If you used the broken openssl to create the private key, then you should assume somewhere who cares enough will be able to guess the private key. Since generating your own CA and generating a certificate request using openssl implies with almost 100% certainty you created the private key using openssl as well, then you should regenerate all of these. I think your problem #3 will be affected, too, assuming it uses openssl.
Thus, this is a huge friggin deal. I'm not sure what's appalling: the fact such a change was implemented in the first place, or the fact that it took 2 years before they powers that be realised it was a problem and fixed it. This is definitely a huge blow to Debian's credibility.
On the plus side, I appreciate how honest and direct the advisory was - no flowery language to try to downplay the fuckup. It can be really hard to admit when you've done something stupid, especially something as colossally stupid and far-reaching as this. Still, W. T. F. I guess the important thing is that they (and others) learn from this mistake.
Actually, Cisco hardware has remained astronomically expensive because a lot of people (network admins) keep buying them regardless of the fact that there's much cheaper equipment out there that does the same job just as well.
I know, because I'm one of them -- and even I don't know why I keep getting Cisco hardware! I think there may be witchcraft involved.
Have you ever thought of... just NOT playing the game. As in, at all?
Sure, and sometimes I don't. But tell me: what difference does it make to the publisher/developer of the game if I choose to never ever play it, rather than to pirate it? The obvious answer is nothing, as they don't make a sale to me either way. That's not entirely accurate though, as a person who plays the game will at least be more aware of the company and their previous products. They might even find themselves pleasantly surprised by the quality of the game, as I was with Portal, and decide to buy it.
Obviously, this does require a lot of honesty from the pirater, as it's easy to say "well it was fairly good but not $90 good" after playing a game, particularly if it's a game like Portal that you're unlikely to play again.
How much innovation is stifled by illegal filesharing? I'm sure it is quite a lot.
I'm sure it's quite a lot less than the amount of innovation that is stifled by patents and even copyright laws, particularly within the various computer software industries. The problem with blaming piracy for everything is that they fail to account for the fact that there's many, many people who are willing to try something for the cost of downloading (i.e. virtually zero) who would otherwise not be interested.
People argue that patents and copyright, despite the costs to society, have a nett benefit. I believe this is likely true, although we probably don't have the optimal solution at present. But have you considered that some piracy is also of nett benefit to society? It certainly improves the quality of my life!
Most of the games I play are pirated, but I only buy a handful of those for the simple fact that many don't represent sufficient value for money. Sometimes they do, though. I recently bought The Orange Box after finally getting around to pirating and playing Portal. This is actually a great example of a game I not only wouldn't have considered buying based purely on word of mouth, but even if I'd played the demo, since it was only the last level that turned it from "an okay way to spend some time" into "holy freaking jesus this is the greatest thing ever". I had played a pirated Half Life many eons ago, and I tried Half-Life 2 when it first came out, but unlike the legions of Valve sycophants I find them to be pretty run-of-the-mill shooters and never bought any before. I didn't even bother finishing HL2. However now that I own them, I figured I'd give them another chance and have played through HL2, Episode 1 and Episode 2; so now I'm hooked on the story (particularly with the references to Aperture Science).
Okay, that was a bit rambling, but the point is I now have a Steam account and am interested in the continuation of HL2 purely because I pirated Portal. Certainly not everybody who pirates games ever buys the ones they really liked, but it's not a rare phenomena.
The fact of the matter is that people who choose to download software illegally cause prices to go up and/or quality to go down for those of us who purchase it legitimately.
I don't see how you can state this as "fact". Every piece of software, be it an application or a video game, is unique unto itself. This effectively creates a monopoly market. Sure, you can play other games instead, but if you want to play a particular game and piracy isn't an option, your only choice is to pay whatever the publisher has decided that game is "worth". This isn't a situation that naturally leads to lower prices or improved quality.
The increasing cost of games is more an indication of an industry caught in a feedback loop of unrealistic expectations, resulting in spiralling costs of development in an effort to try to meet those expectations. The games market is massively larger than it was 10 or 15 years ago, and there's massively more "competition" amongst producers. In a normal market, the retail cost should have plummeted.
Honestly, I think you have it a bit backwards. This might be how The Industry thinks, but the lessening of DRM suggests that they've actually realised it's not quite so simple as that.
You state it yourself: Assuming that you both want to play this game and don't want to deal with the DRM, would you pirate it?
The answer to that is clearly "yes", which means "I would have bought if it weren't for the DRM". Publishers are realising that not only do most copy protection schemes not hugely inconvenience pirates, but it actively inconveniences your paying customers.
if you pirate this game simply to spite the paid version which has DRM you're probably not doing the cause any help.
If you buy it regardless of the DRM, what incentive does that give the publishers to stop using it?
I think it's more accurate to say that this is the ONLY thing* you can do to help, but it only helps if you make sure they know that they are losing sales specifically because of the DRM measures. Mentioning it on forums is a good and semi-anonymous way to get the point across. If they're reading "yeah I love the game, the copy protection is annoying but it's worth the hassle" then they'll get the message that... their paying customers think it's worth the hassle, and they'll keep using it so long as they think it helps reduce piracy*. If they're constantly reading "I would've bought it, but the protection was too invasive" then their attitude toward it will change.
What it comes down to is that they make a list of pros and cons for and against their protection schemes. In the pros list, they have "might reduce piracy, for a little while". In the cons list they have "increases development and support costs, inconveniences users".
So, make sure they add "reduces sales" to the cons list, and it starts looking like a very poor return on investment.
* - since nobody knows how many people have pirated a game, not buying it is effectively the same as pirating it. The fact that any piracy figures are (by necessity) made up means that it gets the blame any time sales are lower than hoped.
Trackers only track what the client tells it about - they periodically tell it how much they've uploaded and downloaded. Obviously if you alter your client to report unbelievable statistics to the tracker people can and will notice, but if you keep it in the realm of believability you can cheat quite effectively. I've actually once been banned from a private tracker for uploading too much; apparently they weren't expecting anyone to upload at several megabytes per second, and it triggered an auto-ban on the assumption that the stats my client was sending were fake.
I think the big issue though is that the Scene (at least, as I understand it) is a bunch of very small, closed communities (where a "community" is probably a single FTP server and the people who have access to it), with tightly controlled conduits between them. While you see this is as being inefficient (and clearly it is, if you're just considering distribution of data), they see it as a feature. They specifically want a minimum of people to have access to multiple communities. So there is no "one central server" that everyone trusts, but multiple servers that a subset of the people involve trust.
Another problem is that BT is largely a "pull" system - you publish a torrent and people can go get it. Most of the Scene operates as a "push" system, with people uploading new content in order to get credit. This means you'd have to have a separate system in order to make the.torrent available to the community you want to upload it to, preferably in an automated manner.
Yet another problem is the fact that individuals are trying to get as much credit as they can, and sharing the upload burden is not an effective way to maximise your own credit. Since most people are going to be trying to do 1:1 transfers anyway, BitTorrent and similar protocols actually introduce inefficiencies in transfer - not to mention a heck of a lot of overhead and complication.
