But except for admiting it was a bug they pointed me to a workaround site that still didn't work 100%.
I don't know, between those two things, which would you really rather have -- the smug satisfaction of knowing you're right, and Linux has a bug? Or a workaround that might actually give you a working Linux?
I've seen extreme cases of Linux fanboyism, but that doesn't sound like one. That sounds like people trying to be genuinely helpful.
Swapfiles (or partitions, usually) are something the OS is designed specifically to be able to deal with, and are required by the fact that there's not actually enough RAM to accomplish the task. It's a hack around the fact that modern filesystems aren't designed for the scenario where you have terabytes of RAM and a UPS -- in other words, it's a hack around a software problem, rather than around a lack of hardware.
And, md/dmraid will do asynchronous writes.
You've got me there. Still, I call it a "hack" because I can't help but imagine it's not going to work as well as the system TFA describes -- and it does exactly the same thing.
This is all nice and well until someone misses the fact that a function which was expecting an array was expecting a specific kind of array.
That's an argument against duck typing, not necessarily against operator overloading.
At this point someone will interject "testing! testing!". Okay, but when using significant software written by third parties in python or Ruby or similar languages I've experienced several cases of exceptions being raised because one part of the software passed an unexpected object to another part. So I guess the testing did not catch that, now, did it?
I don't know. You haven't provided nearly enough information.
Were these exceptions raised in production? (One way testing catches problems like this is through exceptions.) Is it the kind of thing which was able to sleep for five years or so, and then make the production site blow up? Or is it something that you see with routine Q&A on a staging server?
How complete was the test coverage? Was a code coverage tool used? Which one?
Was the development driven by the tests, or were tests added as an afterthought?
I don't think one is better than the other in any kind of absolute sense (i.e. irrespective of context).
Fair enough.
Here's what I think: Strict typing is a specific kind of unit test. It's pervasive, and maybe helpful, but you need unit tests anyway.
And just as strict typing won't catch every kind of typing mistake you could make, unit tests won't catch every kind of bug you might have. However, strict typing also makes your code more verbose -- and there are other things about Java which also tend to make code verbose. Wasn't there a study done somewhere which showed that bugs per line of code stays constant across languages? That seems to be an argument for whichever language is the least verbose.
Arguing exceptions like MMOs, Linux machines and Macs doesn't really advance things.
Agreed. My point is that PCs could take it back. What's holding it back right now is Windows.
And MMOs will likely keep them from becoming irrelevant anytime soon, which is what a lot of people like to imply -- that consoles will completely replace PCs as a gaming platform.
Show me where I can buy World of Warcraft for a console.
Oh, and while I don't often bring this up, updating my drivers is:
apt-get update apt-get dist-upgrade reboot
That's all drivers, on the entire system. On OS X, it's even simpler: When your computer asks you if you want to update now, type your password and click OK. When it finishes, it asks you to reboot; click "reboot".
Introducing classes for all of the Java programmers who can't understand a Self-like language.
I do agree with you there -- we don't need classes. And if we did, we could implement them, in JavaScript.
Operator overloading? Great, now you can enjoy C++ style code, where left shift and print are the same command.
If you really, really want to, then yes. But that it can be abused doesn't strike me as a serious reason for not including it.
I had a problem awhile ago. It was developing for HD-DVD, meaning we didn't have full JavaScript -- meaning I couldn't, say, extend Array. Even if I could, I'm not sure how much better I could make this... I had created a control for displaying a scrolling menu of widgets of some kind. The point is, the control itself shouldn't care about what those widgets are, or even that it's operating on something that's actually an array. It would've been nice to let it work with an array for now, knowing I could always roll something that pretends to be an array later.
Instead, what I ended up doing was wrapping the Array in a monstrosity which contained methods like getValue(), setValue(), etc. It was hideously ugly, but it would tend to work, and would even automatically wrap an Array if needed.
