I and many people like me love physical media, however it should only be used to install, then you can put it back in the case.
But then, what's the point? Other than this:
As a bonus, for a fair majority of the people in the world, their dvd drive or blu-ray drive is significantly faster than their net connection.
Possible, but not necessarily true, especially when you factor in the delay to get that physical media. An ideal application of Steam -- which I'd love if I had a Linux port -- would be to see a game I like, order it, go do something else for an hour or two, then play it. That gets the game to me a lot faster than ordering online, and takes a lot less of my time than driving to the store.
And that's on my crappy college connection -- 10 mbit, half duplex. At home I get 100 mbit full duplex (so figure less than 15 minutes -- can't even make it to the store and back before my game's downloaded), and Japan has higher speeds still. I'm basing this on my recent purchase of Mirror's Edge, which is 7-8 gigs -- there are all kinds of indie games which are relatively tiny. Sure, if you filled a DVD with a few hundred of em, it might be faster, but you couldn't play through them fast enough -- the second game would download before you were through the first level of the first one.
Also, if games would learn to deliver just enough to play -- as Steam did for the original Half-Life -- the delay gets a lot smaller. Even for modern games, as soon as I can start playing, it's going to take far longer for me to play through than for the game to download.
Maybe it's just not as concentrated? I don't think we really need a facebook of porn when pretty much any idiot with a camera and a decent body can make their own website.
Sure they can, many paid versions of Linux have come with closed source tools.
Might that word have something to do with it? Those aren't necessarily distributed under the same license as Ubuntu.
More importantly, which closed source tools? Specifically, do they come with firmware for your wireless card already? Last time I had to do this, I had to find the firmware inside the Windows driver -- I'm sure that if it was legal, that firmware would be available for download, if not already incorporated into the distro. Ubuntu does incorporate the nvidia drivers, for example.
I would require a check box for Ubuntu's own sake saying "This is closed source software, we can't support it, if you have problems with it contact the producer of your wireless card".
I'm pretty sure this is how they handle the nvidia drivers.
99% of commercial games don't even have a Linux version,
99% don't have a Mac version, either, but that didn't stop Valve from porting Steam and a bunch of their games to Mac, making it a viable platform.
You can't really rely on emulation either,
Bullshit. First, Wine isn't emulation, and second, there are several packages (including Cedega) which you can license and include in your game. If you're supporting it, that effectively is a port, and you can then selectively port anything you need to make it feel more "native".
Then again, a game is probably one of the easiest things to port. Anyone who's already made a Mac port already has an OpenGL renderer, so most of the work for a Linux port is done.
Then there are hundreds of different Linux distros and configurations
That's a conservative estimate, but there are just as many possible configurations of Windows. As others have pointed out, we've got games like Quake3 -- the original Quake3 binaries still work after 10 years, on pretty much any distro.
Also, just imagine the outcry about DRM and Valve not open sourcing Steam or it's games.
You mean, like the outcry about not getting an open-source World of Goo, or UT2004, or...? There are now four or five posts on this thread, all modded insightful, which have pointed out the hypothetical outcry. Is there a single person on Slashdot who's going to be more annoyed at a Linux port which still has DRM and isn't open source than no Linux port at all?
I mean, earth to odies, nearly all of Oracle's shit runs on Linux, and it's more proprietary and more expensive than anything Steam has. Maya runs on Linux. Pretty much everything Google does runs on Linux, and it's almost all proprietary, including their own proprietary extensions to Linux, and that's before we start talking about things like Android.
So I have to wonder where you're getting this...
The whole open source and everything-must-be-free mentality
...Stallman doesn't speak for me, and he doesn't speak for pretty much anyone running Ubuntu -- or they wouldn't be running Ubuntu, they'd be running GNewSense.
You can already read here on slashdot how some people refuse to use Steam because it might go down in 50 years.
Or in five. Or tomorrow. But these people exist on every platform, and I know plenty of Linux people who use things like GMail.
This thinking is 100x worse with Linux users.
Really? Can you point me to any posts to that effect on any of the other stories speculating about a Linux port of Steam? Can you give me a citation for the 100x number?
I don't have a problem with proprietary games. I would much rather have open source games, but I'd rather play a proprietary game on Linux than have to reboot and play an open source one. I prefer DRM-free, but I like Steam, and I own several Steam games on Windows -- I'd probably own several times as many if there was a native Linux port.
It's not a surprise, but it is disappointing, and I really, truly hope it was because of something boring like market research, and not because of delusions like yours.
I don't mind proprietary things, but nvidia is a special case -- it's a proprietary binary blob in my kernel which has been known to cause instability and crashes, and which the kernel devs can't do anything about.