I think another difference between the BitTorrent communities and the "Scene" is that BT communities reward people for continuing to upload content someone else provided in the first place, whereas the Scene specifically only rewards people for contributing new content.
Seems to me they'd have already unrared it, so at that point, it makes much more sense to simply ditch the rar and make a torrent from the unrared file.
Well I can only speculate, but my guess is that a lot of initial seeders have obtained the release as a multi-rar on a remote system where they can't watch it; probably a shell on a server with fat pipes. Some may not actually have the ability to unrar on the remote system to at least verify it's intact. Or if they're using a system they're not authorized to use, they might feel that their bandwidth usage won't be noticed but if they start unrarring massive files someone's likely to get wise. Others may simply be trying to be the first to get the torrent up, and feel that spending a few minutes unrarring it may cost them that honour.
Regardless, my hypothesis is that when a new release first makes its way to P2P land, a small group of initial uploaders will have a big impact on the future of the torrent. If the multi-rar happens to be better seeded than the unrarred one early in the game, it's likely to continue to grow in popularity as most people will choose more active torrents in preference of less active ones.
Transferring the files from a well-connected remote server to an end-users' home connection takes time. Unpacking the archive takes time. Watching the video - even just skipping through it to check for obvious problems - takes time. During all that time, the multi-rar can already be being seeded, giving it a significant initial advantage.
I think the mob really hasn't said anything.
Okay, that's true enough. I guess it's more correct to say that the mob doesn't really care one way or the other, otherwise rarred torrents wouldn't be around.
Not particularly. Every decent BitTorrent client includes PeerGuardian-like functionality now. All you'd need to do is point its blacklist at some obfuscated https URL. Problem solved.
Except a blacklist is the opposite of what they'd want to use.
Remember, the Scene members tend to belong to well-known groups who are definitely of interest to law enforcement. Not only do they need to minimize exposure of the Scene as a whole to LE, but they also need to minimize the knowledge each individual has about any others. It's not reasonable to expect someone who gets caught not to cooperate in order to get a reduced sentence.
Also, distribution within the Scene is based on a credit system. It might be possible to retrofit such things into something like BT without having a centralised server, but it's an awful lot of work for little gain. The way they do things now works fine.
Right, I'm hijacking this troll to get eyeballs. Let's all post anecdotes about our funny/stupid bosses.
So we have this website, and a few Sundays ago the database that drives it fell over. It was around 1.30am. Around 11.30pm just before going to bed I decided to check my work email, just in case there was anything I needed to know for Monday. Turns out I needed to know our website was down all day. Of course, I have Nagios monitoring this, but since it was a weekend I didn't check my work email and never knew.
We happened to have a meeting the next day, so I mentioned that the site had been down for nearly 24 hours. Naturally nobody knew, because it was on a Sunday. So I said I wanted to get a GSM modem so I could receive SMS notification if important things went South, and after assuring the boss it'd only be a few hundred dollars he said okay.
Later a colleague (who was going to actually get quotes and buy it) told me he'd been asked to defer it. We had security auditors coming in soon, and the boss wanted to get their okay, because he thought it was a security risk. Bit strange I thought, but fair enough; we're hooking a wireless modem up to a server on our internal network, I can see how that can be perceived as a risk. Easy enough to explain how it wouldn't be possible to access the server using it.
But no, that wasn't the risk. The risk was we were using Nagios. It's open source!, he says. Doesn't that mean it's less secure?
Wow. Just wow. I'm disappointed I heard this 2nd hand, otherwise I would've been in there with a "1998 called; they want their FUD back". He might not have understood, but it would've amused me a great deal. And that's what matters, after all.
So it just goes to show there are still people buying the Microsoft-styled spin hook, line and sinker. I sure hope he doesn't discover the Brocade SAN switch we use runs Linux. Or our ESX servers. He might have a heart attack or something. Maybe it's okay if there's some proprietary code on it, though?
Bonus anecdote: my colleague also wanted to get pricing on sides for the racks in our server room, in order to improve airflow (they're completely open at the moment). The same boss said we don't need them, because hot air won't be coming out the back of the servers. It appears our boss feels that if cold air goes in the front, then cold air ought to come out the back. Sadly my colleague was too awed by this to press further, so we'll probably never know the full thought process behind this.
I actually look for some "group names" in the torrents I get - because they provide one file, not a RAR. In other words, provide what people want, and they will respect you for that. Make their life hard, and they will not care about your 1998 social customs. Like anything else in life.
Firstly, if you use torrents than nobody in the "Scene" gives a flying toss about whether you respect them or not. I have nothing to do with the Scene, and even I know that. They are not ripping things for us, they're ripping things for themselves. We're feeding from their scraps, if you like.
Once you understand that, all the other arguments become moot. Yes, multi-part RARs in torrents annoys me as well, but the people making them aren't doing it for us. Most (all?) Scene members would much prefer their releases never ever made it onto BT or USENET. Telling them that you disapprove of their distribution practices is, well, hilarious. Like a bank robber telling the cops he disapproves of their regular patrols of the street with all the banks on it. Actually, it's more like a bank robber in the US complaining about a pre-school teacher in Japan because he doesn't like the colour of the crayons they use. Thanks for the input, but who asked you, anyway?
So you're left trying to convince the people who do upload to more public services to unrar before they upload. More power to you, and I wish you luck. But I think the mob has largely spoken on this matter, and the mob says: "I don't give a crap if I have to unrar it first, so long as it's a) complete and b) a fast download". The torrents with multi-part archives tend to be seeded better than those which contain the extracted file, and therefore more people download the multi-part; which results in more seeds on it, resulting in more people downloading it...
As for using BT in the Scene -- it's up to them, it's their resources and they can do what they want with them -- so the following is purely mental masturbation. I would think BT would make it harder to keep "safe" and maybe easier to infiltrate. Password-protecting the servers (assuming most BT clients and trackers even support such) is probably insufficient; you'd likely want a local firewall to ensure only other Scene members can connect to your client. Keeping such a list updated in a secure manner would be somewhat tricky, I think, and telling everyone else the IP address of every other member sounds like a no-go.
I'm complaining about a lot of directed debugging that leads to nothing but more stalling. Even basic things like handling of times was broken when it was released so timer jobs wouldn't fire; even if content deployment actually worked reliably for us, that's a pretty significant problem. I'll also throw "completely useless error messages" into my list of complaints, while I'm ranting.
Microsoft chose to sell us -- to aggressively sell us -- their complicated product, at a pretty high price compared to other content management systems, with the promise that it does all these wonderful things that'll save us time and money and enhance our workflows, and so on. For that kind of price, I'd expect things to just fucking work, and if we do have problems then I'd expect a decent level of support to be provided, not bought as an extra.
So having bought the support as an extra, you're damned right I'm complaining about having to spend even more of my time debugging their software for them. For what we paid for it, I should be able to just set up a test farm with the same configuration and custom code as our production environment, give them RDP access, and let them debug the shit out of it. (Actually it would probably be better if we told them how we'd configured it, then they set up their own farm and tested our code in it.)
Commercial software houses are always spouting the "you get what you pay for" line. It's nice and pithy, but it means nothing if they don't deliver on it. And my experience with most vendors says they don't.