But it's ugly hacks like that which drove me away from Java in the first place. I'd rather duck-type it properly, like I do in Ruby. If it claims to have a working [] operator, and has methods like size() or length(), either it's an array, or it's pretending to be one, so treat it like an array.
By your comment about performance being "slightly" slower, and that you "won't notice if you're running on modern hardware", I take it that this is the pseudo read/write mode provided by the kernel NTFS drivers -- that is, you can only read/write to a file, but you can't change any of its attributes, including size, and you can't create or destroy files.
In other words, it's a mode that's really only useful for creating disk images, for things like a Linux filesystem, or swap. Not really like umsdos at all.
I don't actually know this to be true, but I do know that ntfs-3g (which provides full read/write access) is slow enough that I think I would notice the difference if my root filesystem was mounted loopback off of it.
burning a DVD, to watch at a friend's place, or on a portable DVD player
loading onto your laptop hard drive, so you can leave the physical disc at home
playing it on non-HDCP-enabled TVs with no degradation
playing it on multiple TVs at once, within the same house (we have a TV room with a couch, and a kitchen)
taking a screenshot, and using it as your desktop
sampling an audio clip, and using it as a ringtone, or desktop sound effect, or person-specific IM alert
Any new use you think of which should be legal, but which hasn't yet been blessed by the MPAA
Skipping unskippable commercials
Skipping that retarded "You wouldn't steal a car" ad, especially if you actually did buy it
Oh, and by the way: The non-legit side doesn't need this. There have been torrents of Blu-Ray movies almost as long as there have been torrents of HD-DVD movies, and there's always an insider somewhere. The problem is that non-legit people only need to crack it once to seed that torrent, and no one else ever has to crack it again; what SlySoft is doing is catering to people who want to do things like what I listed above, and thus need to be able to rip their own disc reliably.
Personally, I'm probably waiting until it's as easy to do on Linux, with free software, so that I can just pop a disc into my "media server of some sort" and have it rip away. I want it to be as thoroughly cracked as DVD, so that I never have to check some "supported" list to make sure that the particular disc and edition I've got is crackable, or wake up to find that Slysoft has shut down and AnyDVD won't work anymore.
Erm... you seem to have just described a hack to achieve their solution; that is, a block-device-level solution. Except yours is worse; raid would attempt to write to both at once, right? The whole point here is to queue writes up (but keep writing) until power fails.
I thought I'd explained why I think there are problems with any block-level solution, though?
While I've brought up Steam several times in this discussion, I'm going to advise against it.
Absolutely provide a downloadable version, and maybe use something like Steam. But don't use Steam itself.
As a user, no, I can't take them with me anywhere I play. Sometimes, I play on Linux. Sometimes, on OS X. Support on each is limited, and is hacked together with things like Wine and Parallels.
Or digital distribution, period. The places where people have money to spend on games are also getting faster and better Internet. And yes, game data is getting bigger, but if you remember, Half-Life 2 was a little under a gigabyte -- and the original Half-Life is playable without downloading the whole game first.
So, your users can handle it. The only problem left is you managing all that bandwidth, and Amazon pretty much has you covered -- it'll cost you $0.18 per gig, flat rate, less if you let them be a torrent seed. So your raw material cost is $0.18.
Mentioning Reznor in TFA really clouds the issue, I think.
I don't think any gamers are claiming that games should be free. Sure, it's great when they are, but we understand (somewhat) that it costs money, and always will. Some of us even want to be game developers when we grow up. (I'm grown up, and I still want to be a game developer when I grow up.)
The more important point is, games should not come with copy protection. It is a waste of time and money, and won't significantly change the number of people who pay for your game.
Now, your point about Steam is a valid one, but I think the more important point about Steam isn't that it could be an anti-piracy tool, but that something like it, even if there was no actual DRM in place, would still be a reason to buy the game. With Steam, you get to hook into a community of people -- your friends are on the friends list, you can click on them to join a game, you can actually play online with them on any decent server (instead of having to maintain your own pirate server, and memorize the IP address or register a hostname)... All of this can be done with very basic username/password authentication, and that's really it.
