It's a problem because DNS is used by more things than web browsers with human operators. A "this host does not exist" response at DNS-level contains information that a "404 not found" response at HTTP-level does not provide. And that's even assuming they have the common sense to make their "default search page" return an error status code; it's highly likely it'll return an OK status, since as a general rule the people who understand how the internet works at a technical level will refuse to be involved in these kind of projects, which means people who don't really understand what they're breaking are in charge of it all.
When Verisign did this a few years ago, they set up an SMTP rejection service so that mistyped domain names in email addresses would result in an immediate bounce, rather than sitting in the mail queue attempting to be delivered to an address that didn't accept mail for a few days before finally being bounced. This service didn't actually work properly, with the result that if you had more than one incorrect domain in the recipient list, you would get a bounce for only some of the wrong domains. This is because the people that implemented the service didn't think it was necessary to actually parse the SMTP commands, and instead just responded with a scripted "Hello, Ok, Reject" over and over again regardless of what the input was. Needless to say, this was very confusing for actual mail servers.
In addition, people using web browsers that are configured to do something useful in the case of a non-existent domain name get screwed, because now every domain resolves and serves up web pages. If Comcast's "not found" service is not as good as whatever their browser was previously doing, too bad.
At least Comcast provide an opt out, and most of their customers are presumably using Comcast's SMTP relay servers, which one would hope use real DNS servers, so the problems should not be as widespread as when Verisign did it to the entire.com namespace. However whenever you change how a fundamental part of anything works (and has worked for decades) there will always be fallout and unanticipated issues. This is also complicated by the fact you can't differentiate DNS lookups by web browsers from DNS lookups from anything else; with a result being that even when you do anticipate issues, you can't provide a 100% adequate solution to mitigate it.
The provider of the service (Gaiku) and the people making the games aren't necessarily the same people. There's also no reason why Gaiku would be the only provider of a service like this.
This provides two levels of competition: game developers and streaming game service providers. At some point, these services would reach saturation, i.e. everyone who would want to play games on the service will be playing games on it. If you, as a game developer, want to profit from it, you'll have to provide a better game than your competitors. Or at least better marketing.
Also, you assume that sequels do in fact decline in quality. While I tend to agree, you must always keep in mind the possibility that the majority of people do like the crappy sequels, which is why Westwood continues to make vast fortunes from selling crappy spinoffs to the Command and Conquer franchise. Popular culture may be awful, but it's unlikely to be popular just because everyone except me is a sheep with bad taste.
The twitch gamers are likely to be in the "hardcore" gaming market which this service wouldn't try to touch. They won't be interested in a service like this anyway, and the games that appeal to them won't appear on it.
It probably ties in nicely with the ever-expanding "casual" games market. One problem casual games face is that they need to support lowest-common-denominator hardware (and software). Many are written in Flash, and virtually none require a half-decent 3D accelerator. This limits what can be done by the developer, because you have to write it to run on non-gaming PCs.
If you can stream the video from a powerful cluster then you can do much more fancy things without pushing the system requirements way beyond what a casual gamer will actually have. You could do a fancy 3D environment with realistic physics on every object, on a bog-standard desktop PC.
Something like The Sims would be fine on it, and those games have massive audiences -- despite their pretty steep system requirements for smooth gameplay. Offloading the heavy lifting to a centralised cluster could be a huge success. The main issue would be the cost of putting sufficient processing power at the network edge to handle peak demand. The network has a lot of edges.
I think the freedom argument is because it's not possible to stop downloading without massively infringing on people's legitimate, ethical freedoms. In short, you can't effectively stop individuals downloading copyrighted materials they have no rights to without monitoring them 24/7. Nobody is suggesting that yet -- the closest is automated filters which people just work around. But that's the end game, and any move made towards that is alarming to many people.
Even in the case of things like Rapidshare, they provide a useful service to the internet but how can they reasonably provide an equivalent service if they HAVE to filter all content for potential copyright violations? Aside from logistical problems, there's technical ones: many of the things uploaded are in password-protected RARs, so Rapidshare are completely unable to check its contents, even if there was a technically feasible means for them to do so. Ban all password-protected stuff? But why shouldn't I be able to upload some of my own data with a strong password that I provide to whoever I want to have access it? What about uploading content that Rapidshare doesn't recognise? It'd be easy to, say, flip a few bits around and have a small file on another site to flip them back. The version Rapidshare would seem to be "data" of some kind and useless; but with the small bit of extra info it can be converted into a FLAC-encoded Top 40 album.
Finally, there's economics. The argument against piracy is that it costs content-producing industries so much money. The amounts they claim are often dubious, as people do buy things they've pirated sometimes, or they might pirate the first one in a series, get hooked, and then buy later ones which they otherwise wouldn't have been interested in. This can occur for music records, TV shows, games, and so on. So it's very hard to quantify the actual loss, as some piracy is unquestionably beneficial, even if only in the long run. Some is, of course, a loss in the sense that the person pirating it would've paid the requested price if they'd had to.
But it should also be realised that trying to protect against piracy has an economic cost, too. Consider how much time/money would be spent retooling sites like Rapidshare to enable copyright filtering. Now expand that to any site that accepts any kind of user content, and the sheer amount of inefficiency introduced is mind-blowing. Is the economic cost of piracy actually greater than the economic cost of preventing it?
Then back to freedom. Even if you can implement effective measures, how do you prevent them from being abused? Unless you decide that "privacy" is an antiquated notion and teach people not to expect to have any kind of privacy ever, this is a real problem. Someone has to do the watching, but who watches the watchers? What's to stop the copyright enforcement technology being used by governments or other powerful entities to control the other information that is available? Or simply fabricating claims of copyright infringement against people who are in there way?