"Maybe the product is free, but you still have to put in the staff hours to basically support the product yourself."
Sorry to hijack your post, I didn't RTFA but this ticked me off.
They strongly imply that you don't have to support commercial software yourself. As an aside: the fact that companies can go around making these sorts of claims and not get sued into oblivion for blatantly lying to the public (or hell, thrown in jail) says some pretty bad things about the state of our societies, IMHO.
Unless you need someone to babysit you while you do basic tasks with the software, any reasonably competent tech is going to be able to install and use the software, regardless of whether it uses arcane text files or pretty tick boxes to configure it. If you have problems, you search the 'net for it -- chances are good someone else (or a lot of someones) have already encountered and solved the problem. The only times we've ever resorted to paid support from the vendor is when we have a really unusual problem we can't diagnose or fix ourselves; and guess what? The people providing the support are themselves simply reasonably competent (if you're lucky) techs who end up being just as stumped as you.
This means that for servers and infrastructure, paid support is a fucking joke. This bodes poorly for the Open Source companies that want to make money from providing support, but that's just how it is. Maybe it'll work out different when there's actually competition: in theory, since everyone can see the internals there's nothing stopping anyone else from becoming an expert at the software, and anyone can find and fix problems in the code. So possibly for popular software there'd be enough competition to provide paid support that they'd have to be competent and actually fix things, instead of fobbing off the customer until they give up. Maybe the current state of paid support is simply a symptom of monopoly inefficiency.
But I'm not certain about that. At its essence, paid support rewards good marketing of bad software. If the software does what you say it does, does it well, and is straightforward to set up, then there's not going to be any market for support.
<rant>
We use Sharepoint 2007 at work for our websites, and the licensing ain't cheap. Over $50,000 for each internet-facing server, another $20k for SQL 2005, and a bit of change for Windows licenses to run the servers and AD (plus additional licensing for the authoring environment). But that money's just a drop in the bucket compared to the money spent on developers to customize it and training of staff.
I'm pretty sure Microsoft knows this, and that's why they're not afraid to release a half-finished product whose key features (like content deployment) don't actually work. So we spend even more money on their Premier Support service, who proceed to waste my time over the course of several months collecting gigabytes of trace files, doing repetitive "tests" and sending them the error logs (which are, of course, incomplete; seems that part's a bit broken too) and then stalling me for a while until they come up with some other pointless exercise to waste some more of my time.
I'm positive they do this deliberately, because they know that eventually I'll get fed up with getting nowhere and resign myself to working around the defects. As I have, of course. But it's ridiculous that companies can charge you for the software, and then charge you again for (no) help with its problems, and then act like you're getting fantastic value for money.
If I'm getting free support from mailing lists or forums or what have you, then I'm happy to go through all the debugging shit -- installing minimal clean environments to see if the problem is reproducible there, etcetera. It annoys me having to spend my time doing this if I'm paying for support from someone else, though. Isn't that what they're being paid to do?!
Never forget that DRM means you are dependent on a company... as long as you want to be able to access that music, the company has to let you.
Another problem is that the economic system of the western world is based upon competition between different people/businesses offering essentially the same product, and thus having to compete with each other on quality and price.
As a natural result of this, it is expected that the majority of businesses will fail to be profitable and therefore go under. While unfortunate for the people involved in the business, for most people this is fine -- it simply means they weren't offering a compelling enough product compared to the competition.
For a consumer of DRM-encumbered media, this is a highly undesirable event. Therefore, consumers will actually desire to minimise competition -- new players offering better pricing or better quality of service are a lot less compelling, because if they fail to become profitable before their initial cash reserves run out, anything you've bought from them instantly becomes worthless.
As such, DRM is essentially anti-capitalist and will naturally lead to a monopoly, or at best oligopoly situation. It's no wonder Big Business is so desperate to get people to accept it. It's never been about protecting artists' rights, it's always been about creating lock-in.
I much rather risk getting a fine than voting for half wits that I myself look down on.
And you know what the fine for not voting in a federal election is? $20. That's it! Pay it at the post office, or even online. I'll never vote again! (Except in the state election so I can vote no in the DST referendum.)
Okay, I understand your reasoning, however I don't think it's quite right. They don't have infinite bandwidth, and while you're downloading you are actually consuming a portion of their limited bandwidth. When you stop downloading that portion becomes available for someone else to use. This makes it a bit different to regular resources like oil, but the fundamental principle remains the same: the more you use, the less there is for others to use. Therefore, in the interest of fairness, it makes sense to charge those who regularly use a higher portion of the available bandwidth more than those who regularly use less.
Ideally you'd be charged depending on how much bandwidth you're using while there is bandwidth contention. This could be done with a real-time display of the used capacity in their pipes, and if its at less than say 50% utilisation you pay nothing for the traffic, and above that you start paying a fee based on the portion you're using, at a rising rate until at 100% utilisation you're paying a mighty premium.
However such a system would be too convoluted for most people to understand, and also technically difficult to implement, so a simpler "amount of downloads per month" system is usually used instead. Not quite as easy for ISPs to manage as they end up with peak periods where they need to provision more bandwidth than is used at other times of the day, but it's a lot easier for consumers to grasp.
It's not economically feasible to provide sufficient bandwidth across the entire network for every end-user to use their full bandwidth 24/7 - users absolutely have to share the limited upstream bandwidth in order to keep costs at a reasonable level. You can get products with committed data rates which do let you use it at full capacity 24/7 without incurring additional charges, but these are very much more expensive than consumer internet access and most people would be unable to pay that; hence, it's not feasible for the consumer market.
Assuming you got a CDR service, it still costs them in terms of hardware and labour to add more capacity to their network, and they need to recover their costs from somewhere. Again the fair thing to do is to charge heavier users more -- since they're the reason additional bandwidth is needed.
Ultimately, you are actually paying for hardware maintenance, if you accept that "hardware maintenance" also includes all the other things required for the network to operate and to expand the capacity of it to meet demand.
Who's your provider, and why are you using them? That seems too slow to be cable or something, and if you're on 1.5Mbit ADSL you should have a much better choice of ISPs. Even Internode offer 55 GB/mo on 1.5Mbit at that price point, and they're not the cheapest around by a long shot. They only meter downstream too. Heck, for $105/mo you could get an 8 mbit service with 40 Gb per month (though if you're into online gaming you'd probably better staying on 1.5).
Mostly for reasons of geography and population distribution. For one thing, Moscow's population of 10 million is half of Australia's entire population. I don't know what the housing density in Moscow is like, but with a population like that I'm assuming that medium to high-density living is common, if not the norm. Running high-capacity backhaul links to higher-density dwellings is a lot more cost-effective than doing the same for the suburbs where most people here live. In addition, sheer numbers make "unlimited" plans more tenable, although still a bit of a risky game to play. With sufficient numbers of users even wildly varying download behaviour in individuals won't make much, if any, impact.
Secondly, Australia is a long way from anywhere, particular the US & Europe where the majority of data originates. It takes a lot of money to lay fibre optics across the Pacific, and there's essentially nothing along the way to soak up part of the cost. This is very different to Europe.