Even if you could create your own pirate Steam network, it's still probably not going to catch on, again because of this network effect. WoW has been reverse-engineered, and there are multiple open source WoW server projects, and many instances of them live on the Internet. But the real thing still has some 10 million subscribers. That's subscribers, not one-time purchasers, so there's that much more motivation to pirate it, if it were that simple. WoW does use some DRM-like schemes locally, but that seems to be really geared more towards stopping cheating, rather than towards preventing piracy.
Either of those are valid ways of providing even indie games (Peggle, Darwinia, and others) without actually relying on any DRM at all. And it's not a shareware model -- Peggle looks like it could be shareware, and indeed, you get a limited version of it with the Orange Box. But there's also Half-Life 2 (and episodes), arguably one of the larger productions, and Portal, arguably one of the best games, ever.
The point is not to make the average user unable to pirate the game, or too afraid to. It's to make the average user not even want to pirate the game.
If you'll allow me my moment of 2 AM hyperbole, part of the reason Portal exists is that Valve is a game development company, not a DRM-development company. They spend resources on building games, not on locking them down.
If all it takes to get a free piece of software is downloading some file, you'll have many more people copying.
That is all it takes. That will always be all it takes.
Do everything you can without impeding upon the normal, legal user's experience.
Which means, do absolutely nothing. Even entering CD keys is kind of a pain in the ass.
Now, I'd suggest a middle ground -- CD keys are fine, network authentication is fine (either opportunistically, or require it exactly once), but spyware, driver/hardware examination/mutilation, and process blacklists are just going to be an endless arms race between you and the pirates, and will drive legitimate users away.
But presumably someone who pirates the game and plays it won't buy the game. That's not a bad argument.
It only makes sense if someone who can no longer pirate the game will buy the game. And that is a pretty bad argument, because I honestly don't see how that works, in most cases. Exactly what demographic are you targeting who will download a game, and surf around shady sites for a crack, but who would also be willing to buy the game if you made it just a little bit harder? And how, exactly, do you think you're going to make it that little bit harder to where there isn't a crack for it?
The first few bucks (say, on actually having a CD key) stop the 8-12 year olds who would just download it and play it.
Those 8-12 year olds likelyknow more about piracy than you do. They can and will just download it and play it, if it's at all popular.
The next large chunk of money (some online authentication) stops another class of people from just getting the iso and the crack and running it.
Except the iso and the crack will usually stop that online authentication.
The only thing which has been shown to work is to either not use copy protection, or not much copy protection; see just about any Linux port of a game, for some reason, they never port the copy protection, and you don't even need an ISO, it'll just play with no CD -- or to turn it into a service; see WoW.
Now, I'll buy the shareware argument, simply because these games probably aren't popular enough to be cracked -- although that doesn't always stop people; see Uplink (and later Darwinia).
But I'll tell you what -- I don't care about CD keys, and I don't care about online authentication. As long as you don't run a rootkit, or make me physically push in the CD, or do stupid things like check to make sure you're not running on a network share, I'm willing to jump through a couple of hoops. But your best bet is to make it as easy or easier than console games -- Steam being an example of this.
I believe that its illegal to let a crawlers run 'wild' on the internet in the first place(unless you have permission to do so). at least that was what i was told in my cs class where we made a web crawler.
Where was this class? I'd like to know what resumes to ignore...
And for your own sake, get out now.
That, or it's possible that you've misinterpreted something which was said -- maybe something like robots.txt -- which still doesn't say anything about the legality of it.
If i hack my friend's computer and install a VNC program on it and use VNC to use there computer to go to the 'porn' site. then they get there house searched!
And during the search, they'd likely discover that you'd "hacked" their computer.
But it seems like it'd be even easier -- just troll them. Post the link anonymously, just like you might Rick Roll or Goatse.
But my point was, rather, that I want to be able to read my own documents for the rest of my own lifetime, and I want them to be readable for arbitrarily long after that.
