by doctorshady on 6/17/22, 1:50 AM with 229 comments
by roenxi on 6/17/22, 4:17 AM
> "A privatized internet will always amount to the rule of the many by the few"
Giving control to the government is rather explicitly giving control to the few - the government is busy responding to issues voters see as key (eg, inflation, war, environmental concerns, skirmishes between different political groups, etc). Unless a voter is willing (and stupid enough) to prioritise cheap internet access above the important issues then public internet is just asking for regulatory capture.
The government doesn't have the bandwidth to do what this quote implies they can. It will just result in an incompetent, politically connected few in control. We know there is a group who are willing to get involved in corrupting the regulations - it is the people currently profiting off poor internet regulation. How many people are going to use voting as a signal for their displeasure in corruption of a Board of the Internet?
by AlbertCory on 6/17/22, 3:59 PM
I actually kicked off a lengthy thread on the internet-history mailing list about this, and the conclusion was that by 1990, the war was already over and TCP had won.
Network operators said, in effect, "OK, we support OSI. But GOSIP doesn't say we actually have to run it on our network."
One guy from The Wollongong Group said that his company offered a package to assist in converting from TCP to OSI, since obviously everyone would have to do it. They found that in Europe, supposedly the hotbed of international standards, there was only demand for a package to convert the other way: OSI --> TCP.
Note that this is not an object lesson proving either that "government works" OR "government always screws things up."
The reason TCP worked and OSI didn't was that regular engineers and grad students and postdocs, not politicians and big telecoms, built it. One reason they had such success is that the Defense Department applied time pressure: "give us something that works now, not in the glorious future." So that's also the government -- just a different part of it.
by WalterBright on 6/17/22, 6:07 AM
Um, networks communicating with each other was commonplace in those days, they were called "gateways".
Also, there were many privately developed networks, like the bulletin board system, that were global.
The internet also runs on Ethernet - all those internet cables you've got laying around are Ethernet, developed by Xerox.
by bmmayer1 on 6/17/22, 3:46 AM
Ok, I'll bite.
The privatized internet has failed us...compared to what?
There's two prime examples of a publicly owned internet in the world today. One is run by the CCP and is sardonically known as the Great Firewall of China. The other is the state-run intranet of North Korea.
Do we really think that these are desirable alternatives to a privately run and distributed internet where unfortunately some people say and do crazy things, and some internet providers happen to have geographical monopolies that really have nothing to do with the internet in the first place but local protectionism?
Come on.
by Gentil on 6/17/22, 6:24 AM
Remember the Internet Society planning to sell .org? - https://www.theregister.com/2019/11/20/org_registry_sale_sha.... ICANN and Internet Society has been under the spotlight for shitty behavior since a very long time.
W3C is also heavily lobbied by companies like Google for instance is abusing it's browser market share against to make the internet less private. Then there is Manifest v3 - https://docs.google.com/document/d/1nPu6Wy4LWR66EFLeYInl3Nzz....
This article is about things like that. The infrastructure privatization and lobbying. Not individual business privatization.
While I didn't like the author's website's patterns, that shouldn't change the fact about what the author is trying to convey.
by mathlover2 on 6/17/22, 5:01 AM
I think there should be regulation of ISPs coupled with more community broadband efforts. I also like protocol-based social media. I also think there should be scrutiny of larger companies for anti-competitive behavior. But you can do these things, and other efforts, without banning private investment in the Internet entirely.
Also, this article's assertion that the Internet's golden age ended in 1995 is rather laughable. I know Usenet pre-Eternal September was nice, but I'd hardly call it the "golden age" of the internet, especially since it was barely accessible to the public at that time.
by ketzu on 6/17/22, 4:32 AM
* Infrastructure
* Services
On the infrastructure side it proposes community owned access to the internet as an alternative to few, often a single, commercial provider for access. These partially already exist (as discussed in the article) but not as widespread as people would want and, worse, got hampered by local laws forbidding these local organizations in many US cities (as far as I remember at least).