My ISP does offer unlimited ADSL plans, but only on the 1.5Mbit/256Kbit service, and that's at $299/month which is far more than most people are willing or able to pay for internet. It's also at a relatively low speed, and I'd be very reluctant to go from 20Mbit/2Mbit down to 1.5 even if it was unlimited. Honestly, even 6Mbit seems a bit slow to me now, though I could live with it for truly unlimited and performant traffic.
As a point of comparison, the datacentre hosting our servers charge us around $5,000 per month for 10mbit unmetered traffic. Virtually all of this cost is in the bandwidth itself. This is orders of magnitude different from hosting providers in the US and Europe, and there's good reasons for it.
Having a monopoly on the infrastructure won't solve the problem. In fact, that's pretty much the very definition of the problem. What incentive would this private company have to sell access to the fibre to other ISPs at a reasonable price?
We pretty much have this situation in Australia, as Telecom Australia (as a taxpayer funded government department) built and owns all the telephone exchanges around the country, backhaul capacity between them and the cities, and so forth. All the ISPs have to buy backhaul capacity from Telecom (now "Telstra", a non-government, listed company), beg them for access to exchanges in order to install their own DSLAMs, and so on.
Granted, it doesn't help that Telstra are also in the retail market directly competing with ISPs, but even if they weren't -- what incentive do they have to keep prices at commercially fair and competitive levels? They don't. There is no competition, so they can charge whatever the hell they want. If someone does try to compete, they can then lower prices in that small region, which means no other company is going to invest in infrastructure to compete with Telstra because they'll never make money on it.
Here, the government heavily regulates Telstra and generally does all the things that governments should never do to businesses. This is a mess and it is clearly not a long-term solution, because we (the taxpayers who paid for the infrastructure in the first place) are getting screwed anyway, and it's also incredibly unfair to Telstra.
I haven't seen the interstates partly because I don't drive, and also partly because I don't live in the US. (We do have interstate highways here, but we don't call them interstates.) I've also never been a toll road, as there aren't many of those in Australia, but they do seem like kind of a pain.
Anyway, your point is well taken and if you have a better proposal I'm all ears, but there's two problems I have with your argument.
First, just because the government has failed to properly maintain the roads doesn't automatically mean they would fail to maintain internet infrastructure. They very well may do, but it is a different situation where users (ISPs) can be charged for their use, making it easier to recover costs, and we have an example of something that doesn't work that we could hopefully learn from.
Second, your post makes the assumption that privately owned road networks would necessarily be better than what we have now. It's easy to look at what we have and laugh at it when there's nothing to compare it to, but would a country where every road is a toll road operated by upteen different operators be that much better than the current state of affairs? And what happens when the inevitable buyouts result in a handful of "big road" operating 99% of the roadways, no room to build any more (unless you bulldoze a crapload of homes and businesses), and no incentive to keep prices down or quality up?
Even if you have a lot of competition, this means massive amounts of money spent on parallel roads from competing companies, which is enormously wasteful. Add to that the fact that every operator would almost certainly charge for entry to (or more likely, exit from) their own network of roads, and you wind up with traffic snaking around the city trying to avoid hopping across different operators networks and thereby avoiding tolls, rather than taking the most direct route.
The most obvious analogy is the roads, and the majority of roads are paid for by the government (i.e. taxpayers). The roads provide the infrastructure needed to enable a lot of competition, for example taxis and courier services can all use these roads and compete on something other than "we own more roads than the other guys!".
This system works pretty well, and you always have the option to pay for your own private amazing road if you really want to, or if you need a road where nobody else needs one. All the things you mention like faults and maintenance or paid for by taxpayers, of course. So as you said:
I don't recall anyone ever saying "To have a free market, it must be provided by public Government services", a free market can never have any Government regulation or intervention, else it is not a free market.
However it's not necessarily in our interest to have a free market for the provision of roads; the services that can be provided using roads are for more valuable, and we can get better competition in those services by footing the bill for ourselves and giving everybody equal access.
Now, roads are paid for by everyone because a) they're incredibly important to the functioning of our society as a whole and b) it would be absolutely ridiculous for every major corporation to run its own (redundant) roads everywhere. So, do these apply to the internet?
It's certainly becoming more important, but if the internet disappeared overnight would civilization collapse? Probably not. You can argue the redundancy part of the question either way: redundancy of connectivity is good for reliability, but it's also pretty stupid for 5 different companies to run 5 separate bunches of fibre down a road, particularly since they're probably all going to be in the same physical location and therefore you're gaining very little in terms of redundancy.
I tend to think that we'd be better served by minimal, very (very very) high capacity links going everywhere, provided to retailers at essentially cost price, in order to greatly reduce the barrier to entry for competitors. Still, this sort of thing costs a lot of money and government departments aren't particularly well known for their efficiency.
the world sucks in terms of heavily weighting download packets vs uploads for no legitimate reason except "we can"
I think this is largely to do with the technology and consumer demand. Especially with analog transmissions (modem and DSL) you have a limited spectrum to play with. These have generally evolved in such a way that a particular part of this spectrum is used for transmission and another part for reception, to enable full-duplex operation. In theory your ADSL modem should be able to negotiate the size of each direction, but in practice I think the standards define particular configurations.
These have always tended to favour downstream bandwidth over upstream bandwidth, because that's what the majority of customers want. I think that still applies - people tend to consume more than they produce, but it applies less than it did before. It might now simply be a case that the market for download-centric services is so much larger that upload-centric (or balanced) services becomes something of a special case that needs "exceptional" handling, and therefore attracts higher prices to compensate.
If you move away from these sorts of connections, e.g. to fibre or co-located services, then I've never encountered anyone that charges you more to send data than to receive; the metering is always either "greatest of downloads or uploads" or "sum of downloads and uploads".
What do you mean the "pay for what you use model died"? If you sign up with an ISP in Australia, you choose a plan based on a) the line speed and b) the monthly download quota. If you want a 40 GB/month quota you'll pay more than you'd pay for a 20 GB/month quota at the same speed. Exceed the quota you paid for and your internet gets very, very slow.
In fact, the "pay for what you use" model is massively dominant in the marketplace; it's just been redesigned to be more consumer-friendly. People like having fixed, predictable monthly costs, and this structure lets people choose what fits their budget and needs. It also makes people less wary of downloading things, because they've already paid for their monthly quota and may as well make use of it.
It is a little odd that this scheme has become so accepted, because normally people balk at having to pay up front for a fixed amount of non-refundable, non-transferable usage.
It's a nice idea and might happen in the future - with widespread deployment of IPv6 and IPSEC it'd possibly be viable from a security perspective. There are two main blockers:
Firstly, wireless has a lot less bandwidth than wired networks. Each router would only be able to handle a small number of users, which means you'd need a very dense mesh so each router only handles a tiny fraction of the total traffic, with dynamic distribution over nodes to balance out the load. This has implications for latency/jitter-sensitive applications.
The second and really difficult to surmount problem is that of long links. There's always going to be a need for high bandwidth links over long, unpopulated distances (outside of cities, and of course inter-continental) and these are going to be very expensive to build and maintain. Someone has to pay for that.