I'm assuming, therefore, that my documents may outlast Microsoft.
I think comparing the router to a kid is a bad analogy. Try that analogy with a dog.
Actually, try that analogy with the lock on the front door.
As I understand it, the most you can do is, if you find someone in your house, you can kick them out -- and if they don't leave when asked, then they are tresspassing.
Except it's more than that -- your router is broadcasting that it's open. If your dog could actually, literally invite people in, rather than simply allow them in, I think that defense should hold.
Slashdot Mirror
I don't know, between those two things, which would you really rather have -- the smug satisfaction of knowing you're right, and Linux has a bug? Or a workaround that might actually give you a working Linux?
I've seen extreme cases of Linux fanboyism, but that doesn't sound like one. That sounds like people trying to be genuinely helpful.
RTFA.
Imagine you've already got the 8 megs, and you still want things to happen in realtime. Oh, and you want to actually have drivers.
Suddenly, things like VxWorks are going to cost you in licensing fees and development time, in return for... what, exactly?
Swapfiles (or partitions, usually) are something the OS is designed specifically to be able to deal with, and are required by the fact that there's not actually enough RAM to accomplish the task. It's a hack around the fact that modern filesystems aren't designed for the scenario where you have terabytes of RAM and a UPS -- in other words, it's a hack around a software problem, rather than around a lack of hardware.
You've got me there. Still, I call it a "hack" because I can't help but imagine it's not going to work as well as the system TFA describes -- and it does exactly the same thing.
All well and good, and irrelevant.
I was talking about the NTFS drive. Do you have full read/write access to the Windows drive, too? Or just to your Linux partition/file?
Either way, you have to have full read/write access to the Linux partition/file, that's a given. The question is what NTFS driver is used.
That's an argument against duck typing, not necessarily against operator overloading.
I don't know. You haven't provided nearly enough information.
Were these exceptions raised in production? (One way testing catches problems like this is through exceptions.) Is it the kind of thing which was able to sleep for five years or so, and then make the production site blow up? Or is it something that you see with routine Q&A on a staging server?
How complete was the test coverage? Was a code coverage tool used? Which one?
Was the development driven by the tests, or were tests added as an afterthought?
Fair enough.
Here's what I think: Strict typing is a specific kind of unit test. It's pervasive, and maybe helpful, but you need unit tests anyway.
And just as strict typing won't catch every kind of typing mistake you could make, unit tests won't catch every kind of bug you might have. However, strict typing also makes your code more verbose -- and there are other things about Java which also tend to make code verbose. Wasn't there a study done somewhere which showed that bugs per line of code stays constant across languages? That seems to be an argument for whichever language is the least verbose.
Agreed. My point is that PCs could take it back. What's holding it back right now is Windows.
And MMOs will likely keep them from becoming irrelevant anytime soon, which is what a lot of people like to imply -- that consoles will completely replace PCs as a gaming platform.
Show me where I can buy World of Warcraft for a console.
Oh, and while I don't often bring this up, updating my drivers is:
That's all drivers, on the entire system. On OS X, it's even simpler: When your computer asks you if you want to update now, type your password and click OK. When it finishes, it asks you to reboot; click "reboot".
Only on Windows is this kind of thing a chore.
I do agree with you there -- we don't need classes. And if we did, we could implement them, in JavaScript.
If you really, really want to, then yes. But that it can be abused doesn't strike me as a serious reason for not including it.
I had a problem awhile ago. It was developing for HD-DVD, meaning we didn't have full JavaScript -- meaning I couldn't, say, extend Array. Even if I could, I'm not sure how much better I could make this... I had created a control for displaying a scrolling menu of widgets of some kind. The point is, the control itself shouldn't care about what those widgets are, or even that it's operating on something that's actually an array. It would've been nice to let it work with an array for now, knowing I could always roll something that pretends to be an array later.
Instead, what I ended up doing was wrapping the Array in a monstrosity which contained methods like getValue(), setValue(), etc. It was hideously ugly, but it would tend to work, and would even automatically wrap an Array if needed.
