Sure, it's nice that the guy notice the scenes where they've photoshopped in a few extra Ewoks or fixed the gamma correction, but the article doesn't mention that they're still going with the revisionist title. Back when I first watched Revenge of the Jedi, it was still named Revenge of the Jedi...
Unless they're being prosecuted separately for the fraud, it seems like the fine of $7500 was totally inadequate - at minimum, they ought to be forced to pay back that money as well as any other penalties they pay. He'll probably only do 5 years if he behaves himself in the slammer, but there's no excuse for letting him keep his half of the $400K.
And of course, if the fraud was what it sounds like, those $39.95 make-money-fast kits tell their customers how to be spammers themselves, so they not only annoy millions of people when trolling for those 10,000 suckers, but those suckers will go on to annoy millions more people.
512MB RAM and only 128MB swap? Either that's because you don't have a big enough disk drive (which is odd, because by the time 512MB RAM was affordable, 40GB disk drives were also affordable, and you could afford to blow an extra $2 on a gig or two on swap if you needed it), or you don't have enough swap space configured for the applications you're running. Of course, it used to be that Unix systems recommended having more swap space than RAM (originally 2-3x more), but even though that's no longer necessary, you still should have enough for what you're doing.
Meanwhile, if you're actually using the RAM for active applications, then get enough RAM. If you're on an old PC, that's a separate issue - that 486 or pentium133 motherboard might not have enough room to add more, or the price of old-style RAM may be high enough that it's cheaper to just get a new motherboard, so if you want to keep using the machine, you need to tune your distro appropriately, and maybe Slackware or Debian is the right answer, or maybe just running Blackbox or IceWM instead of the latest Gnome-bloatware is enough to do the job. (X itself doesn't need to be bloated - it ran just fine on a 386 with 8MB RAM:-)
You might be a redneck if.... Moonshiners occasionally used radiators to make cheap stills instead of doing the work of winding copper pipe. It was a really spectacularly bad idea, because they tended to have lead solder in them and other compounds that were really unwise to drink after they'd leached out into your distillate.
Not sure if there's any relationship to the safety of using this for your computer cooler, though. And a 1979 Toyota seems about right for recycling by now - we just got rid of our 1985, which was still running after ~190K miles, albeit pretty roughly.
Gak. Don't you k1dd13z remember your Dilbert history? The Pointy-Haired Boss was originally a devil also, but got watered down. (And that's not even counting his cousin Phil from Heck, the Prince of Inadequate Light, carrier of a big pitch-spoon...) In the PHB's case, that was probably a distinct improvement, but here it's just lame corporate blandness.
Disclaimer: I work for a different Phone Company than the one Scott Adams worked for back when Dilbert was mostly a Phone Company thing.
Probably too late to make the report
on
Press freedom
·
· Score: 3, Informative
That may affect next year's report, but was probably too late to make this year's.
Besides, Nobody's "responsible" for it. Everybody says it was somebody else, or that they're not allowed to talk about pending criminal investigations, or things like that.
At least under the last few years of US procedures for computer search and seizure rules, the Indymedia attacks were mismanaged - they're supposed to take a copy and return everything ASAP for most cases, and they're supposed to be extremely careful of systems containing journalistic works in progress, which Indymedia pretty obviously had. And they didn't handle it that way.
There's no reason that anybody should be expected not to do that, or castigated for doing it. Communications technology makes it relatively possible to work from anywhere in the world, and if a government says you can only work in their turf by paying them a large chunk of money, it's rational and moral to leave, and if that government's subjects have fewer jobs because of it, well maybe they'll take the hint after a while (or not.) The important question should be whether the companies are creating jobs for their subjects, whether directly by hiring them or indirectly by buying their products.
Now, that reasoning doesn't apply for a company that wants to get on the government dole selling them weapons or prisons or eavesdropping equipment or tax-collector data-processing services, especially if they'd be illegal if operated in the purchasing country, or things like that; those guys ought to put up with the bad effects of the Administration they're supporting.
You're not "cheating" on taxes if you arrange your affairs in ways that aren't taxable - even the Supreme Court has declared that over the years. You're only cheating if you're turning in dishonest returns. Enron and their ilk may have been cheating on their taxes, or may have only been cheating their investors - but somebody who moves their headquarters offshore because it's advantageous is no worse than someone who registers their company in Delaware or Nevada because it's advantageous.