Ultimately, internet access isn't all that expensive for the average end-user, and they get a much higher quality service than a constantly changing mesh of thousands of wireless routers could hope to provide.
Do you mean, common sense says that data transfer up to the capacity of your purchased link should be free? If so, how do you arrive at this conclusion?
Slashdot Mirror
If you used the broken openssl to create the private key, then you should assume somewhere who cares enough will be able to guess the private key. Since generating your own CA and generating a certificate request using openssl implies with almost 100% certainty you created the private key using openssl as well, then you should regenerate all of these. I think your problem #3 will be affected, too, assuming it uses openssl.
Thus, this is a huge friggin deal. I'm not sure what's appalling: the fact such a change was implemented in the first place, or the fact that it took 2 years before they powers that be realised it was a problem and fixed it. This is definitely a huge blow to Debian's credibility.
On the plus side, I appreciate how honest and direct the advisory was - no flowery language to try to downplay the fuckup. It can be really hard to admit when you've done something stupid, especially something as colossally stupid and far-reaching as this. Still, W. T. F. I guess the important thing is that they (and others) learn from this mistake.
Actually, Cisco hardware has remained astronomically expensive because a lot of people (network admins) keep buying them regardless of the fact that there's much cheaper equipment out there that does the same job just as well.
I know, because I'm one of them -- and even I don't know why I keep getting Cisco hardware! I think there may be witchcraft involved.
Sure, and sometimes I don't. But tell me: what difference does it make to the publisher/developer of the game if I choose to never ever play it, rather than to pirate it? The obvious answer is nothing, as they don't make a sale to me either way. That's not entirely accurate though, as a person who plays the game will at least be more aware of the company and their previous products. They might even find themselves pleasantly surprised by the quality of the game, as I was with Portal, and decide to buy it.
Obviously, this does require a lot of honesty from the pirater, as it's easy to say "well it was fairly good but not $90 good" after playing a game, particularly if it's a game like Portal that you're unlikely to play again.
I'm sure it's quite a lot less than the amount of innovation that is stifled by patents and even copyright laws, particularly within the various computer software industries. The problem with blaming piracy for everything is that they fail to account for the fact that there's many, many people who are willing to try something for the cost of downloading (i.e. virtually zero) who would otherwise not be interested.
People argue that patents and copyright, despite the costs to society, have a nett benefit. I believe this is likely true, although we probably don't have the optimal solution at present. But have you considered that some piracy is also of nett benefit to society? It certainly improves the quality of my life!
Most of the games I play are pirated, but I only buy a handful of those for the simple fact that many don't represent sufficient value for money. Sometimes they do, though. I recently bought The Orange Box after finally getting around to pirating and playing Portal. This is actually a great example of a game I not only wouldn't have considered buying based purely on word of mouth, but even if I'd played the demo, since it was only the last level that turned it from "an okay way to spend some time" into "holy freaking jesus this is the greatest thing ever". I had played a pirated Half Life many eons ago, and I tried Half-Life 2 when it first came out, but unlike the legions of Valve sycophants I find them to be pretty run-of-the-mill shooters and never bought any before. I didn't even bother finishing HL2. However now that I own them, I figured I'd give them another chance and have played through HL2, Episode 1 and Episode 2; so now I'm hooked on the story (particularly with the references to Aperture Science).
Okay, that was a bit rambling, but the point is I now have a Steam account and am interested in the continuation of HL2 purely because I pirated Portal. Certainly not everybody who pirates games ever buys the ones they really liked, but it's not a rare phenomena.
The fact of the matter is that people who choose to download software illegally cause prices to go up and/or quality to go down for those of us who purchase it legitimately.I don't see how you can state this as "fact". Every piece of software, be it an application or a video game, is unique unto itself. This effectively creates a monopoly market. Sure, you can play other games instead, but if you want to play a particular game and piracy isn't an option, your only choice is to pay whatever the publisher has decided that game is "worth". This isn't a situation that naturally leads to lower prices or improved quality.
The increasing cost of games is more an indication of an industry caught in a feedback loop of unrealistic expectations, resulting in spiralling costs of development in an effort to try to meet those expectations. The games market is massively larger than it was 10 or 15 years ago, and there's massively more "competition" amongst producers. In a normal market, the retail cost should have plummeted.
Honestly, I think you have it a bit backwards. This might be how The Industry thinks, but the lessening of DRM suggests that they've actually realised it's not quite so simple as that.
You state it yourself: Assuming that you both want to play this game and don't want to deal with the DRM, would you pirate it?
The answer to that is clearly "yes", which means "I would have bought if it weren't for the DRM". Publishers are realising that not only do most copy protection schemes not hugely inconvenience pirates, but it actively inconveniences your paying customers.
if you pirate this game simply to spite the paid version which has DRM you're probably not doing the cause any help.If you buy it regardless of the DRM, what incentive does that give the publishers to stop using it?
I think it's more accurate to say that this is the ONLY thing* you can do to help, but it only helps if you make sure they know that they are losing sales specifically because of the DRM measures. Mentioning it on forums is a good and semi-anonymous way to get the point across. If they're reading "yeah I love the game, the copy protection is annoying but it's worth the hassle" then they'll get the message that ... their paying customers think it's worth the hassle, and they'll keep using it so long as they think it helps reduce piracy*. If they're constantly reading "I would've bought it, but the protection was too invasive" then their attitude toward it will change.
What it comes down to is that they make a list of pros and cons for and against their protection schemes. In the pros list, they have "might reduce piracy, for a little while". In the cons list they have "increases development and support costs, inconveniences users".
So, make sure they add "reduces sales" to the cons list, and it starts looking like a very poor return on investment.
* - since nobody knows how many people have pirated a game, not buying it is effectively the same as pirating it. The fact that any piracy figures are (by necessity) made up means that it gets the blame any time sales are lower than hoped.
So let me get this straight: in this extreme police-state, I have a girlfriend? Your ideas intrigue me! How do I subscribe to your newsletter?
Trackers only track what the client tells it about - they periodically tell it how much they've uploaded and downloaded. Obviously if you alter your client to report unbelievable statistics to the tracker people can and will notice, but if you keep it in the realm of believability you can cheat quite effectively. I've actually once been banned from a private tracker for uploading too much; apparently they weren't expecting anyone to upload at several megabytes per second, and it triggered an auto-ban on the assumption that the stats my client was sending were fake.
I think the big issue though is that the Scene (at least, as I understand it) is a bunch of very small, closed communities (where a "community" is probably a single FTP server and the people who have access to it), with tightly controlled conduits between them. While you see this is as being inefficient (and clearly it is, if you're just considering distribution of data), they see it as a feature. They specifically want a minimum of people to have access to multiple communities. So there is no "one central server" that everyone trusts, but multiple servers that a subset of the people involve trust.
Another problem is that BT is largely a "pull" system - you publish a torrent and people can go get it. Most of the Scene operates as a "push" system, with people uploading new content in order to get credit. This means you'd have to have a separate system in order to make the .torrent available to the community you want to upload it to, preferably in an automated manner.