But it's ugly hacks like that which drove me away from Java in the first place. I'd rather duck-type it properly, like I do in Ruby. If it claims to have a working [] operator, and has methods like size() or length(), either it's an array, or it's pretending to be one, so treat it like an array.
There's also example.com.
If it ever actually existed, you might have a shot at buying it here.
Disclaimer: I used to work for that company.
What, specifically, makes you think that they are not trying to make a better Linux for Linux users?
By your comment about performance being "slightly" slower, and that you "won't notice if you're running on modern hardware", I take it that this is the pseudo read/write mode provided by the kernel NTFS drivers -- that is, you can only read/write to a file, but you can't change any of its attributes, including size, and you can't create or destroy files.
In other words, it's a mode that's really only useful for creating disk images, for things like a Linux filesystem, or swap. Not really like umsdos at all.
I don't actually know this to be true, but I do know that ntfs-3g (which provides full read/write access) is slow enough that I think I would notice the difference if my root filesystem was mounted loopback off of it.
There are all kinds of legit reasons:
Oh, and by the way: The non-legit side doesn't need this. There have been torrents of Blu-Ray movies almost as long as there have been torrents of HD-DVD movies, and there's always an insider somewhere. The problem is that non-legit people only need to crack it once to seed that torrent, and no one else ever has to crack it again; what SlySoft is doing is catering to people who want to do things like what I listed above, and thus need to be able to rip their own disc reliably.
Personally, I'm probably waiting until it's as easy to do on Linux, with free software, so that I can just pop a disc into my "media server of some sort" and have it rip away. I want it to be as thoroughly cracked as DVD, so that I never have to check some "supported" list to make sure that the particular disc and edition I've got is crackable, or wake up to find that Slysoft has shut down and AnyDVD won't work anymore.
Erm... you seem to have just described a hack to achieve their solution; that is, a block-device-level solution. Except yours is worse; raid would attempt to write to both at once, right? The whole point here is to queue writes up (but keep writing) until power fails.
I thought I'd explained why I think there are problems with any block-level solution, though?
While I've brought up Steam several times in this discussion, I'm going to advise against it.
Absolutely provide a downloadable version, and maybe use something like Steam. But don't use Steam itself.
As a user, no, I can't take them with me anywhere I play. Sometimes, I play on Linux. Sometimes, on OS X. Support on each is limited, and is hacked together with things like Wine and Parallels.
Or digital distribution, period. The places where people have money to spend on games are also getting faster and better Internet. And yes, game data is getting bigger, but if you remember, Half-Life 2 was a little under a gigabyte -- and the original Half-Life is playable without downloading the whole game first.
So, your users can handle it. The only problem left is you managing all that bandwidth, and Amazon pretty much has you covered -- it'll cost you $0.18 per gig, flat rate, less if you let them be a torrent seed. So your raw material cost is $0.18.
Mentioning Reznor in TFA really clouds the issue, I think.
I don't think any gamers are claiming that games should be free. Sure, it's great when they are, but we understand (somewhat) that it costs money, and always will. Some of us even want to be game developers when we grow up. (I'm grown up, and I still want to be a game developer when I grow up.)
The more important point is, games should not come with copy protection. It is a waste of time and money, and won't significantly change the number of people who pay for your game.
Now, your point about Steam is a valid one, but I think the more important point about Steam isn't that it could be an anti-piracy tool, but that something like it, even if there was no actual DRM in place, would still be a reason to buy the game. With Steam, you get to hook into a community of people -- your friends are on the friends list, you can click on them to join a game, you can actually play online with them on any decent server (instead of having to maintain your own pirate server, and memorize the IP address or register a hostname)... All of this can be done with very basic username/password authentication, and that's really it.