So while combating piracy isn't itself a bad thing, the cost to society of actually preventing it may be much higher than the cost of the piracy itself.
Right... but how do you ensure you have that? Pretty much everyone is susceptible to corruption, and any worthwhile police force or army is going to have to consist of a lot of people. Decent people can easily become indecent, as has been shown time and again in history. Saying "but we'd only have good, uncorruptible, virtuous people in positions of power" is a cop-out that has zero chance of success. You may as well say we're going to solve world hunger and end all wars just be having everyone decide to be nice to each other. Sure it makes sense, but it's obviously not what actually happens.
Regardless, even if you get your wish, what are non-violent protests going to achieve? Why would the government pay any attention? All you're doing is standing around chanting catchy slogans. They're free to continue doing whatever they want. Unless you somehow get the vast majority of citizens to join the protest and down tools, but are you really naive enough to think that'll happen? Even without using any force against protestors, the government can just make offers to people who continue to work and ignore the protestors - tax cuts, or access to useful services like health care and public transport, and so on. Not to mention the fact that if the people in power think what they're doing is a good idea, it's highly likely a significant number of ordinary folk will too and won't even want to resist.
Your post would have been better if you could spell brakes correctly. Although I do like the concept of an "autobreak" system disintegrating the aircraft in mid-air. That would make marketing earn their pay.
I think it's because the war of independence -- where the citizens of North America took up arms against the then government and won their independence -- is a significant part of the identity of many North American citizens. Also don't forget that the founding fathers made it very clear that all citizens must have the right to possess arms, for the explicit reason of being able to overthrow a future corrupt government. While one hopes that would never come to pass, it's still an important part of the very identity of US citizens.
Or to put the question back to you: if your government ignores the apparent will of the majority of the people, how will you react? If they start imposing martial law to keep the peace and outlaw protests against their clearly unpopular policies, what can the ordinary citizen do?
Armed revolution is most definitely a last resort, but if you don't have an unpalatable last resort then what's stopping those in power from abusing that power in the extreme?
I've never seen a slashdot article with no comments before. Okay, so it's only idle so it doesn't really count, but it's still pretty eery. Like being in the middle of a deserted metropolis.
How dangerous are these decades-old unexploded ordinances? I mean, they haven't exploded for 50 years, so they probably won't explode in the next few days... right? Or is it a case that they may become very fragile after sitting undisturbed and unmoved for so long, and even the slightest bump can set it off?
I wouldn't think Akamai would be doing any of the actual work behind the iTunes store. I seem to recall they do have that capability, but it would be really hard to take advantage of unless you designed for it from the start, and even then I doubt anyone, especially a company as large as Apple, would be happy to give their content distribution network access to any of their actual user data.
Our website is served by Akamai as well, but nearly all the content is served by Windows web servers. If you do a simple GET and the page is in the cache of the Akamai server you're using, then you could maybe say it was served by Linux or whatever. If you do a search or anything that requires actual work, your request will be getting funneled back to our Windows servers.
I would say it's extremely unlikely iTunes works any differently.
This sounds too much like presumption of guilt to me. In a perfect world perhaps it would be reasonable, but this isn't a perfect world and if the police decide to pin a crime on someone because they know they play these types of games, that person is going to suffer unreasonable harassment.
Also, the very existence of this information and its ability to be used for these purposes means that your arguments are conflicting with each other: those who do have these types of fantasies and think there's even a remote chance they may act on them one day will avoid ever having any association with these types of products. Therefore, if playing these games does actually have any kind of effect on people's real life behaviour, those "on the border" who would benefit from having a safe, non-harmful outlet will deliberately avoid utilising that outlet.
In addition, the negative stigma that is obviously being attached to it ("you can have the game, but we're putting you on our watch list, you disgusting pervert") means people will avoid them. I think these things are only useful if they can de-stigmatise particular desires, to effect a shift in perception to one of understanding: "yes you can play these games, it's fine to have these fantasies, just be aware that doing it for real will make us all very upset".
Making people feel ashamed of themselves for their thoughts and primal urges seems counter-productive, to me.
I don't think it's safe to let people indulge in... certain fantasies.
Thoughtcrime.
If we could monitor the thoughts of every citizen and punish anyone who thinks bad things (like harming others), would it be a good thing for society?
Is the act of playing a video game in the privacy of your home sufficiently far removed from the act of fantasizing within one's own mind to warrant being treated differently?
If you'd rather keep your SSN to yourself, then... don't enter your SSN on any websites?
Exploits aside, cookies can't be used to share information between websites, so even if a site you trust decides to do something retarded like store your SSN in a cookie, other sites can't access it.
Or was the reference to your Social Security Number just a bad example and you were really thinking of other things that can be automatically collected? Most of that isn't particularly private though, and can be altered if you have reason to think your OS or browser version or screen resolution are things you need to keep secret.
The only thing I found surprising was that Microsoft was actually organised enough to be able to conduct such experiments in a logical and usual fashion.;) Now that I've had a night to get used to the idea, the only surprising thing is I wasn't consciously aware that everyone already does this. In my defense, I'd never given it any though before.
To extend your restraunt analogy would you keep going to a restraunt if you discovered they had been deliberately delaying your order for research or worse to try and extract more money from you?
Of course. I mean, I assume every business is doing stuff like that all the time. If I found the wait annoying I might go elsewhere, of course. But not because they were trying to work out how to increase their profits (newsflash: almost everything any business does is an attempt to increase profits).
I wonder what the people who are experiencing moral outrage over this are going to do when they realise that every single business they interact with is also experimenting on customers? Everything from changing staffing levels to introducing a new kind of burger or changing the colours or fonts or imagery used in their advertising is all an experiment in one form or another.