I dual-boot, but there are very few new games that I'll buy for Windows anymore. I have a limited amount of time and money to spend on games, so it makes sense to focus on games with native Linux versions, as well as good solid indie titles in general.
It may not be a priority for Linux, but I'd guess I'm not alone in this.
Nothing can completely prevent piracy forever. That's not the point.
The point is, Steam at least presents a scenario where if we ignore all moral, legal, and financial reasons, and reduce it to the raw functionality, I'd prefer Steam.
By contrast, almost all other DRM schemes are exactly the opposite. I prefer not to have physical media which can be scratched, so even in its most refined form (console games) where a game can pretty much be treated as a physical object to be bought, sold, lended, rented, etc, I'd still rather have something I can download immediately, back up, and otherwise save from physical harm.
I don't know how many people think like me, but if you add the relative convenience of Steam (click, buy, download faster than a torrent, sometimes start playing before it's even done downloading, no searching for cracks, no worrying about viruses), I would guess there's a large swath of gamers who might consider piracy, but would rather use Steam.
It has been shown many times and it has been shown again: Web 1.0, with all of the glorious unreadable Perl stuff, neatly and cleanly defeats all this Ruby on Rails, gradients-and-rounded-corners, Twitter-compatible, "beta" Web 2.0 nonsense!
I can write that script much quicker and cleaner in Ruby. In nine lines, I might even be able to tweet the results, just to annoy you...
...or maybe Web 2.0 people should stop designing RESTful asynchronous JavaScript-compatible social-media APIs that are too easily abused. It's not that hard!
Agreed. It's actually quite easy to create a RESTFUL AJAX-compatible social-media API which isn't so easily abused.
(This was supposed to be a humorous post, but it's not really working today, is it?)
Unfortunately, aside from being "cute" for a beer or something, it could conceivably be used as evidence to show that you were in a certain place at a certain time. Exploits like these have to become pretty common before we can be reasonably sure a court will throw out the "evidence" that I checked in at the scene of the crime...
It shows what you can expect from their so-called free and open source browser.
What's "so-called" about a browser which is actually free, and actually open source?
In fact, everytime they do it, they prove GNU/FSF "nitpicking" about the FOSS and plain "open source".
The FSF wouldn't have a problem with this, as far as I can tell -- Chromium still is, and always was, Free as in Freedom. Paying Google $5 to list/host your extension hardly counts as making it proprietary -- if you like, you can still release source, and still allow anyone else to either pay the $5 or host it themselves.
I don't want to make anyone feel guilty or cheap but I personally know some open source developers who can't even afford a $30 external disk and development stops until they get some kind of donation.
Where in my post did I say anything about this being "cheap"? I don't care if it's $5, $50, or $5000 -- it's just a listing. If there's a really good extension which can't afford or doesn't want this service, they can always host it elsewhere. It's not as though that's entirely unprecedented -- PuTTY pretty much refuses to get a real domain, and they haven't suffered because of it.
I also have never once called GNU or the FSF "nitpicking" -- I understand why they do what they do, and while I don't agree that all software should be Free, I have no problem with their definition of Free. It seems you do, however -- remember, it's got nothing to do with price.
So, just who did you think you were responding to?
you don't need a WWDC subscription to write safari browser extensions and publish.
Nor do you need this $5 fee -- it's entirely possible to host a Chrome extension yourself, have it auto-update yourself, and all the mechanisms to do this are already in place.
I don't see any benefit to integrating email into the Wave system - I wouldn't want to interactively create an email message,
That's not the only thing you could do with it. Since it had extensions, you could easily embed, say, a map, a calendar meeting, or a survey into a Wave. Tools to embed these things into email are cumbersome, nonstandard, and not necessarily secure. Having the concept built-in has some advantages.
It's also useful in that if someone's not online, it can behave like email, much better than IM offline messages for the same purpose. But when someone is online, it simply and naturally flips to IM. It's nice that Google Talk is in Gmail, but it's not truly integrated -- I can't immediately continue an email conversation as IM, or vice versa.
One of the purposes of Wave was to unify the various means we have of communicating. You wouldn't need forums, Facebook, email, IM, IRC, mailing lists, etc, because Wave does it all.
Unfortunately, the implementation bogged down when we had too many messages, it wasn't nearly streamlined enough, and -- not entirely Google's fault, since there was enough of an API for people to do this -- but since it didn't wire into any of those systems, and since everyone wasn't trying it all at once (partly because of the semi-closed beta), you now had forums, Facebook, email, IM, IRC, mailing lists, Wave, etc, which isn't an improvement.
IPs work even when the DNS server is down/unreachable.
In this case, since the DNS server I'm using is in the device acting as my router, if it's down, it needs to come back up right now.
Oh, and you have to enter the IP of the DNS server.