Even if you could create your own pirate Steam network, it's still probably not going to catch on, again because of this network effect. WoW has been reverse-engineered, and there are multiple open source WoW server projects, and many instances of them live on the Internet. But the real thing still has some 10 million subscribers. That's subscribers, not one-time purchasers, so there's that much more motivation to pirate it, if it were that simple. WoW does use some DRM-like schemes locally, but that seems to be really geared more towards stopping cheating, rather than towards preventing piracy.
Either of those are valid ways of providing even indie games (Peggle, Darwinia, and others) without actually relying on any DRM at all. And it's not a shareware model -- Peggle looks like it could be shareware, and indeed, you get a limited version of it with the Orange Box. But there's also Half-Life 2 (and episodes), arguably one of the larger productions, and Portal, arguably one of the best games, ever.
The point is not to make the average user unable to pirate the game, or too afraid to. It's to make the average user not even want to pirate the game.
If you'll allow me my moment of 2 AM hyperbole, part of the reason Portal exists is that Valve is a game development company, not a DRM-development company. They spend resources on building games, not on locking them down.
That is all it takes. That will always be all it takes.
Which means, do absolutely nothing. Even entering CD keys is kind of a pain in the ass.
Now, I'd suggest a middle ground -- CD keys are fine, network authentication is fine (either opportunistically, or require it exactly once), but spyware, driver/hardware examination/mutilation, and process blacklists are just going to be an endless arms race between you and the pirates, and will drive legitimate users away.
It only makes sense if someone who can no longer pirate the game will buy the game. And that is a pretty bad argument, because I honestly don't see how that works, in most cases. Exactly what demographic are you targeting who will download a game, and surf around shady sites for a crack, but who would also be willing to buy the game if you made it just a little bit harder? And how, exactly, do you think you're going to make it that little bit harder to where there isn't a crack for it?
Those 8-12 year olds likelyknow more about piracy than you do. They can and will just download it and play it, if it's at all popular.
Except the iso and the crack will usually stop that online authentication.
The only thing which has been shown to work is to either not use copy protection, or not much copy protection; see just about any Linux port of a game, for some reason, they never port the copy protection, and you don't even need an ISO, it'll just play with no CD -- or to turn it into a service; see WoW.
Now, I'll buy the shareware argument, simply because these games probably aren't popular enough to be cracked -- although that doesn't always stop people; see Uplink (and later Darwinia).
But I'll tell you what -- I don't care about CD keys, and I don't care about online authentication. As long as you don't run a rootkit, or make me physically push in the CD, or do stupid things like check to make sure you're not running on a network share, I'm willing to jump through a couple of hoops. But your best bet is to make it as easy or easier than console games -- Steam being an example of this.
Well, Microsoft ignores piracy, but there's still the legal pressure. So it's not a Reznor thing...
But I think this would work much better for Apple than it would for Microsoft. Specifically:
Fixed it for you.
Not that it makes any sense, and not that they'd ever do that, but I think it might make money.
Where was this class? I'd like to know what resumes to ignore...
And for your own sake, get out now.
That, or it's possible that you've misinterpreted something which was said -- maybe something like robots.txt -- which still doesn't say anything about the legality of it.
And during the search, they'd likely discover that you'd "hacked" their computer.
But it seems like it'd be even easier -- just troll them. Post the link anonymously, just like you might Rick Roll or Goatse.
Medical science may well catch up to that.
But my point was, rather, that I want to be able to read my own documents for the rest of my own lifetime, and I want them to be readable for arbitrarily long after that.
I'm assuming, therefore, that my documents may outlast Microsoft.
Oh, to clarify...
Are you talking about artists or customers here?
Labels won't. Artists might -- after all, look at Radiohead and NiN. Customers will, for the convenience.
Isn't this also true if you simply lock the screen?
Actually, try that analogy with the lock on the front door.
As I understand it, the most you can do is, if you find someone in your house, you can kick them out -- and if they don't leave when asked, then they are tresspassing.
Except it's more than that -- your router is broadcasting that it's open. If your dog could actually, literally invite people in, rather than simply allow them in, I think that defense should hold.