My local convenience store rearranged their cashier section a while ago. It was pretty obvious, too; they set up a snaking laneway to the registers by using shelves with confectionary as walls. Later they changed it so those shelves were perpendicular to the registers. The first way worked better IMHO, because now at busy times multiple queues start forming, whereas before there was only one queue. I guess some people didn't like being channelled through the confectionary aisle, so they changed it in response to customer's reactions. Was doing this unethical?
The line for me is, "would I consider this to be unethical (or just plain nasty) if they did it on a permanent basis, or put in place policies that would have this effect on a permanent basis".
So to use the examples we've been using so far: slowing down the loading time of a web page or the response time at a restaurant. If they made a decision to put less emphasis on page optimisation or to reduce staff, knowing full well it'd result in longer waits during busy times, would I think that's unethical or nasty? No, of course not. Every business has to make trade-offs as to where to spend money, and if they determine that shaving an extra half a second off of page load times is not worth the expense, then it's a perfectly fair decision to make.
If they decide to put in place a policy whereby they punch their customers in the face (even if it is only very infrequently) or show seizure-inducing flashing imagery on their website to some of their customers, then yes I would consider that to be very unethical.
I grok the "slippery slope" argument, but it is sometimes used inappropriately. There is a world of difference between simulating the effects of very possible and legitimate trade-offs in spending priorities, and causing physical or mental trauma to customers.
The "slippery slope" argument also works the other way; and I did actually pose some questions to you in that vein in my last post. Every business experiments on its customer base all the time. Some do it on a small group so they can compare results. Others do it on every customer. So again I ask: is it unethical for a company to reduce the staffing levels at peak times throughout every restaurant (assuming they're a chain) in order to save some costs and see the effect of it? Or is it only unethical if they do it at a single store? Or is that still okay, and it only becomes unethical if they are simulating it on a randomly selected group of customers at a particular store or stores? What do you draw the line?
Or is there no line, and it's simply unethical for any business to ever try anything different and measure the impact it has in order to try to improve their business?
A few people seem to be implying some kind of ethical issue with this practice, but try as I might I don't see the problem. It's not as if they're subjecting customers to some kind of degrading experience (despite the slashdot headline). So the site loads slightly slow, or the page is a little less optimised, or the waitstaff are a bit slower. How is this at all unethical? These are all perfectly normal things that can and do happen "by accident" all the time. I don't see how artificially causing them to occur so you can determine their impact is in any way a violation of anyone's rights or moral or ethical obligations.
Your example was absurd because you're clearly inflicting actual harm on the customer. You can argue until you're blue in the face, but I will never accept that an additional delay when loading a web page or ordering food is in any way a meaningful "harm". These are things that happen all the time for a variety of reasons, some within the control of company and some outside.
Experimentation happens all the time, and is a vital part of the economy. In my restaurant example, what if management simply decided (based on gut feel) that the cost of having additional staff exceeded the additional revenue they'd receive by having timely service? So they cut shifts. After a few weeks of this they realise that the poor service did actually cost them more than they saved on wages, so they abandon that policy and put on extra staff. Unethical bastards! They've just conducted an experiment on their entire customer base in order to determine the most effective use of their funds. How could anyone be so morally bankrupt?!
Or on the subject of the article, what if management decided that the time it takes to optimise page load times was costing them too much money and stopped doing that? Hey presto, every single visitor gets a slower, more bloated page and a slightly "degraded" user experience. Only it's not really degraded because it's the normal experience for that website. Is that ethical because they're giving all their customers the same suboptimal experience? What if it turns out the whole thing was actually an experiment to see what effect it had -- is that now suddenly unethical?
If so, your position seems to imply that a company doing anything less than dedicating every available resource to ensure the optimal satisfation of every single customer is unethical. However, if a company conducts any kind of measured testing to determine what's most beneficial for their customers, they're also being unethical.
Or is your only issue with this that it's occuring to only a small random (or not) sample of the customer base?
It sounds absurd because what you're saying is absurd.
If you ran an experiment where some customers had their orders delayed by a few minutes more than was necessary and had some kind of metric to determine their enjoyment of their dining experience, it wouldn't be so absurd. Perhaps you provide free internet access in your store, and the extra delay results in a greater chance of people making use of it. And once they've started using it, there's a greater chance they'll decide to order a coffee after their meal and stick around for a bit longer.
Or maybe you find they're less likely to return to the store. That might be hard to track, but the point stands. There are some things which are interesting and which may or may not give unexpected results when tried in real life. If an experiment like this shows that a few minutes delay significantly upsets customers, then it becomes clear that spending extra money to have more staff on is probably actually worth the expense. On the other hand if you can show that most people don't notice, then it makes sense to risk having a shortage of staff at peak periods if you can save a bit of money.
You might even find unexpected results, for example maybe a lot of people after waiting a few minutes with nothing to look at but the menu end up ordering more than they initially would, so it's actually profitable to make people wait longer. Who knows? The only way to find out is to experiment.
If you're using something like password safe and have access to its database, then you won't need to be recovering your password in the first place, will you?
Ubuntu and most other distributions let you select which packages to install, so that's the main reason it's not a trivial file copy operation - you want to make sure the package database matches the installed components so that later updates will work. The other configuration that takes place is things like the system hostname, IP addresses, display resolution, keyboard type, local users, etc. Some of that can be left at defaults (i.e. DHCP) while others would need to be done once per model you're pushing out. I don't particularly see why you'd have to do more than a) copy your standard installation to the hard drive, b) run GRUB or whatever bootloader you're using to set up the MBR and c) copy any per-model configuration files over. Most of X11 can configure itself automatically as pretty much all monitors being sold with PCs (or built in) can report their supported and preferred resolutions.