David Brin's book "The Transparent Society" addresses the topic fairly well. The cost of computing is going to zero, and the cost of cameras have already pretty much gone to zero, and the cost of image processing is pretty much going to zero, and the cost of data tranmission, even wireless, is going to zero, and the cost of data storage is going to zero, so you can pretty much expect that any data that's even vaguely available to the public, and any events that happen in public view, are going to be kept and stored somewhere if anybody at all feels like it, and they'll probably be stored somewhere easily findable. Get used to it.
The real issue becomes whether the public can force the government to make its activities transparent also, or whether the government, who are the most important organization that might have the power to keep other people from watching it, will succeed at doing its surveillance in secret. It certainly has the resources to do it, and it has the motives to do it, and if the public lets it get away with it, they'll go gung-ho and do it.
Now think about the Bush Administration, or about Blunkett's Home Office if you're British, and the public's tolerance for secrecy in the name of "protecting us from terrorism", and for secret "no-fly lists", and for governments storing data at Caribbean companies that don't even have the US's minimal data privacy laws, and governments running secret prisons like Guantanemo Bay, and governments preventing media access to more normal prisons like Pelican Bay, and the rapid increase in classified information.
And then go read Brin's book again, and remember that it was written in 1999, before Bush got ~elected, and think how much Moore's Law has done since then and how terrorism is being used as an excuse for everything.
And if you're an American, next Tuesday you should go vote Early and Often, and if you're not in a swing state you don't have to vote for Kerry and Edwards (who are certainly no friends of privacy, but aren't rabid about killing it like Bush and Cheney are.)
[Number of students] x [Probability that a student pisses somebody off] x [Probability that pissed-off person can DDOS]
is a relatively high percentage, especially if lots of the students are gamers (who seem to score fairly high in both probabilities. That's not a flame, just an observation.)
A large ISP that I know fairly well (insert typical disclaimers here) tracks a lot of security problems on the net, and DDOS attacks against single targets seem to be fairly common, especially at universities, where the students have lots of bandwidth and computing resources and lots of unsecured PCs around to borrow. Another popular target is individual gamers (who are usually easier to overwhelm).
When did you first see a LAN in a car? A friend of mine had one at San Diego Usenix 1993. He was working on portable wireless data-transmission technology, and the alpha version of the box was in his trunk, with the beta version in a portable computer (too big to really call a laptop) in the front seat, and a thinwire ethernet neatly hidden away between them. We were able to telnet to Bell Labs research from there (didn't have a login on that machine, so we just tried "berferd"...
Here in Collie-Fornia, display screens like that in the front seat have been illegal since the beginning of the year. While much of the motivation was televisions and DVD players that can distract the driver, it's also illegal for the passenger in the front seat to be using a laptop.
I recently had a crown built and installed, and my dentist had a new milling machine in his office that built the tooth while I waited, which lets you get everything done in one visit. Unlike the previous process, where the dentist makes a cast, sends it off to a milling-machine lab, puts a temporary crown on your tooth, and a week or two later the crown is done and you go back and get it installed, the dentist scans your tooth into a computer, which tells the milling machine what to make. No need for RFID or other tracking, because the dentist is working on one of these at a time and knows which patient is getting a crown vs. which patient is getting their teeth cleaned or xrayed. (And the XRays are now hi-res digital too:-)
Being a geek, I of course had to go watch the machine. It starts with an oblong block of tooth material and has a couple of spinning files that grind on the tooth. However, after a minute or two of position-calibrating, it started spraying water and became impossible to see anything, so I went back to the chair and breathed nitrous until the tooth was ready to install.
In addition to being RFID-less, and single visit, it's able to make the crown in shapes that fit better to your tooth's surface, without the need for a post or the various other surface shapes that you used to need to get the crown to attach; my dentist said that the shaping the new technology uses goes against everything he learned in dental school, but he's done a few hundred of these by now and they work pretty well (as well as amortizing his machine cost.)
Being a geek doesn't help you win at poker if all your friends are geeks too. It just evens out the mathematical-skill part of the game, leaving you with the psychological behaviour-prediction parts of the game that you're not so good at. After all, the game's not about manipulating cards, it's about manipulating people.
On the other hand, back during the boom, the main instigator of our poker games also liked very good single-malts, so any money I lost was more than made up for by a cheerful evening with friends drinking his whisky.
My Friends would like you to join us for Poker
on
Geeks Playing Poker?