This seems relatively straight forward and with removal of those laws this could be practices fairly easily for the last-mile problem. For the backbone this is not a solution though. If that was to be "community owned" it would be equivalent to the government running it on the national scale and unclear on how to run the international connections and who shouold own and operate them. As far as I can tell, the privat sector doesn't do a too shabby job at running those (besides major security problems with BGP...).
On the services side, the article seems to take a "protocols over platforms" stance, that seems to be popular on HN, too. It leaves open some questions for me, though: What if people do not switch? How to handle existing platforms with severe vendor lock-in or network effects?
I think those open questions are not easy or straight forward to answer and that's why the article only goes for the anyways-popular approaches. Although, I don't think the north-korea and china examples are good "deprivatized internet" examples. I have no idea of the north korean internet, but for chinas great firewall my imperssion is it is an enforcement tool for locality and local laws, the internet itself is as privatized as it is outside of it - centralized platforms, some p2p things, ads, tracking and run by mostly private companies.
Maybe something like minitel and other original competitors are more of an example for state managed internet. I am by no means educated well on minitel, but afair companies had to register with a central authority to host a service, which cuased severe friction and protectionism problems.
by motohagiography on 6/17/22, 3:51 AM
by dasil003 on 6/17/22, 5:46 AM
by teucris on 6/17/22, 4:10 AM
by kwatsonafter on 6/17/22, 9:09 PM
by tbrownaw on 6/17/22, 3:53 AM
Blaming it all on the presence of business being inherently corruptive of what would have otherwise been a utopia is also a bit sketchy.
by voz_ on 6/17/22, 9:00 PM
by voz_ on 6/17/22, 4:25 AM
by heydemo on 6/17/22, 4:25 AM
The problem here is that network effects and economies of scale generally make centralization more practical. Capitalism's competition between multiple centrally managed models, although not perfect, is tough to beat.
"Rather than lay out a concrete plan for a deprivatized internet, Tarnoff explains that experimentation will be key."
It's tough to clap your hands and summon a decentralized system, but Jacobin maintains typical faith something never tried will work well, as long as it's anti-capitalist.
Seems like regulation within capitalism is a much more dependable way of addressing flaws in the free market.
by helloworld11 on 6/17/22, 5:18 AM
edit: I mean, what would you expect of a magazine that named itself after a bunch of "heroes" of the French Revolution whose own extreme, fanatical leftism was a pioneer for mass murder of civilians in the name of ideological purity.
by osigurdson on 6/17/22, 4:01 AM
No, that's Canada
by 0daystock on 6/17/22, 2:05 PM
> Instead of waiting to see what Google or Amazon hand us, technology is produced by communities and collectives to serve very different needs and ends.
You can already do that - right now. But it's easier to write a whiny diatribe than it is to learn how to program though, isn't it?
by perryizgr8 on 6/17/22, 7:07 AM
by aanthracite on 6/17/22, 5:12 AM
Look at what's going on right now in Russia, North Korea or China (and these are only firsts that come to my mind). It's really hard for people there to get any information beyond the state propaganda.
Good times create dumb people -> dumb people make communism -> communism creates bad times -> bad times creates smart people -> smart people create capitalism -> capitalism creates good times -> good times create dumb people...
So, let us avoid the red color...
by g8oz on 6/17/22, 10:19 AM
Ouch
by sovietcattle on 6/17/22, 1:07 PM
by Paris Marx
by walrus01 on 6/17/22, 8:59 PM
The economics textbook version of the term "regulatory capture" is what has failed us in the large telecom and large ISP industry.
The private internet has failed us? No shit, maybe we shouldn't allow entities like the combined Centurylink/Level3 to acquire various mid sized players and reduce the market competition. Maybe we shouldn't allow Rogers and Shaw to merge in Canada. Things like that.
Maybe when the US federal government hands out subsidy money to companies like Frontier and Verizon to build suburban and rural FTTH they should be held accountable when they just take the money and don't actually build the service promised.