I don't think OEMs would really care about anything except the current mainstream kernel and related HAL. Linux tends to have fairly good backwards compatibility, but if you're imaging a fleet of thousands of machines from different eras you'd likely to run into trouble. I don't think an OEM would really have much problem; most of their hardware is going to be pretty similar at any point in time.
As far as drivers being in the right place, it's pretty much just a matter of them being in an appropriate directory under/lib/modules/kernel-version/ isn't it? The only driver configuration I've had to do since the ISA days was for the network bonding driver, which I don't think applies to many OEMs.
The whole premise of your post seems off to me. Linux loads just about everything at runtime. You don't need a sysprep equivalent because it doesn't store the driver it's going to use. For example consider Window's weird USB support; I think this might be fixed in Vista, but I'm not 100% sure. Certainly with XP if you plug a USB storage device into a USB port, it'll load drivers and then present it to you. Remove it and plug the exact same thing into a different USB port... it'll load drivers and then present it to you. Plug it in to the same port and it's instantly available.
This goes to the core of driver support, even well into the "Plug and Play" era: Windows always associates drivers with particular hardware device addresses and has to store configuration information whenever that changes. No such issues on Linux. The closest you'd get is having to clean up the udev files which ensure particularly hardware gets assigned the same device name each boot (i.e. the various _persistent_ rulefiles).
The only other issue you might have is if the kernel is unable to boot on the hardware, though pretty much all distributions use large initrds which include drivers for virtually everything.
Once upon a time I rebuilt my PC, and decided to see if I could get away with not having to re-install Windows as the build was very similar. It did in fact work quite well. I had a dual boot system. Linux booted up as normal, just a bit faster because of the faster processor etc. Windows booted up okay, then futzed around saying it was installing drivers for my new hardware and needed a reboot or two before it was happy. It wasn't quite right though, as from thereafter it never shut down properly. It would shut down Windows, but wouldn't turn the power off or reboot. I guess the power management was slightly different with the new motherboard, and Windows had at some point installed something specific for the previous chipset. The Linux kernel just works out what needs to be done each time it's booted, and so it all worked perfectly fine.
At work I've upgraded a Linux server installed on an HP DL360 to a DL380 just by moving the drives to the new system. The only complication I would ever imagine facing is if the hardware RAID controller doesn't recognise the drives, but I didn't have that issue as they were similar-generation. I wouldn't even try that with a Windows install, because even if the hardware seems to be 100% identical Windows will still notice different device IDs and have a hissy fit. The only problem I encountered with the Linux install was that the network interfaces were assigned silly names because it was reserving eth0 and eth1 for the previous IDs; again, just nuking those persistent config files and rebooted sorted it out.
You do make a good point about kickbacks from pre-installing all the garbage you get with a big manufacturer PC. While they could do the same thing with Linux, I'd imagine most people opting for Linux at this stage would find that to be a complete deal-breaker. In addition, the fact that Windows and Linux are in many ways very different platforms does add complications -- they've had many many years to organise their deployment strategies and toolchains around Windows' peculiarities, and adapting to the peculiarities of any other system will obviously involve some cost.
I would also imagine that they make some amount of profit by including commercial software, in the same way a retail shop selling boxes of software makes a bit of profit. If everything you're including is free software, then it's harder to profit off of that -- the natural end-game would seem to be vendors competing purely on the basis of hardware costs, which I don't think any of them particularly want to do.
Firstly, how do you know they're not using Linux or something other than Windows? I mean, I assume they're not like you're doing, but we don't know. The BSAs methodology is basically to estimate the number of PCs and then look at sales figures in the region and assume that there should be corresponding software sales. Since Linux isn't sold, if Linux usage was 99% there'd still be a massive disparity between the number of PCs and the number of OS sales. Very little piracy, though.
Additionally you argue the point as if "stealing" Windows is somehow harder than using Linux for free. In reality, pirated copies of Windows would be as easy to obtain as copies of a Linux distribution. If Windows has already achieved a significant marketshare (e.g. by Microsoft choosing to ignore the piracy because they'd rather people pirate Windows than use an alternative OS) then it would probably be easier to obtain a pirated copy of Windows than a legitimate copy of Linux. High marketshare tends to self-reinforce, since most people want whatever most other people are using.
Fuck me that's bizarre. I've had Symbian phones for ages and they have loads of apps available for them, but I pretty much don't buy any because I've found that 99% of the time I use it a few times and then never touch it again. And that's for things that actually seem useful at first.
iSnort? Maybe the poster that linked to that as an example of why they wanted an iPhone was taking the piss. Why not just use a drawing application to draw some white chalky lines on a black background and then switch to the eraser tool? Why would you pay 5 anythings for that app?
One thing that always annoyed me about huntsman spiders is that they do, in fact, chase you around. I think they have a climbing instinct, and when they're sitting on a floor they really really want to climb something; and if a person happens to be nearby it must look like a tree or something equally climbable.
I've noticed it a few times, but one time in particular I remember was in our tiled entry (which was basically just a room that happened to have the front door to the house) and I was trying to catch a huntsman that had ended up on the floor. I guess I was a bit slow and it started moving about, coming straight for me. Slightly freaky but I figured it was just a chance thing, so I backed off and it kept coming, then when I ran out of room I stepped over it to give some space. So it stopped, turned around, and started running toward me again.
That continued for some time, but it eventually stopped. I guess it was starting to wonder why that tree was so difficult to get to and wanted to have a think about it for a while. So I used that opportunity to catch it and relocate it. Again I don't think it was aggressive or anything, it wasn't showing any signs of aggression. Pretty sure if I'd let it reach me it would've just started scaling me, rather than trying to eat me. Still, I didn't particularly want it to do either.
Slashdot Mirror
So... it was a bad analogy, then?
