Perhaps NAT played a part in not charging per computer by the ISPs but I would say they charged per public IP address which before NAT there would have to be a public IP address assigned to each computer. So you're effectively paying for each computer. We still have this today in reality. A single public IP address in almost all cases costs less than more than a single IP address for an otherwise identical service. Additionally I do not believe that a true per-computer charge separate from the IP addresses would have ever been a realistic or successful model even without NAT and would have eventually failed if ISPs tried to enforce it in any meaningful way. <joke> Would they have the Personal Computer Police come by and make a sweep of your business. I can just picture it, "This is the PCP, we know you're in there, please keep your hands off your keyboards. Don't you dare touch that 'any' key or we will be forced to take more aggressive action. Sir, there's a flood of people running out the back carrying PCs, I think they're trying to hide the number of computers they have." </joke> On Mon, 2011-02-14 at 13:58 -0600, Mark Katerberg wrote: > On 02/14/2011 01:33 PM, Justin Krejci wrote: > > > > I don't know if circumventing or quietly violating a company's EULA/AUP > > with NAT is a good argument on why NAT is good. Without knowing the > > details of the case referenced (a quick search on what Security Now > > turned up a GRC archives page and I did not find "NAT" anywhere in the > > shows list) I can only make conjecture. I would hate to have to pay for > > each computer I connect to my ISP and if a company did that I would find > > another who didn't. Vote with my feet. > > I don't think it violates anyone's EULA, it just made it impossible for > them to realistically have anything to gain with it. And as we've seen > by the near monopolistic control of these companies, it can often be > very difficult to vote with a dollar due to the massive startup costs > for these companies. > > If there were a reasonably priced ISP giving out static IPs, I'd be > there in a second, but due to the cartel-like control, I haven't found one. > > > > > Can you cite the show specifically or point out any relevant links or > > sources to the case you reference? > > > > Steve Gibson talks about it a little bit in Episode 285, but the > detailed description was from Episode 284 of Security Now. From the > transcript: > > LEO: Yeah, thinker. And this, he said, in these days the ISPs wanted > to charge you for each user in the house. And I do remember those days > where basically each one would have a static IP address, and you would > have to pay the full freight, or maybe a slightly discounted rate. He > said, we were starting to create routers at the time. Or it was his > suggestion to put NAT in. And he said, I knew, but I didn't tell > anyone, that this would effectively make it impossible for ISPs to > charge per user. They'd have to charge per household because of course > the NAT would hide all the additional users. He said, I never mentioned > that feature. I just said it'd be a good thing to put NAT in these routers. > > STEVE: And that's what connection sharing is, of course. > > LEO: Right. And so he says, you can thank me for the fact that you are > paying what you're paying for your Internet access. He had some other > very interesting things to say. We decided to invite him in for a full > hour interview for our triangulation show because he's a fascinating > guy. And what I like about him, yeah, he created history 40 years ago > with VisiCalc, but he's not sat on his laurels. And he's doing > something very interesting right now, really about taking back the > Internet, that I thought was fascinating. He said the Net Neutrality > conversation is misguided. That's not what we need. He said there's a > way to handle this. > > STEVE: The other thing, too, is that by putting NAT in all these > routers, not only were we preventing ISPs from charging per user, but we > did hugely slow down the depletion of the IPv4 address space. Because, > you know, we've got, I mean, I'm sure probably all of our listeners have > many different machines behind their single IP that's out there, their > public IP. And we always are talking about 192.168.x.x addresses, which > we're all sharing, but which are kept separate. > > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > tclug-list at mn-linux.org > http://mailman.mn-linux.org/mailman/listinfo/tclug-list