulate all flights with commercial purpose.
Slashdot Mirror
I don't think it's really a peering issue, much as they want folks to think it is.
No, this isn't a typical peering situation. They shouldn't be trying to ding Netflix for sending more than they receive, they should be trying to ding me for requesting more data. But they don't want to do that because they want to keep pretending they charge less for data than the other guy. So instead they try to charge me _through_ Netflix (they know durn well that the only person Netflix can charge to pay this fee is me, so it's coming out of my wallet either way) in the hopes that I'll get mad at Netflix instead.
They should be charging the person who ordered the sofa (me), just like they charge me for everything else that comes down that road. The only reason that truck is there is me, so just go ahead and bill me. What they're trying to do is bill UPS and make UPS bill me, which will hopefully make me mad at UPS instead of making me mad(der) at how much they're overpricing the road.
There's that too. But the first step away from 'all regulation is bad' is realizing that 'no regulation' is not good :)
I think the key is to limit the loan duration and add a hardship clause. With a 5 year loan, you're only stuck for at most 5 years (and how stuck you are is decreasing over time), and the more stuck they have you for those 5 years the more they're going to be paying you afterwards. (Unless they just keep you for 5 and then drop you, which seems a bit shortsighted of them.) But yeah, it's not a perfect solution.
I'll freely agree that too much regulation is a problem. But too little is also a problem in a non-free market, and telecom in general is almost as non-free as it gets (bettered only by electric, water, sewer, etc.) Since there's not enough competition to force a given broadband provider to not gouge their customers and partners, we (as a society) either have to use regulation or settle for getting gouged.
I say as a society because an individual does have the option (however unpalatable it is) to simply do without internet; "take it or leave it". But that's becoming less and less of a viable solution as more and more of our day to day interactions with each other, with companies, and with government move to the internet. If improvement only happens after substantial numbers quit and quitting is infeasible, then nothing will ever improve.
Of course, I believe it is possible to have an amount of regulation which is neither too much nor too little, though it's harder to maintain that balance as the bureaucratic empires grow and harden. If you don't feel that such a sweet spot is possible, then I can see how less regulation would be preferable to more. But in that case we need to drop this districted monopoly system too, so we can actually hope for some competitors.
Perhaps they need a contract stating "Training will cost you $x. We will loan you this amount and extract loan payments from your paycheck for 5 years. If you quit before the loan is paid, you owe us the balance immediately. If we fire you or lay you off, the loan is cancelled. Once the loan is paid off, your takehome pay will go up by the amount of the loan payment."
It's not a total loss. I'm now substantially more likely to shell out for a Fluke next time I need a multimeter.
I don't see what you mean. Assuming revenue of 100 and 40% expenses (taxes, sales channel), 5% of gross is 5 bucks, but 5% of net is only 3 bucks. 2 bucks difference on the bottom line may not sound like a lot, but scale it up and it starts to matter more.
Since it seems I'm misinterpreting what you said, could you please clarify?
It's easier to distinguish between 0xfffffe and 0xffffff as integers than as colors. Unless your eyes are a crapload better than mine, anyway.
It's essentially a quote from RSA (from 2003); see the wikipedia entry.
which conveniently leads to my main worry about the book - how's he going to get image hovertext on paper?
Of course they do. It's not their only data source, certainly, but why would they not make use of a publicly available database? They probably use the Yellow Pages (tm)(R)(c)(pat pending) too. It gives them a small shred of deniability, even.
And even if you don't trust them to not grab that key anyway, nothing says the files you feed it have to be cleartext.
This page has their three plans. One is free, but it assumes your backup locations are not Crashplan's systems (friends, family, other computers you control). That's probably the catch you're looking for; you have to come up with the backup location yourself. Being able to back up one system to CP's servers (unlimited quantity) is on the order of 5 a month; being able to back up 2-10 systems is on the order of 15.
You may also consider the encryption options to be a catch; as you note, the free plan has less encryption. On the other hand, the data is being stored somewhere you presumably trust. The plans that save to CP's systems have a couple of key management options, and in one they don't get the key (barring sneakiness, but if you're that worried about it you're likely going to rig it so what they're backing up is encrypted already).
Personally, I back up one system (my file server) to Crashplan, and most of my other ones to the file server. NOTE! This does not mean my other systems are getting backed up to Crashplan indirectly! If I want it backed up to Crashplan I have to copy it to the fileserver as a regular file. It just means that stuff on my laptop that I don't think is important enough to go offsite still gets automatically backed up to a different disk.
I also use crashplan. Backblaze is also a flat-rate-for-whatever option.
This. She has no problem with the executive branch messing with the public, just with the legislative branch.
No worries. Get the master key, decrypt it, extract the parts you want. Isn't that how they handle blurays now?
I've thought about going tape for local backup, but I never felt comfortable with it, because I have the impression that the lifespan of tape is poor. What's your experience in tape lifespan (which I suppose also involves how often/whether you rewrite instead of getting a new tape)?
(Currenly local backup for most of my stuff is "put it on the file server", where it's mirrored, which is of course not perfect. But the file server is also backed up offsite through CrashPlan.)
sub-par? You mean it gets pricier than "grass fed"?
from what I've read, the goal of dumping this water was to keep any smelt from getting caught in the pumps that would have taken the water to the farming area. The same article/blogpost (I don't remember which) asserted that the loss of smelt in pumps was lower than the loss to federal monitoring practices, which at least _sounds_ like the population could probably stand a _little_ more peril.
I think that perhaps it's not as firmly equivalent as you imply; a concept in a book cannot be used in the same ways as a concept in human memory without being copied to human memory. At which point it's not the concept in the book getting used any more.
Just remember, dark matter is for explaining why very big things don't fly apart, and dark energy is for explaining why even bigger things do.
To another number! It's numbers all the way down!! :)
maybe. looks solid so far :)
I agree with you. My statement was not about what they should regulate but what they are allowed to regulate, and currently they regulate all flights with commercial purpose.