It's a problem because DNS is used by more things than web browsers with human operators. A "this host does not exist" response at DNS-level contains information that a "404 not found" response at HTTP-level does not provide. And that's even assuming they have the common sense to make their "default search page" return an error status code; it's highly likely it'll return an OK status, since as a general rule the people who understand how the internet works at a technical level will refuse to be involved in these kind of projects, which means people who don't really understand what they're breaking are in charge of it all.
When Verisign did this a few years ago, they set up an SMTP rejection service so that mistyped domain names in email addresses would result in an immediate bounce, rather than sitting in the mail queue attempting to be delivered to an address that didn't accept mail for a few days before finally being bounced. This service didn't actually work properly, with the result that if you had more than one incorrect domain in the recipient list, you would get a bounce for only some of the wrong domains. This is because the people that implemented the service didn't think it was necessary to actually parse the SMTP commands, and instead just responded with a scripted "Hello, Ok, Reject" over and over again regardless of what the input was. Needless to say, this was very confusing for actual mail servers.
In addition, people using web browsers that are configured to do something useful in the case of a non-existent domain name get screwed, because now every domain resolves and serves up web pages. If Comcast's "not found" service is not as good as whatever their browser was previously doing, too bad.
At least Comcast provide an opt out, and most of their customers are presumably using Comcast's SMTP relay servers, which one would hope use real DNS servers, so the problems should not be as widespread as when Verisign did it to the entire .com namespace. However whenever you change how a fundamental part of anything works (and has worked for decades) there will always be fallout and unanticipated issues. This is also complicated by the fact you can't differentiate DNS lookups by web browsers from DNS lookups from anything else; with a result being that even when you do anticipate issues, you can't provide a 100% adequate solution to mitigate it.
If monitoring everything your sysadmins do is important because they have a habit of wiping their histories, I think you need new sysadmins.
The provider of the service (Gaiku) and the people making the games aren't necessarily the same people. There's also no reason why Gaiku would be the only provider of a service like this.
This provides two levels of competition: game developers and streaming game service providers. At some point, these services would reach saturation, i.e. everyone who would want to play games on the service will be playing games on it. If you, as a game developer, want to profit from it, you'll have to provide a better game than your competitors. Or at least better marketing.
Also, you assume that sequels do in fact decline in quality. While I tend to agree, you must always keep in mind the possibility that the majority of people do like the crappy sequels, which is why Westwood continues to make vast fortunes from selling crappy spinoffs to the Command and Conquer franchise. Popular culture may be awful, but it's unlikely to be popular just because everyone except me is a sheep with bad taste.
The twitch gamers are likely to be in the "hardcore" gaming market which this service wouldn't try to touch. They won't be interested in a service like this anyway, and the games that appeal to them won't appear on it.
It probably ties in nicely with the ever-expanding "casual" games market. One problem casual games face is that they need to support lowest-common-denominator hardware (and software). Many are written in Flash, and virtually none require a half-decent 3D accelerator. This limits what can be done by the developer, because you have to write it to run on non-gaming PCs.
If you can stream the video from a powerful cluster then you can do much more fancy things without pushing the system requirements way beyond what a casual gamer will actually have. You could do a fancy 3D environment with realistic physics on every object, on a bog-standard desktop PC.
Something like The Sims would be fine on it, and those games have massive audiences -- despite their pretty steep system requirements for smooth gameplay. Offloading the heavy lifting to a centralised cluster could be a huge success. The main issue would be the cost of putting sufficient processing power at the network edge to handle peak demand. The network has a lot of edges.
I think the freedom argument is because it's not possible to stop downloading without massively infringing on people's legitimate, ethical freedoms. In short, you can't effectively stop individuals downloading copyrighted materials they have no rights to without monitoring them 24/7. Nobody is suggesting that yet -- the closest is automated filters which people just work around. But that's the end game, and any move made towards that is alarming to many people.
Even in the case of things like Rapidshare, they provide a useful service to the internet but how can they reasonably provide an equivalent service if they HAVE to filter all content for potential copyright violations? Aside from logistical problems, there's technical ones: many of the things uploaded are in password-protected RARs, so Rapidshare are completely unable to check its contents, even if there was a technically feasible means for them to do so. Ban all password-protected stuff? But why shouldn't I be able to upload some of my own data with a strong password that I provide to whoever I want to have access it? What about uploading content that Rapidshare doesn't recognise? It'd be easy to, say, flip a few bits around and have a small file on another site to flip them back. The version Rapidshare would seem to be "data" of some kind and useless; but with the small bit of extra info it can be converted into a FLAC-encoded Top 40 album.
Finally, there's economics. The argument against piracy is that it costs content-producing industries so much money. The amounts they claim are often dubious, as people do buy things they've pirated sometimes, or they might pirate the first one in a series, get hooked, and then buy later ones which they otherwise wouldn't have been interested in. This can occur for music records, TV shows, games, and so on. So it's very hard to quantify the actual loss, as some piracy is unquestionably beneficial, even if only in the long run. Some is, of course, a loss in the sense that the person pirating it would've paid the requested price if they'd had to.
But it should also be realised that trying to protect against piracy has an economic cost, too. Consider how much time/money would be spent retooling sites like Rapidshare to enable copyright filtering. Now expand that to any site that accepts any kind of user content, and the sheer amount of inefficiency introduced is mind-blowing. Is the economic cost of piracy actually greater than the economic cost of preventing it?
Then back to freedom. Even if you can implement effective measures, how do you prevent them from being abused? Unless you decide that "privacy" is an antiquated notion and teach people not to expect to have any kind of privacy ever, this is a real problem. Someone has to do the watching, but who watches the watchers? What's to stop the copyright enforcement technology being used by governments or other powerful entities to control the other information that is available? Or simply fabricating claims of copyright infringement against people who are in there way?