·
· Score: 1
As they say, "If you think poker is a game of luck rather than skill, my friends would love to have you come over and play poker with us."
Maybe 5 years ago, every sales person I knew got a Palm Pilot, and lots of techies also got them. That was partly because it was the Boom, and they were cool toys, but if you're in any sort of mobile profession, whether it's sales or consulting or whatever, it's far more convenient to have your calendar and contact list and notepad in a pocket-sized instant-on device. Laptops in the front seat of your car were really annoying even before California banned them. Universities aren't really the usage market for this sort of device - except for the first week of classes, you pretty much know where you're going and when, and if you're going to take notes on a computerized device, laptops are better than most PDAs - though some people might use a PDA and a desktop machine.
PDA/Cellphone combos are an obvious extension, at least if they've got decent screens and don't cost too much more per month than a regular cellphone, and mean you only have to carry one device in your pocket instead of two. Basic PDAs are ~$100 now, but I don't know if any of the cheap ones have MP3 players in them.
Spam is a social problem, but it's fundamentally a business. It's a social problem because the business depends on the availability of suckers and the willingness of spammers to exploit suckers and annoy non-suckers, but the spammers aren't doing it for fun, they're doing it to make money. The ROKSO list of the Top 200 spammers isn't filled with those small-time fools - it's mostly filled with people who are really making money. If you can cut off the money, you can cut off most of the spam; otherwise you can't.
Legislation is too easy to avoid, since the Internet lets you work from everywhere in the world - it can only work if it changes the fundamental economics, and they're hard to change. Technology is an arms race, and it also works for everybody if it changes the economics, but it can also work for _you_ if it cuts down the amount of spam that you actually see, as opposed to the amount of spam that spammers attempt to send to you. One big problem with technological solutions is that it's hard not to interfere with real email, and false positives are really annoying.
Bio. Spamford didn't start his life of evil selling spyware. He got his nickname from being one of the early big spammers, but he'd been evil before that. The reason you don't get inundated with junk faxes is that Spamford was also one of the early big junk faxers, and this annoyed enough people that Congress made a law against it. It hasn't gone away entirely, but it's at least a relatively well-defined problem, and the economics at the time were such that a law could make it relatively uneconomical.
Sorry.. At first glance, it looked like www.frikingalaska.com , which seemed to make sense given that you were talking about how cold it gets:-().
A friend of mine moved to North Pole AK a decade or so ago; I think she's now in Florida or somewhere else in the lower 49.
Ouch, yes, that was a typo
on
Replacing TCP?
·
· Score: 1
Thanks.
TCP's shortcomings, IPv6, and TCP for IPv6
on
Replacing TCP?
·
· Score: 1
TCP does have real shortcomings that need to be fixed.
The standard implementations have window sizes too small for environments with high bandwidth*delay products
but there are various hacks and workarounds people have done for this
There's also a need to play in the IPv6 world, including all the different layers like DNS.
IPv6 does provide bigger addresses, though classless addressing, firewalls, and NAT have given us an extra decade or two of slack. It's also supposed to give us mandatory built-in security, though IPSEC is a stopgap for that, and it's supposed to improve routing by providing hooks for things like geographical aggregation of IP addresses, though I don't think the real world has really deployed any implementations of that.
As you say, however, they don't provide much justification for this being better than TCP (and I couldn't find much besides the Slashdot article claiming they did), much less being a general-purpose replacement for it. In particular, it doesn't appear to do any congestion control, at least in the parts that are documented - it does have an option for varying transmission rate so that it can share bandwidth with other protocols rather than hogging the whole pipe, but there's no documentation indicating that it attempts to solve simple problems like transmission from a fat port to a skinny port, much less sharing bandwidth between large numbers of sessions or accommodating buffer congestion on the receiving device, which TCP has done lots of work on over the years.
There are some things it does that are interesting, and it may be an especially useful tool for multicasts where TCP isn't appropriate (there's been tons of work on reliable multicast protocols over the last decade or two, some of which is actually usefully implementable - this stuff may help, if integrated well, because you'd really like to minimize packet losses and retransmissions in a multicast environment, as well as not maintaining large numbers of 1:1 sessions.)
TFTP performance is just fine
on
Replacing TCP?