Yet another problem is the fact that individuals are trying to get as much credit as they can, and sharing the upload burden is not an effective way to maximise your own credit. Since most people are going to be trying to do 1:1 transfers anyway, BitTorrent and similar protocols actually introduce inefficiencies in transfer - not to mention a heck of a lot of overhead and complication.
I think another difference between the BitTorrent communities and the "Scene" is that BT communities reward people for continuing to upload content someone else provided in the first place, whereas the Scene specifically only rewards people for contributing new content.
Well I can only speculate, but my guess is that a lot of initial seeders have obtained the release as a multi-rar on a remote system where they can't watch it; probably a shell on a server with fat pipes. Some may not actually have the ability to unrar on the remote system to at least verify it's intact. Or if they're using a system they're not authorized to use, they might feel that their bandwidth usage won't be noticed but if they start unrarring massive files someone's likely to get wise. Others may simply be trying to be the first to get the torrent up, and feel that spending a few minutes unrarring it may cost them that honour.
Regardless, my hypothesis is that when a new release first makes its way to P2P land, a small group of initial uploaders will have a big impact on the future of the torrent. If the multi-rar happens to be better seeded than the unrarred one early in the game, it's likely to continue to grow in popularity as most people will choose more active torrents in preference of less active ones.
Transferring the files from a well-connected remote server to an end-users' home connection takes time. Unpacking the archive takes time. Watching the video - even just skipping through it to check for obvious problems - takes time. During all that time, the multi-rar can already be being seeded, giving it a significant initial advantage.
I think the mob really hasn't said anything.Okay, that's true enough. I guess it's more correct to say that the mob doesn't really care one way or the other, otherwise rarred torrents wouldn't be around.
Not particularly. Every decent BitTorrent client includes PeerGuardian-like functionality now. All you'd need to do is point its blacklist at some obfuscated https URL. Problem solved.Except a blacklist is the opposite of what they'd want to use.
Remember, the Scene members tend to belong to well-known groups who are definitely of interest to law enforcement. Not only do they need to minimize exposure of the Scene as a whole to LE, but they also need to minimize the knowledge each individual has about any others. It's not reasonable to expect someone who gets caught not to cooperate in order to get a reduced sentence.
Also, distribution within the Scene is based on a credit system. It might be possible to retrofit such things into something like BT without having a centralised server, but it's an awful lot of work for little gain. The way they do things now works fine.
Wikipedia has some info on the subject.
Right, I'm hijacking this troll to get eyeballs. Let's all post anecdotes about our funny/stupid bosses.
So we have this website, and a few Sundays ago the database that drives it fell over. It was around 1.30am. Around 11.30pm just before going to bed I decided to check my work email, just in case there was anything I needed to know for Monday. Turns out I needed to know our website was down all day. Of course, I have Nagios monitoring this, but since it was a weekend I didn't check my work email and never knew.
We happened to have a meeting the next day, so I mentioned that the site had been down for nearly 24 hours. Naturally nobody knew, because it was on a Sunday. So I said I wanted to get a GSM modem so I could receive SMS notification if important things went South, and after assuring the boss it'd only be a few hundred dollars he said okay.
Later a colleague (who was going to actually get quotes and buy it) told me he'd been asked to defer it. We had security auditors coming in soon, and the boss wanted to get their okay, because he thought it was a security risk. Bit strange I thought, but fair enough; we're hooking a wireless modem up to a server on our internal network, I can see how that can be perceived as a risk. Easy enough to explain how it wouldn't be possible to access the server using it.
But no, that wasn't the risk. The risk was we were using Nagios. It's open source!, he says. Doesn't that mean it's less secure?
Wow. Just wow. I'm disappointed I heard this 2nd hand, otherwise I would've been in there with a "1998 called; they want their FUD back". He might not have understood, but it would've amused me a great deal. And that's what matters, after all.
So it just goes to show there are still people buying the Microsoft-styled spin hook, line and sinker. I sure hope he doesn't discover the Brocade SAN switch we use runs Linux. Or our ESX servers. He might have a heart attack or something. Maybe it's okay if there's some proprietary code on it, though?
Bonus anecdote: my colleague also wanted to get pricing on sides for the racks in our server room, in order to improve airflow (they're completely open at the moment). The same boss said we don't need them, because hot air won't be coming out the back of the servers. It appears our boss feels that if cold air goes in the front, then cold air ought to come out the back. Sadly my colleague was too awed by this to press further, so we'll probably never know the full thought process behind this.
Firstly, if you use torrents than nobody in the "Scene" gives a flying toss about whether you respect them or not. I have nothing to do with the Scene, and even I know that. They are not ripping things for us, they're ripping things for themselves. We're feeding from their scraps, if you like.
Once you understand that, all the other arguments become moot. Yes, multi-part RARs in torrents annoys me as well, but the people making them aren't doing it for us. Most (all?) Scene members would much prefer their releases never ever made it onto BT or USENET. Telling them that you disapprove of their distribution practices is, well, hilarious. Like a bank robber telling the cops he disapproves of their regular patrols of the street with all the banks on it. Actually, it's more like a bank robber in the US complaining about a pre-school teacher in Japan because he doesn't like the colour of the crayons they use. Thanks for the input, but who asked you, anyway?
So you're left trying to convince the people who do upload to more public services to unrar before they upload. More power to you, and I wish you luck. But I think the mob has largely spoken on this matter, and the mob says: "I don't give a crap if I have to unrar it first, so long as it's a) complete and b) a fast download". The torrents with multi-part archives tend to be seeded better than those which contain the extracted file, and therefore more people download the multi-part; which results in more seeds on it, resulting in more people downloading it...
As for using BT in the Scene -- it's up to them, it's their resources and they can do what they want with them -- so the following is purely mental masturbation. I would think BT would make it harder to keep "safe" and maybe easier to infiltrate. Password-protecting the servers (assuming most BT clients and trackers even support such) is probably insufficient; you'd likely want a local firewall to ensure only other Scene members can connect to your client. Keeping such a list updated in a secure manner would be somewhat tricky, I think, and telling everyone else the IP address of every other member sounds like a no-go.
I'm complaining about a lot of directed debugging that leads to nothing but more stalling. Even basic things like handling of times was broken when it was released so timer jobs wouldn't fire; even if content deployment actually worked reliably for us, that's a pretty significant problem. I'll also throw "completely useless error messages" into my list of complaints, while I'm ranting.
Microsoft chose to sell us -- to aggressively sell us -- their complicated product, at a pretty high price compared to other content management systems, with the promise that it does all these wonderful things that'll save us time and money and enhance our workflows, and so on. For that kind of price, I'd expect things to just fucking work, and if we do have problems then I'd expect a decent level of support to be provided, not bought as an extra.
So having bought the support as an extra, you're damned right I'm complaining about having to spend even more of my time debugging their software for them. For what we paid for it, I should be able to just set up a test farm with the same configuration and custom code as our production environment, give them RDP access, and let them debug the shit out of it. (Actually it would probably be better if we told them how we'd configured it, then they set up their own farm and tested our code in it.)
Commercial software houses are always spouting the "you get what you pay for" line. It's nice and pithy, but it means nothing if they don't deliver on it. And my experience with most vendors says they don't.
Sorry to hijack your post, I didn't RTFA but this ticked me off.