So while combating piracy isn't itself a bad thing, the cost to society of actually preventing it may be much higher than the cost of the piracy itself.
Right... but how do you ensure you have that? Pretty much everyone is susceptible to corruption, and any worthwhile police force or army is going to have to consist of a lot of people. Decent people can easily become indecent, as has been shown time and again in history. Saying "but we'd only have good, uncorruptible, virtuous people in positions of power" is a cop-out that has zero chance of success. You may as well say we're going to solve world hunger and end all wars just be having everyone decide to be nice to each other. Sure it makes sense, but it's obviously not what actually happens.
Regardless, even if you get your wish, what are non-violent protests going to achieve? Why would the government pay any attention? All you're doing is standing around chanting catchy slogans. They're free to continue doing whatever they want. Unless you somehow get the vast majority of citizens to join the protest and down tools, but are you really naive enough to think that'll happen? Even without using any force against protestors, the government can just make offers to people who continue to work and ignore the protestors - tax cuts, or access to useful services like health care and public transport, and so on. Not to mention the fact that if the people in power think what they're doing is a good idea, it's highly likely a significant number of ordinary folk will too and won't even want to resist.
Your post would have been better if you could spell brakes correctly. Although I do like the concept of an "autobreak" system disintegrating the aircraft in mid-air. That would make marketing earn their pay.
I think it's because the war of independence -- where the citizens of North America took up arms against the then government and won their independence -- is a significant part of the identity of many North American citizens. Also don't forget that the founding fathers made it very clear that all citizens must have the right to possess arms, for the explicit reason of being able to overthrow a future corrupt government. While one hopes that would never come to pass, it's still an important part of the very identity of US citizens.
Or to put the question back to you: if your government ignores the apparent will of the majority of the people, how will you react? If they start imposing martial law to keep the peace and outlaw protests against their clearly unpopular policies, what can the ordinary citizen do?
Armed revolution is most definitely a last resort, but if you don't have an unpalatable last resort then what's stopping those in power from abusing that power in the extreme?
I've never seen a slashdot article with no comments before. Okay, so it's only idle so it doesn't really count, but it's still pretty eery. Like being in the middle of a deserted metropolis.
How dangerous are these decades-old unexploded ordinances? I mean, they haven't exploded for 50 years, so they probably won't explode in the next few days... right? Or is it a case that they may become very fragile after sitting undisturbed and unmoved for so long, and even the slightest bump can set it off?
I wouldn't think Akamai would be doing any of the actual work behind the iTunes store. I seem to recall they do have that capability, but it would be really hard to take advantage of unless you designed for it from the start, and even then I doubt anyone, especially a company as large as Apple, would be happy to give their content distribution network access to any of their actual user data.
Our website is served by Akamai as well, but nearly all the content is served by Windows web servers. If you do a simple GET and the page is in the cache of the Akamai server you're using, then you could maybe say it was served by Linux or whatever. If you do a search or anything that requires actual work, your request will be getting funneled back to our Windows servers.
I would say it's extremely unlikely iTunes works any differently.
This sounds too much like presumption of guilt to me. In a perfect world perhaps it would be reasonable, but this isn't a perfect world and if the police decide to pin a crime on someone because they know they play these types of games, that person is going to suffer unreasonable harassment.
Also, the very existence of this information and its ability to be used for these purposes means that your arguments are conflicting with each other: those who do have these types of fantasies and think there's even a remote chance they may act on them one day will avoid ever having any association with these types of products. Therefore, if playing these games does actually have any kind of effect on people's real life behaviour, those "on the border" who would benefit from having a safe, non-harmful outlet will deliberately avoid utilising that outlet.
In addition, the negative stigma that is obviously being attached to it ("you can have the game, but we're putting you on our watch list, you disgusting pervert") means people will avoid them. I think these things are only useful if they can de-stigmatise particular desires, to effect a shift in perception to one of understanding: "yes you can play these games, it's fine to have these fantasies, just be aware that doing it for real will make us all very upset".
Making people feel ashamed of themselves for their thoughts and primal urges seems counter-productive, to me.
I don't think it's safe to let people indulge in... certain fantasies.
Thoughtcrime.
If we could monitor the thoughts of every citizen and punish anyone who thinks bad things (like harming others), would it be a good thing for society?
Is the act of playing a video game in the privacy of your home sufficiently far removed from the act of fantasizing within one's own mind to warrant being treated differently?
If you'd rather keep your SSN to yourself, then... don't enter your SSN on any websites?
Exploits aside, cookies can't be used to share information between websites, so even if a site you trust decides to do something retarded like store your SSN in a cookie, other sites can't access it.
Or was the reference to your Social Security Number just a bad example and you were really thinking of other things that can be automatically collected? Most of that isn't particularly private though, and can be altered if you have reason to think your OS or browser version or screen resolution are things you need to keep secret.
The only thing I found surprising was that Microsoft was actually organised enough to be able to conduct such experiments in a logical and usual fashion. ;) Now that I've had a night to get used to the idea, the only surprising thing is I wasn't consciously aware that everyone already does this. In my defense, I'd never given it any though before.
To extend your restraunt analogy would you keep going to a restraunt if you discovered they had been deliberately delaying your order for research or worse to try and extract more money from you?
Of course. I mean, I assume every business is doing stuff like that all the time. If I found the wait annoying I might go elsewhere, of course. But not because they were trying to work out how to increase their profits (newsflash: almost everything any business does is an attempt to increase profits).
I wonder what the people who are experiencing moral outrage over this are going to do when they realise that every single business they interact with is also experimenting on customers? Everything from changing staffing levels to introducing a new kind of burger or changing the colours or fonts or imagery used in their advertising is all an experiment in one form or another.