·
· Score: 1
TCP performance isn't blazingly fast, because as you say it's a really simple protocol. But that just means that you shouldn't use it for things it's not designed for - it's a really fine protocol for downloading operating system code to a chip that's running a minimal bootstrap environment and doesn't have anything better to do until it's booted, as long as you're running it in a relatively benign environment.
no, no, no. Crashing is not error handling, and crashing is not strict HTML handline, crashing is LACK of competent error handling. Rendering incorrect pages liberally _is_ better than crashing, though error handling would be better.
This isn't a general TCP Replacement
on
Replacing TCP?
·
· Score: 4, Insightful
As far as I can tell from the documentation, this isn't really a TCP replacement. It's basically a set of efficient forward error correction codes, so if you've got a relatively wide open pipe you can transmit on without much packet loss, it can blast away and recover from occasional losses, but it doesn't do any actual congestion control.
They have a "TCP-friendliness" option that varies the transmission rate in a way that TCP windowing can probably cooperate with, so you can set the rate knobs to something less than full blast,
but nothing they've documented appears to address the problem of multiple users of this application trying to use a transmission path at the same time, and
they also don't document anything that does path rate discovery - so it may work fine if you've got a couple of small pipes feeding a fat network, but if you've got a fat pipe on the sending end and a skinny pipe on the receiving end, they don't document anything that figures out what rate is safe to transmit at.
They also don't document when you would want to use this and when you would want to use TCP and when you would want to use this on top of TCP.
Of course regulation is a terrible idea, but it's while decentralization is usually better at creating a better variety of policies, in this case it's much better to have the Feds who have half a clue in charge rather than 50+ sets of state regulators who've got even less of a clue and who are totally in bed with the local telcos, and who'll make it difficult to build nationwide communications systems, much less global ones.
Most of the POTS providers are also trying to get into the VOIP game, because that's where the low-cost emerging technology is, so they've got mixed positions here. The biggest costs in typical VOIP-to-POTS environments are the customer premises equipment (VOIP routers or whatever) at the VOIP end and the local telco's per-minute price for delivering analog voice to houses at the destination. The telcos are often still charging 2 cents per minute for that, in spite of the amortized cost of the long-haul portion being 1/10 cent per minute, and unless something is done to change that regulatory structure, everybody's going to be building ugly workaround architectures.
The FCC's original job was partly to eliminate unregulated competition for the radio spectrum, ostensibly in the name of "protecting" the public's commons from interference, but in reality protecting the power of the early corporate interests who grabbed lots of spectrum, and regulating the content of speech you were allowed to broadcast. There were better alternatives - letting the market evolve formal or informal property rights, which happened in many places before government takeovers, and happened in Italy during much of the 80s and 90s when radio stations basically ignored the regulators but got along fine with each other. There were also worse alternatives - too many governments totally nationalized their airwaves, taking control away from the public and giving it to government propaganda stations. (Some of those produced some high quality material, like the BBC, but that was largely the exception.)
Additionally, they became the Federal regulators of the interstate aspects of the telephone monopolies, though those had already become largely state-regulated because the "regulated monopoly" tradeoff of exclusive power to offer a service in return for politically correct implementation and pricing is basically a geographical monopoly at the local scale.
Much of the New Deal really worked that way - trading off favors for regulation while telling the public that they were beating up the evil nasty monopolies.
Slashdot Mirror
Sure, it's nice that the guy notice the scenes where they've photoshopped in a few extra Ewoks or fixed the gamma correction, but the article doesn't mention that they're still going with the revisionist title. Back when I first watched Revenge of the Jedi, it was still named Revenge of the Jedi...
And of course, if the fraud was what it sounds like, those $39.95 make-money-fast kits tell their customers how to be spammers themselves, so they not only annoy millions of people when trolling for those 10,000 suckers, but those suckers will go on to annoy millions more people.
Meanwhile, if you're actually using the RAM for active applications, then get enough RAM. If you're on an old PC, that's a separate issue - that 486 or pentium133 motherboard might not have enough room to add more, or the price of old-style RAM may be high enough that it's cheaper to just get a new motherboard, so if you want to keep using the machine, you need to tune your distro appropriately, and maybe Slackware or Debian is the right answer, or maybe just running Blackbox or IceWM instead of the latest Gnome-bloatware is enough to do the job. (X itself doesn't need to be bloated - it ran just fine on a 386 with 8MB RAM :-)
Moonshiners occasionally used radiators to make cheap stills instead of doing the work of winding copper pipe. It was a really spectacularly bad idea, because they tended to have lead solder in them and other compounds that were really unwise to drink after they'd leached out into your distillate.