They strongly imply that you don't have to support commercial software yourself. As an aside: the fact that companies can go around making these sorts of claims and not get sued into oblivion for blatantly lying to the public (or hell, thrown in jail) says some pretty bad things about the state of our societies, IMHO.
Unless you need someone to babysit you while you do basic tasks with the software, any reasonably competent tech is going to be able to install and use the software, regardless of whether it uses arcane text files or pretty tick boxes to configure it. If you have problems, you search the 'net for it -- chances are good someone else (or a lot of someones) have already encountered and solved the problem. The only times we've ever resorted to paid support from the vendor is when we have a really unusual problem we can't diagnose or fix ourselves; and guess what? The people providing the support are themselves simply reasonably competent (if you're lucky) techs who end up being just as stumped as you.
This means that for servers and infrastructure, paid support is a fucking joke. This bodes poorly for the Open Source companies that want to make money from providing support, but that's just how it is. Maybe it'll work out different when there's actually competition: in theory, since everyone can see the internals there's nothing stopping anyone else from becoming an expert at the software, and anyone can find and fix problems in the code. So possibly for popular software there'd be enough competition to provide paid support that they'd have to be competent and actually fix things, instead of fobbing off the customer until they give up. Maybe the current state of paid support is simply a symptom of monopoly inefficiency.
But I'm not certain about that. At its essence, paid support rewards good marketing of bad software. If the software does what you say it does, does it well, and is straightforward to set up, then there's not going to be any market for support.
<rant>
We use Sharepoint 2007 at work for our websites, and the licensing ain't cheap. Over $50,000 for each internet-facing server, another $20k for SQL 2005, and a bit of change for Windows licenses to run the servers and AD (plus additional licensing for the authoring environment). But that money's just a drop in the bucket compared to the money spent on developers to customize it and training of staff.
I'm pretty sure Microsoft knows this, and that's why they're not afraid to release a half-finished product whose key features (like content deployment) don't actually work. So we spend even more money on their Premier Support service, who proceed to waste my time over the course of several months collecting gigabytes of trace files, doing repetitive "tests" and sending them the error logs (which are, of course, incomplete; seems that part's a bit broken too) and then stalling me for a while until they come up with some other pointless exercise to waste some more of my time.
I'm positive they do this deliberately, because they know that eventually I'll get fed up with getting nowhere and resign myself to working around the defects. As I have, of course. But it's ridiculous that companies can charge you for the software, and then charge you again for (no) help with its problems, and then act like you're getting fantastic value for money.
If I'm getting free support from mailing lists or forums or what have you, then I'm happy to go through all the debugging shit -- installing minimal clean environments to see if the problem is reproducible there, etcetera. It annoys me having to spend my time doing this if I'm paying for support from someone else, though. Isn't that what they're being paid to do?!
</rant>
Right. So how many hertz does a 2 GHz CPU run at?
How many bits can a 100 MBit/sec network connection transfer each second?
How many bytes can be stored on a 1 GB memory stick?
Another problem is that the economic system of the western world is based upon competition between different people/businesses offering essentially the same product, and thus having to compete with each other on quality and price.
As a natural result of this, it is expected that the majority of businesses will fail to be profitable and therefore go under. While unfortunate for the people involved in the business, for most people this is fine -- it simply means they weren't offering a compelling enough product compared to the competition.
For a consumer of DRM-encumbered media, this is a highly undesirable event. Therefore, consumers will actually desire to minimise competition -- new players offering better pricing or better quality of service are a lot less compelling, because if they fail to become profitable before their initial cash reserves run out, anything you've bought from them instantly becomes worthless.
As such, DRM is essentially anti-capitalist and will naturally lead to a monopoly, or at best oligopoly situation. It's no wonder Big Business is so desperate to get people to accept it. It's never been about protecting artists' rights, it's always been about creating lock-in.
And you know what the fine for not voting in a federal election is? $20. That's it! Pay it at the post office, or even online. I'll never vote again! (Except in the state election so I can vote no in the DST referendum.)
Okay, I understand your reasoning, however I don't think it's quite right. They don't have infinite bandwidth, and while you're downloading you are actually consuming a portion of their limited bandwidth. When you stop downloading that portion becomes available for someone else to use. This makes it a bit different to regular resources like oil, but the fundamental principle remains the same: the more you use, the less there is for others to use. Therefore, in the interest of fairness, it makes sense to charge those who regularly use a higher portion of the available bandwidth more than those who regularly use less.
Ideally you'd be charged depending on how much bandwidth you're using while there is bandwidth contention. This could be done with a real-time display of the used capacity in their pipes, and if its at less than say 50% utilisation you pay nothing for the traffic, and above that you start paying a fee based on the portion you're using, at a rising rate until at 100% utilisation you're paying a mighty premium.
However such a system would be too convoluted for most people to understand, and also technically difficult to implement, so a simpler "amount of downloads per month" system is usually used instead. Not quite as easy for ISPs to manage as they end up with peak periods where they need to provision more bandwidth than is used at other times of the day, but it's a lot easier for consumers to grasp.
It's not economically feasible to provide sufficient bandwidth across the entire network for every end-user to use their full bandwidth 24/7 - users absolutely have to share the limited upstream bandwidth in order to keep costs at a reasonable level. You can get products with committed data rates which do let you use it at full capacity 24/7 without incurring additional charges, but these are very much more expensive than consumer internet access and most people would be unable to pay that; hence, it's not feasible for the consumer market.
Assuming you got a CDR service, it still costs them in terms of hardware and labour to add more capacity to their network, and they need to recover their costs from somewhere. Again the fair thing to do is to charge heavier users more -- since they're the reason additional bandwidth is needed.
Ultimately, you are actually paying for hardware maintenance, if you accept that "hardware maintenance" also includes all the other things required for the network to operate and to expand the capacity of it to meet demand.
Who's your provider, and why are you using them? That seems too slow to be cable or something, and if you're on 1.5Mbit ADSL you should have a much better choice of ISPs. Even Internode offer 55 GB/mo on 1.5Mbit at that price point, and they're not the cheapest around by a long shot. They only meter downstream too. Heck, for $105/mo you could get an 8 mbit service with 40 Gb per month (though if you're into online gaming you'd probably better staying on 1.5).
Mostly for reasons of geography and population distribution. For one thing, Moscow's population of 10 million is half of Australia's entire population. I don't know what the housing density in Moscow is like, but with a population like that I'm assuming that medium to high-density living is common, if not the norm. Running high-capacity backhaul links to higher-density dwellings is a lot more cost-effective than doing the same for the suburbs where most people here live. In addition, sheer numbers make "unlimited" plans more tenable, although still a bit of a risky game to play. With sufficient numbers of users even wildly varying download behaviour in individuals won't make much, if any, impact.
Secondly, Australia is a long way from anywhere, particular the US & Europe where the majority of data originates. It takes a lot of money to lay fibre optics across the Pacific, and there's essentially nothing along the way to soak up part of the cost. This is very different to Europe.