My local convenience store rearranged their cashier section a while ago. It was pretty obvious, too; they set up a snaking laneway to the registers by using shelves with confectionary as walls. Later they changed it so those shelves were perpendicular to the registers. The first way worked better IMHO, because now at busy times multiple queues start forming, whereas before there was only one queue. I guess some people didn't like being channelled through the confectionary aisle, so they changed it in response to customer's reactions. Was doing this unethical?
The line for me is, "would I consider this to be unethical (or just plain nasty) if they did it on a permanent basis, or put in place policies that would have this effect on a permanent basis".
So to use the examples we've been using so far: slowing down the loading time of a web page or the response time at a restaurant. If they made a decision to put less emphasis on page optimisation or to reduce staff, knowing full well it'd result in longer waits during busy times, would I think that's unethical or nasty? No, of course not. Every business has to make trade-offs as to where to spend money, and if they determine that shaving an extra half a second off of page load times is not worth the expense, then it's a perfectly fair decision to make.
If they decide to put in place a policy whereby they punch their customers in the face (even if it is only very infrequently) or show seizure-inducing flashing imagery on their website to some of their customers, then yes I would consider that to be very unethical.
I grok the "slippery slope" argument, but it is sometimes used inappropriately. There is a world of difference between simulating the effects of very possible and legitimate trade-offs in spending priorities, and causing physical or mental trauma to customers.
The "slippery slope" argument also works the other way; and I did actually pose some questions to you in that vein in my last post. Every business experiments on its customer base all the time. Some do it on a small group so they can compare results. Others do it on every customer. So again I ask: is it unethical for a company to reduce the staffing levels at peak times throughout every restaurant (assuming they're a chain) in order to save some costs and see the effect of it? Or is it only unethical if they do it at a single store? Or is that still okay, and it only becomes unethical if they are simulating it on a randomly selected group of customers at a particular store or stores? What do you draw the line?
Or is there no line, and it's simply unethical for any business to ever try anything different and measure the impact it has in order to try to improve their business?
A few people seem to be implying some kind of ethical issue with this practice, but try as I might I don't see the problem. It's not as if they're subjecting customers to some kind of degrading experience (despite the slashdot headline). So the site loads slightly slow, or the page is a little less optimised, or the waitstaff are a bit slower. How is this at all unethical? These are all perfectly normal things that can and do happen "by accident" all the time. I don't see how artificially causing them to occur so you can determine their impact is in any way a violation of anyone's rights or moral or ethical obligations.
Your example was absurd because you're clearly inflicting actual harm on the customer. You can argue until you're blue in the face, but I will never accept that an additional delay when loading a web page or ordering food is in any way a meaningful "harm". These are things that happen all the time for a variety of reasons, some within the control of company and some outside.
Experimentation happens all the time, and is a vital part of the economy. In my restaurant example, what if management simply decided (based on gut feel) that the cost of having additional staff exceeded the additional revenue they'd receive by having timely service? So they cut shifts. After a few weeks of this they realise that the poor service did actually cost them more than they saved on wages, so they abandon that policy and put on extra staff. Unethical bastards! They've just conducted an experiment on their entire customer base in order to determine the most effective use of their funds. How could anyone be so morally bankrupt?!
Or on the subject of the article, what if management decided that the time it takes to optimise page load times was costing them too much money and stopped doing that? Hey presto, every single visitor gets a slower, more bloated page and a slightly "degraded" user experience. Only it's not really degraded because it's the normal experience for that website. Is that ethical because they're giving all their customers the same suboptimal experience? What if it turns out the whole thing was actually an experiment to see what effect it had -- is that now suddenly unethical?
If so, your position seems to imply that a company doing anything less than dedicating every available resource to ensure the optimal satisfation of every single customer is unethical. However, if a company conducts any kind of measured testing to determine what's most beneficial for their customers, they're also being unethical.
Or is your only issue with this that it's occuring to only a small random (or not) sample of the customer base?
It sounds absurd because what you're saying is absurd.
If you ran an experiment where some customers had their orders delayed by a few minutes more than was necessary and had some kind of metric to determine their enjoyment of their dining experience, it wouldn't be so absurd. Perhaps you provide free internet access in your store, and the extra delay results in a greater chance of people making use of it. And once they've started using it, there's a greater chance they'll decide to order a coffee after their meal and stick around for a bit longer.
Or maybe you find they're less likely to return to the store. That might be hard to track, but the point stands. There are some things which are interesting and which may or may not give unexpected results when tried in real life. If an experiment like this shows that a few minutes delay significantly upsets customers, then it becomes clear that spending extra money to have more staff on is probably actually worth the expense. On the other hand if you can show that most people don't notice, then it makes sense to risk having a shortage of staff at peak periods if you can save a bit of money.
You might even find unexpected results, for example maybe a lot of people after waiting a few minutes with nothing to look at but the menu end up ordering more than they initially would, so it's actually profitable to make people wait longer. Who knows? The only way to find out is to experiment.
If you're using something like password safe and have access to its database, then you won't need to be recovering your password in the first place, will you?
Ubuntu and most other distributions let you select which packages to install, so that's the main reason it's not a trivial file copy operation - you want to make sure the package database matches the installed components so that later updates will work. The other configuration that takes place is things like the system hostname, IP addresses, display resolution, keyboard type, local users, etc. Some of that can be left at defaults (i.e. DHCP) while others would need to be done once per model you're pushing out. I don't particularly see why you'd have to do more than a) copy your standard installation to the hard drive, b) run GRUB or whatever bootloader you're using to set up the MBR and c) copy any per-model configuration files over. Most of X11 can configure itself automatically as pretty much all monitors being sold with PCs (or built in) can report their supported and preferred resolutions.