Not sure if there's any relationship to the safety of using this for your computer cooler, though. And a 1979 Toyota seems about right for recycling by now - we just got rid of our 1985, which was still running after ~190K miles, albeit pretty roughly.
Disclaimer: I work for a different Phone Company than the one Scott Adams worked for back when Dilbert was mostly a Phone Company thing.
Besides, Nobody's "responsible" for it. Everybody says it was somebody else, or that they're not allowed to talk about pending criminal investigations, or things like that.
At least under the last few years of US procedures for computer search and seizure rules, the Indymedia attacks were mismanaged - they're supposed to take a copy and return everything ASAP for most cases, and they're supposed to be extremely careful of systems containing journalistic works in progress, which Indymedia pretty obviously had. And they didn't handle it that way.
Now, that reasoning doesn't apply for a company that wants to get on the government dole selling them weapons or prisons or eavesdropping equipment or tax-collector data-processing services, especially if they'd be illegal if operated in the purchasing country, or things like that; those guys ought to put up with the bad effects of the Administration they're supporting.
You're not "cheating" on taxes if you arrange your affairs in ways that aren't taxable - even the Supreme Court has declared that over the years. You're only cheating if you're turning in dishonest returns. Enron and their ilk may have been cheating on their taxes, or may have only been cheating their investors - but somebody who moves their headquarters offshore because it's advantageous is no worse than someone who registers their company in Delaware or Nevada because it's advantageous.
The real issue becomes whether the public can force the government to make its activities transparent also, or whether the government, who are the most important organization that might have the power to keep other people from watching it, will succeed at doing its surveillance in secret. It certainly has the resources to do it, and it has the motives to do it, and if the public lets it get away with it, they'll go gung-ho and do it.
Now think about the Bush Administration, or about Blunkett's Home Office if you're British, and the public's tolerance for secrecy in the name of "protecting us from terrorism", and for secret "no-fly lists", and for governments storing data at Caribbean companies that don't even have the US's minimal data privacy laws, and governments running secret prisons like Guantanemo Bay, and governments preventing media access to more normal prisons like Pelican Bay, and the rapid increase in classified information.
And then go read Brin's book again, and remember that it was written in 1999, before Bush got ~elected, and think how much Moore's Law has done since then and how terrorism is being used as an excuse for everything.
And if you're an American, next Tuesday you should go vote Early and Often, and if you're not in a swing state you don't have to vote for Kerry and Edwards (who are certainly no friends of privacy, but aren't rabid about killing it like Bush and Cheney are.)
A large ISP that I know fairly well (insert typical disclaimers here) tracks a lot of security problems on the net, and DDOS attacks against single targets seem to be fairly common, especially at universities, where the students have lots of bandwidth and computing resources and lots of unsecured PCs around to borrow. Another popular target is individual gamers (who are usually easier to overwhelm).
When did you first see a LAN in a car? A friend of mine had one at San Diego Usenix 1993. He was working on portable wireless data-transmission technology, and the alpha version of the box was in his trunk, with the beta version in a portable computer (too big to really call a laptop) in the front seat, and a thinwire ethernet neatly hidden away between them. We were able to telnet to Bell Labs research from there (didn't have a login on that machine, so we just tried "berferd"...
Here in Collie-Fornia, display screens like that in the front seat have been illegal since the beginning of the year. While much of the motivation was televisions and DVD players that can distract the driver, it's also illegal for the passenger in the front seat to be using a laptop.
Being a geek, I of course had to go watch the machine. It starts with an oblong block of tooth material and has a couple of spinning files that grind on the tooth. However, after a minute or two of position-calibrating, it started spraying water and became impossible to see anything, so I went back to the chair and breathed nitrous until the tooth was ready to install.
In addition to being RFID-less, and single visit, it's able to make the crown in shapes that fit better to your tooth's surface, without the need for a post or the various other surface shapes that you used to need to get the crown to attach; my dentist said that the shaping the new technology uses goes against everything he learned in dental school, but he's done a few hundred of these by now and they work pretty well (as well as amortizing his machine cost.)
On the other hand, back during the boom, the main instigator of our poker games also liked very good single-malts, so any money I lost was more than made up for by a cheerful evening with friends drinking his whisky.
As they say, "If you think poker is a game of luck rather than skill, my friends would love to have you come over and play poker with us."