My ISP does offer unlimited ADSL plans, but only on the 1.5Mbit/256Kbit service, and that's at $299/month which is far more than most people are willing or able to pay for internet. It's also at a relatively low speed, and I'd be very reluctant to go from 20Mbit/2Mbit down to 1.5 even if it was unlimited. Honestly, even 6Mbit seems a bit slow to me now, though I could live with it for truly unlimited and performant traffic.
As a point of comparison, the datacentre hosting our servers charge us around $5,000 per month for 10mbit unmetered traffic. Virtually all of this cost is in the bandwidth itself. This is orders of magnitude different from hosting providers in the US and Europe, and there's good reasons for it.
Having a monopoly on the infrastructure won't solve the problem. In fact, that's pretty much the very definition of the problem. What incentive would this private company have to sell access to the fibre to other ISPs at a reasonable price?
We pretty much have this situation in Australia, as Telecom Australia (as a taxpayer funded government department) built and owns all the telephone exchanges around the country, backhaul capacity between them and the cities, and so forth. All the ISPs have to buy backhaul capacity from Telecom (now "Telstra", a non-government, listed company), beg them for access to exchanges in order to install their own DSLAMs, and so on.
Granted, it doesn't help that Telstra are also in the retail market directly competing with ISPs, but even if they weren't -- what incentive do they have to keep prices at commercially fair and competitive levels? They don't. There is no competition, so they can charge whatever the hell they want. If someone does try to compete, they can then lower prices in that small region, which means no other company is going to invest in infrastructure to compete with Telstra because they'll never make money on it.
Here, the government heavily regulates Telstra and generally does all the things that governments should never do to businesses. This is a mess and it is clearly not a long-term solution, because we (the taxpayers who paid for the infrastructure in the first place) are getting screwed anyway, and it's also incredibly unfair to Telstra.
I haven't seen the interstates partly because I don't drive, and also partly because I don't live in the US. (We do have interstate highways here, but we don't call them interstates.) I've also never been a toll road, as there aren't many of those in Australia, but they do seem like kind of a pain.
Anyway, your point is well taken and if you have a better proposal I'm all ears, but there's two problems I have with your argument.
First, just because the government has failed to properly maintain the roads doesn't automatically mean they would fail to maintain internet infrastructure. They very well may do, but it is a different situation where users (ISPs) can be charged for their use, making it easier to recover costs, and we have an example of something that doesn't work that we could hopefully learn from.
Second, your post makes the assumption that privately owned road networks would necessarily be better than what we have now. It's easy to look at what we have and laugh at it when there's nothing to compare it to, but would a country where every road is a toll road operated by upteen different operators be that much better than the current state of affairs? And what happens when the inevitable buyouts result in a handful of "big road" operating 99% of the roadways, no room to build any more (unless you bulldoze a crapload of homes and businesses), and no incentive to keep prices down or quality up?
Even if you have a lot of competition, this means massive amounts of money spent on parallel roads from competing companies, which is enormously wasteful. Add to that the fact that every operator would almost certainly charge for entry to (or more likely, exit from) their own network of roads, and you wind up with traffic snaking around the city trying to avoid hopping across different operators networks and thereby avoiding tolls, rather than taking the most direct route.
The most obvious analogy is the roads, and the majority of roads are paid for by the government (i.e. taxpayers). The roads provide the infrastructure needed to enable a lot of competition, for example taxis and courier services can all use these roads and compete on something other than "we own more roads than the other guys!".
This system works pretty well, and you always have the option to pay for your own private amazing road if you really want to, or if you need a road where nobody else needs one. All the things you mention like faults and maintenance or paid for by taxpayers, of course. So as you said:
I don't recall anyone ever saying "To have a free market, it must be provided by public Government services", a free market can never have any Government regulation or intervention, else it is not a free market.However it's not necessarily in our interest to have a free market for the provision of roads; the services that can be provided using roads are for more valuable, and we can get better competition in those services by footing the bill for ourselves and giving everybody equal access.
Now, roads are paid for by everyone because a) they're incredibly important to the functioning of our society as a whole and b) it would be absolutely ridiculous for every major corporation to run its own (redundant) roads everywhere. So, do these apply to the internet?
It's certainly becoming more important, but if the internet disappeared overnight would civilization collapse? Probably not. You can argue the redundancy part of the question either way: redundancy of connectivity is good for reliability, but it's also pretty stupid for 5 different companies to run 5 separate bunches of fibre down a road, particularly since they're probably all going to be in the same physical location and therefore you're gaining very little in terms of redundancy.
I tend to think that we'd be better served by minimal, very (very very) high capacity links going everywhere, provided to retailers at essentially cost price, in order to greatly reduce the barrier to entry for competitors. Still, this sort of thing costs a lot of money and government departments aren't particularly well known for their efficiency.
I think this is largely to do with the technology and consumer demand. Especially with analog transmissions (modem and DSL) you have a limited spectrum to play with. These have generally evolved in such a way that a particular part of this spectrum is used for transmission and another part for reception, to enable full-duplex operation. In theory your ADSL modem should be able to negotiate the size of each direction, but in practice I think the standards define particular configurations.
These have always tended to favour downstream bandwidth over upstream bandwidth, because that's what the majority of customers want. I think that still applies - people tend to consume more than they produce, but it applies less than it did before. It might now simply be a case that the market for download-centric services is so much larger that upload-centric (or balanced) services becomes something of a special case that needs "exceptional" handling, and therefore attracts higher prices to compensate.
If you move away from these sorts of connections, e.g. to fibre or co-located services, then I've never encountered anyone that charges you more to send data than to receive; the metering is always either "greatest of downloads or uploads" or "sum of downloads and uploads".
What do you mean the "pay for what you use model died"? If you sign up with an ISP in Australia, you choose a plan based on a) the line speed and b) the monthly download quota. If you want a 40 GB/month quota you'll pay more than you'd pay for a 20 GB/month quota at the same speed. Exceed the quota you paid for and your internet gets very, very slow.
In fact, the "pay for what you use" model is massively dominant in the marketplace; it's just been redesigned to be more consumer-friendly. People like having fixed, predictable monthly costs, and this structure lets people choose what fits their budget and needs. It also makes people less wary of downloading things, because they've already paid for their monthly quota and may as well make use of it.
It is a little odd that this scheme has become so accepted, because normally people balk at having to pay up front for a fixed amount of non-refundable, non-transferable usage.
It's a nice idea and might happen in the future - with widespread deployment of IPv6 and IPSEC it'd possibly be viable from a security perspective. There are two main blockers:
Firstly, wireless has a lot less bandwidth than wired networks. Each router would only be able to handle a small number of users, which means you'd need a very dense mesh so each router only handles a tiny fraction of the total traffic, with dynamic distribution over nodes to balance out the load. This has implications for latency/jitter-sensitive applications.
The second and really difficult to surmount problem is that of long links. There's always going to be a need for high bandwidth links over long, unpopulated distances (outside of cities, and of course inter-continental) and these are going to be very expensive to build and maintain. Someone has to pay for that.
Ultimately, internet access isn't all that expensive for the average end-user, and they get a much higher quality service than a constantly changing mesh of thousands of wireless routers could hope to provide.
Do you mean, common sense says that data transfer up to the capacity of your purchased link should be free? If so, how do you arrive at this conclusion?