I don't -- on IPv4, that's what DHCP is for. On IPv6, DHCP is still available, but it's got a stateless autoconf built in, which does provide DNS.
Of course, if I'm doing either of these, that's an additional point of failure. Currently, the same server -- even the same process (dnsmasq) -- provides both of these, and it's been stable and reliable, but again, if it's down, that's an immediate priority.
Since this is a home network, I don't have terribly many things which need to connect to each other which shouldn't be configured dynamically -- in particular, the convenience of not having to reconfigure my laptop based on which network it's plugged into is worth the risk of having to deal with DHCP not working, which hasn't happened yet.
I do still assign sane IPs, and I do still have some memorized, but I also have DNS configured to where I'm not sure I would notice or care if I flipped over to IPv6.
A typical university student needs a word processor, maybe a spreadsheet.
And a web browser, at the very least. The ability to watch videos online, while not strictly a requirement, could definitely be helpful. All of these bump the specs up a bit -- in particular, I don't care how efficient your browser is, that 200 mhz Pentium is going to be useless on the modern Internet.
A cheap Linux Netbook is fine for most students, but won't run Portal.
It depends on the netbook. It's not going to be fast, but it could work. It's certainly more likely to than most modern, commercial games they could've picked.
A lot of students these days have Macs, but I think there is now a Mac port of Portal - was there back when this decision was made?
I don't know, but as part of my computer science program, I get a Windows license. Boot Camp is free, if I had a Mac.
There isn't a Linux port of Portal, yet, but one is in the works. I don't really see it being more of an issue here than in other disciplines, where, for example, LabVIEW is still required. It's certainly better than if they had required an iPad game.
It is still not possible to buy Portal without invasive DRM (you need either Steam or an XBox),
For some value of "invasive". I'd prefer no DRM at all, but Steam is a fair compromise.
Add to all these considerations the fact that most schools have computer labs of some sort, and I don't really see this being a problem.
I hope it's an optional module...
From TFA, the course is required, but not all sections of it require Portal yet.
Not everyone has a computer, but I think it's fair to say that everyone who has a computer capable of the kinds of things needed for college these days, probably has a computer capable of playing Portal. Just crank the settings down.
Some people have never been exposed to WASD, but everyone knows how to read a book. Will people be expected to game to be culturally literate these days?
I'm not sure if that would be a bad thing, but it would be different.
That would not change with IPv6 because now only the VPN service is exposed (a.k.a. anyone can try to hack it), but the internal services can be unencrypted because if someone hacks the VPN then he most likely would be able to hack whatever services there are.
Yeah, I thought so, too, but you at least want authentication working properly. That goes doubly for localhost -- it amazes me how many people assume something's "secure" just because it's running on localhost, or behind a firewall -- hello local escalation.
I use a VPN currently, but I do so for two main reasons. First, it's nice to be able to route all traffic through a trusted network -- or at least, one relatively more trusted than whoever happens to be listening at my local coffee shop. And second, because it's way more convenient than forwarding ports, thanks to ipv4 and NAT.
Still, it should be possible to mandate ipsec, and I suspect that would solve this issue. Even better, some higher-end network cards can implement ipsec in hardware.
It looks like the designers of SIP were actually trying to make it impossible to use with NAT.
It's possible NAT wasn't a consideration at the time. Same goes for FTP -- again, there are kludges, but compare this to a protocol like Skype, which has to deliberately punch holes through NAT "firewalls" to connect two cooperating parties.
I would still try to use NAT with IPv6 (if iptables or other sotware supported it), if not for anything else then to have my internal IPs constant and independent of the ISP
What? Why are you using IPs directly? That's what DNS is for.
Currently, with IPv4, I create a dnsmasq rule for the mac address of each machine I want to connect, giving it a fixed IP (though dynamic) and a DNS entry. I'm not sure exactly how I'd do this with v6, but it should be possible.
One ISP I know, offers 3 IPs with its fastest plan. I honestly would not be able to make use of them, well, I could assign separate IPs for 2 of my computers, but that would not become more useful to me than just forwarding ports.
For SSH alone, it'd be useful to me. Since I only use SSH keys, my SSH connection is at least as secure as my VPN connection, so I see no particular reason to tunnel it.
NAT works well enough and IPv6 is not (currently) much better than IPv4+NAT.
It is better in that, at the very least, I do not want ISPs getting the same idea. Some already are. Basically, think of your entire setup becoming no longer viable because your ISP ran out of IPs, and has put you behind a NAT -- now you need to pay extra for some third-party service which has IPs in order to deal with this (for example, I've got a VPS with its own IP, I can run a VPN through that), or you need to pay extra to your ISP for a real IP.
Or, again, all these problems go away with IPv6.