PDA/Cellphone combos are an obvious extension, at least if they've got decent screens and don't cost too much more per month than a regular cellphone, and mean you only have to carry one device in your pocket instead of two. Basic PDAs are ~$100 now, but I don't know if any of the cheap ones have MP3 players in them.
Legislation is too easy to avoid, since the Internet lets you work from everywhere in the world - it can only work if it changes the fundamental economics, and they're hard to change. Technology is an arms race, and it also works for everybody if it changes the economics, but it can also work for _you_ if it cuts down the amount of spam that you actually see, as opposed to the amount of spam that spammers attempt to send to you. One big problem with technological solutions is that it's hard not to interfere with real email, and false positives are really annoying.
Bio. Spamford didn't start his life of evil selling spyware. He got his nickname from being one of the early big spammers, but he'd been evil before that. The reason you don't get inundated with junk faxes is that Spamford was also one of the early big junk faxers, and this annoyed enough people that Congress made a law against it. It hasn't gone away entirely, but it's at least a relatively well-defined problem, and the economics at the time were such that a law could make it relatively uneconomical.
A friend of mine moved to North Pole AK a decade or so ago; I think she's now in Florida or somewhere else in the lower 49.
Thanks.
IPv6 does provide bigger addresses, though classless addressing, firewalls, and NAT have given us an extra decade or two of slack. It's also supposed to give us mandatory built-in security, though IPSEC is a stopgap for that, and it's supposed to improve routing by providing hooks for things like geographical aggregation of IP addresses, though I don't think the real world has really deployed any implementations of that.
As you say, however, they don't provide much justification for this being better than TCP (and I couldn't find much besides the Slashdot article claiming they did), much less being a general-purpose replacement for it. In particular, it doesn't appear to do any congestion control, at least in the parts that are documented - it does have an option for varying transmission rate so that it can share bandwidth with other protocols rather than hogging the whole pipe, but there's no documentation indicating that it attempts to solve simple problems like transmission from a fat port to a skinny port, much less sharing bandwidth between large numbers of sessions or accommodating buffer congestion on the receiving device, which TCP has done lots of work on over the years.
There are some things it does that are interesting, and it may be an especially useful tool for multicasts where TCP isn't appropriate (there's been tons of work on reliable multicast protocols over the last decade or two, some of which is actually usefully implementable - this stuff may help, if integrated well, because you'd really like to minimize packet losses and retransmissions in a multicast environment, as well as not maintaining large numbers of 1:1 sessions.)
TCP performance isn't blazingly fast, because as you say it's a really simple protocol. But that just means that you shouldn't use it for things it's not designed for - it's a really fine protocol for downloading operating system code to a chip that's running a minimal bootstrap environment and doesn't have anything better to do until it's booted, as long as you're running it in a relatively benign environment.
no, no, no. Crashing is not error handling, and crashing is not strict HTML handline, crashing is LACK of competent error handling. Rendering incorrect pages liberally _is_ better than crashing, though error handling would be better.
- They have a "TCP-friendliness" option that varies the transmission rate in a way that TCP windowing can probably cooperate with, so you can set the rate knobs to something less than full blast,
- but nothing they've documented appears to address the problem of multiple users of this application trying to use a transmission path at the same time, and
- they also don't document anything that does path rate discovery - so it may work fine if you've got a couple of small pipes feeding a fat network, but if you've got a fat pipe on the sending end and a skinny pipe on the receiving end, they don't document anything that figures out what rate is safe to transmit at.
They also don't document when you would want to use this and when you would want to use TCP and when you would want to use this on top of TCP.Most of the POTS providers are also trying to get into the VOIP game, because that's where the low-cost emerging technology is, so they've got mixed positions here. The biggest costs in typical VOIP-to-POTS environments are the customer premises equipment (VOIP routers or whatever) at the VOIP end and the local telco's per-minute price for delivering analog voice to houses at the destination. The telcos are often still charging 2 cents per minute for that, in spite of the amortized cost of the long-haul portion being 1/10 cent per minute, and unless something is done to change that regulatory structure, everybody's going to be building ugly workaround architectures.
Additionally, they became the Federal regulators of the interstate aspects of the telephone monopolies, though those had already become largely state-regulated because the "regulated monopoly" tradeoff of exclusive power to offer a service in return for politically correct implementation and pricing is basically a geographical monopoly at the local scale.
Much of the New Deal really worked that way - trading off favors for regulation while telling the public that they were beating up the evil nasty monopolies.