Maybe people in their ordinary homes in ordinary neighborhoods should have better options than degraded DSL from the local "phone" company on 30 year old copper POTS lines or the near monopoly local Comcast DOCSIS3 coax cable service, squeezing every last dollar of ROI out of that legacy coax plant.
bias/point of view: I do network engineering for a small/mid-size ISP that directly competes with the telecom dinosaurs.
by istillwritecode on 6/17/22, 4:14 AM
by kristopolous on 6/17/22, 10:09 PM
Anyone could just fraudulently send email as generalSmith@dod.mil in 1975 and you'd have no way of knowing if it was real.
And then it would traverse in a nondeterministic unencrypted way over any machine that claims it can get it there with no way of knowing whether it succeeded or whether the message received was the message sent.
It was built by academics for tasks like remote timesharing and it ran mostly on minicomputers, not even mainframes. All the early nodes were at academic institutions. Exactly 0 were on military bases.
The project goals, people involved, sites it was installed at, technologies built, all the founders, Cerf, Kahn, Taylor, Roberts, Linkletter - zero military people - 100% academics. None of this suggests military purpose
Look at the abysmal security the network had. Do you think email, rcp, ftp and telnet was designed for military use?
It was openly bridged to the Soviet research network through IIASA, you know, cause that's how cold war things happened - open door policy to the enemy
Or what about the routing protocols where a rogue network switch could just announce itself and then start soliciting for traffic to pass through it.
In 1997, a misbehaving router singlehandedly took down the net https://en.m.wikipedia.org/wiki/AS_7007_incident any enemy could have easily done this.
Look at DNS host transfer up to about 2002 - you could just query for all records dumping your entire network topology, to just anyone - extremely valuable information for your enemies.
Look at finger and the original whois, an email and personnel lookup tool. You could use it to get people's schedule, all the people who work under them, what they're doing, how to contact them, where they last logged in at - do you know what I'd really like to have as your military enemy?
Heck let's cite Wikipedia as if reality matters:
"20th century WHOIS servers were highly permissive and would allow wild-card searches. A WHOIS query of a person's last name would yield all individuals with that name. A query with a given keyword returned all registered domains containing that keyword. A query for a given administrative contact returned all domains the administrator was associated with."
Sending out spies, espionage, sabotage, all unnecessary if you're enemy is using this technology. You could do it all from a terminal.
There's zero security in any of these. The doors are unlocked and swinging open with a giant honking welcome sign blinking.
Edit: Apparently reality is unpopular. I'm committed to reality far more than being popular. My politics are on the far left btw, that's why I demand such high standards from these people. They're supposedly playing for my team. But let me tell you, they don't seem to care either.
by Apocryphon on 6/17/22, 10:12 PM
> This is a useful lens for thinking about the evolution of the internet, and for understanding why the dot-coms didn’t succeed. The internet of the mid-to-late 1990s was under private ownership, but it had not yet been optimised for profit. It retained too much of its old shape as a system designed for researchers, and this shape wasn’t conducive to the new demands being placed on it. Formal subsumption had been achieved, in other words, but real subsumption remained elusive.
> Accomplishing the latter would involve technical, social and economic developments that made it possible to construct new kinds of systems. These systems are the digital equivalents of the modern factory farm. They represent the long-sought solution to the problem that consumed and ultimately defeated the dot-com entrepreneurs: how to push privatisation up the stack. And eBay offered the first glimpse of what that solution looked like.
by gyre007 on 6/17/22, 10:15 PM
by Wolfenstein98k on 6/17/22, 8:39 AM
by briga on 6/17/22, 4:49 AM
I guess my point is, if there was every anything that has done more to liberate the average human being, it is the internet. I just don't buy the argument that any of this would have happened without free market competition, or that the state of the internet not is somehow worse than it would be otherwise.
by rank0 on 6/17/22, 9:32 PM
Or I guess we’ll just tack on another Trillion dollars to our annual deficit…infinite government expansion/spending can solve every problem right?! /s