I don't think OEMs would really care about anything except the current mainstream kernel and related HAL. Linux tends to have fairly good backwards compatibility, but if you're imaging a fleet of thousands of machines from different eras you'd likely to run into trouble. I don't think an OEM would really have much problem; most of their hardware is going to be pretty similar at any point in time.
As far as drivers being in the right place, it's pretty much just a matter of them being in an appropriate directory under /lib/modules/kernel-version/ isn't it? The only driver configuration I've had to do since the ISA days was for the network bonding driver, which I don't think applies to many OEMs.
The whole premise of your post seems off to me. Linux loads just about everything at runtime. You don't need a sysprep equivalent because it doesn't store the driver it's going to use. For example consider Window's weird USB support; I think this might be fixed in Vista, but I'm not 100% sure. Certainly with XP if you plug a USB storage device into a USB port, it'll load drivers and then present it to you. Remove it and plug the exact same thing into a different USB port... it'll load drivers and then present it to you. Plug it in to the same port and it's instantly available.
This goes to the core of driver support, even well into the "Plug and Play" era: Windows always associates drivers with particular hardware device addresses and has to store configuration information whenever that changes. No such issues on Linux. The closest you'd get is having to clean up the udev files which ensure particularly hardware gets assigned the same device name each boot (i.e. the various _persistent_ rulefiles).
The only other issue you might have is if the kernel is unable to boot on the hardware, though pretty much all distributions use large initrds which include drivers for virtually everything.
Once upon a time I rebuilt my PC, and decided to see if I could get away with not having to re-install Windows as the build was very similar. It did in fact work quite well. I had a dual boot system. Linux booted up as normal, just a bit faster because of the faster processor etc. Windows booted up okay, then futzed around saying it was installing drivers for my new hardware and needed a reboot or two before it was happy. It wasn't quite right though, as from thereafter it never shut down properly. It would shut down Windows, but wouldn't turn the power off or reboot. I guess the power management was slightly different with the new motherboard, and Windows had at some point installed something specific for the previous chipset. The Linux kernel just works out what needs to be done each time it's booted, and so it all worked perfectly fine.
At work I've upgraded a Linux server installed on an HP DL360 to a DL380 just by moving the drives to the new system. The only complication I would ever imagine facing is if the hardware RAID controller doesn't recognise the drives, but I didn't have that issue as they were similar-generation. I wouldn't even try that with a Windows install, because even if the hardware seems to be 100% identical Windows will still notice different device IDs and have a hissy fit. The only problem I encountered with the Linux install was that the network interfaces were assigned silly names because it was reserving eth0 and eth1 for the previous IDs; again, just nuking those persistent config files and rebooted sorted it out.
You do make a good point about kickbacks from pre-installing all the garbage you get with a big manufacturer PC. While they could do the same thing with Linux, I'd imagine most people opting for Linux at this stage would find that to be a complete deal-breaker. In addition, the fact that Windows and Linux are in many ways very different platforms does add complications -- they've had many many years to organise their deployment strategies and toolchains around Windows' peculiarities, and adapting to the peculiarities of any other system will obviously involve some cost.
I would also imagine that they make some amount of profit by including commercial software, in the same way a retail shop selling boxes of software makes a bit of profit. If everything you're including is free software, then it's harder to profit off of that -- the natural end-game would seem to be vendors competing purely on the basis of hardware costs, which I don't think any of them particularly want to do.
Firstly, how do you know they're not using Linux or something other than Windows? I mean, I assume they're not like you're doing, but we don't know. The BSAs methodology is basically to estimate the number of PCs and then look at sales figures in the region and assume that there should be corresponding software sales. Since Linux isn't sold, if Linux usage was 99% there'd still be a massive disparity between the number of PCs and the number of OS sales. Very little piracy, though.
Additionally you argue the point as if "stealing" Windows is somehow harder than using Linux for free. In reality, pirated copies of Windows would be as easy to obtain as copies of a Linux distribution. If Windows has already achieved a significant marketshare (e.g. by Microsoft choosing to ignore the piracy because they'd rather people pirate Windows than use an alternative OS) then it would probably be easier to obtain a pirated copy of Windows than a legitimate copy of Linux. High marketshare tends to self-reinforce, since most people want whatever most other people are using.
What phones are these? My newish phone came with RoadSync pre-installed, but I don't think that exactly counts.
Fuck me that's bizarre. I've had Symbian phones for ages and they have loads of apps available for them, but I pretty much don't buy any because I've found that 99% of the time I use it a few times and then never touch it again. And that's for things that actually seem useful at first.
iSnort? Maybe the poster that linked to that as an example of why they wanted an iPhone was taking the piss. Why not just use a drawing application to draw some white chalky lines on a black background and then switch to the eraser tool? Why would you pay 5 anythings for that app?
One thing that always annoyed me about huntsman spiders is that they do, in fact, chase you around. I think they have a climbing instinct, and when they're sitting on a floor they really really want to climb something; and if a person happens to be nearby it must look like a tree or something equally climbable.
I've noticed it a few times, but one time in particular I remember was in our tiled entry (which was basically just a room that happened to have the front door to the house) and I was trying to catch a huntsman that had ended up on the floor. I guess I was a bit slow and it started moving about, coming straight for me. Slightly freaky but I figured it was just a chance thing, so I backed off and it kept coming, then when I ran out of room I stepped over it to give some space. So it stopped, turned around, and started running toward me again.
That continued for some time, but it eventually stopped. I guess it was starting to wonder why that tree was so difficult to get to and wanted to have a think about it for a while. So I used that opportunity to catch it and relocate it. Again I don't think it was aggressive or anything, it wasn't showing any signs of aggression. Pretty sure if I'd let it reach me it would've just started scaling me, rather than trying to eat me. Still, I didn't particularly want it to do either.