And yes, you can do NAT. I just can't see a good reason to do that, other than that it's what you're used to.
But other than the conceptual "a IP for each machine", having it is not any better than NAT. Well, unless you are at risk of running out of ports...
Running out? Probably not. But why should I have to connect on a non-standard port?
It's a minor annoyance and a potential security issue to have to run a reverse proxy or specify a port, and expose cookies to all subdomains behind a given NAT. It's a much bigger headache to have to deal with multiple SSH ports routing to multiple target machines (known_hosts doesn't seem to play well with this), and another security risk to use SSH forwarding instead of just connecting directly.
And then there are the systems which actually can't work on a non-standard port. To get around that, you'd need a VPN -- but if you're willing to expose a VPN, why not just expose the machines?
It also means that we can't have truly distributed systems, again, without manually forwarding ports. We can get close with hacks like Skype, but that by nature requires a centralized coordinating party. Compare that to something like SIP, which could be difficult to forward, but becomes simple and natural by simply giving each phone in the house a real, Internet-addressable IP. Hit that IP, the phone rings.
There are kludges, sure. There's also IP-over-DNS to avoid having to pay for wireless in Starbuck's, but aside from saving a few bucks, I can't imagine why anyone would prefer the kludge.
if it weren't for the fact that it eats all your memory, and then craps itself when it doesn't have any memory left to eat, and runs like a snail. Seriously, I love Firefox, except that it just keeps getting worse and worse at garbage collection....
Which is, again, a property of the Javascript runtime.
Of course, it could be that there's crap coding going on elsewhere, but again, this is a speed boost for the entire browser.
I'd also ask what version you're running, but I don't really care, haven't really run Firefox seriously for awhile.
couple places I really do need a working Greasemonkey.
Are these places where Chrome's user script support doesn't work?
If so, have you tried Chrome's extension support? It took me about an afternoon from knowing nothing about Chrome extensions (but having a working knowledge of web development) to having a working adblocker.
Jesus, Internet, can you please latch on to something else?
Like sex, boobs, sex, pics or it didn't happen, sex, really perverted fucked-up shit that should never exist, sex, and lolcats?
The Internet seems to "latch on" to anything and everything. It's an expression of our collective psyches -- so if humans have latched onto anything, anywhere, the Internet has, too.
Slashdot Mirror
I and many people like me love physical media, however it should only be used to install, then you can put it back in the case.
But then, what's the point? Other than this:
As a bonus, for a fair majority of the people in the world, their dvd drive or blu-ray drive is significantly faster than their net connection.
Possible, but not necessarily true, especially when you factor in the delay to get that physical media. An ideal application of Steam -- which I'd love if I had a Linux port -- would be to see a game I like, order it, go do something else for an hour or two, then play it. That gets the game to me a lot faster than ordering online, and takes a lot less of my time than driving to the store.
And that's on my crappy college connection -- 10 mbit, half duplex. At home I get 100 mbit full duplex (so figure less than 15 minutes -- can't even make it to the store and back before my game's downloaded), and Japan has higher speeds still. I'm basing this on my recent purchase of Mirror's Edge, which is 7-8 gigs -- there are all kinds of indie games which are relatively tiny. Sure, if you filled a DVD with a few hundred of em, it might be faster, but you couldn't play through them fast enough -- the second game would download before you were through the first level of the first one.
Also, if games would learn to deliver just enough to play -- as Steam did for the original Half-Life -- the delay gets a lot smaller. Even for modern games, as soon as I can start playing, it's going to take far longer for me to play through than for the game to download.
Maybe. They might also take it as a good thing and use it in their "Oracle Unbreakable Linux".
Maybe it's just not as concentrated? I don't think we really need a facebook of porn when pretty much any idiot with a camera and a decent body can make their own website.
Sure they can, many paid versions of Linux have come with closed source tools.
Might that word have something to do with it? Those aren't necessarily distributed under the same license as Ubuntu.
More importantly, which closed source tools? Specifically, do they come with firmware for your wireless card already? Last time I had to do this, I had to find the firmware inside the Windows driver -- I'm sure that if it was legal, that firmware would be available for download, if not already incorporated into the distro. Ubuntu does incorporate the nvidia drivers, for example.
I would require a check box for Ubuntu's own sake saying "This is closed source software, we can't support it, if you have problems with it contact the producer of your wireless card".
I'm pretty sure this is how they handle the nvidia drivers.
99% of commercial games don't even have a Linux version,
99% don't have a Mac version, either, but that didn't stop Valve from porting Steam and a bunch of their games to Mac, making it a viable platform.
You can't really rely on emulation either,
Bullshit. First, Wine isn't emulation, and second, there are several packages (including Cedega) which you can license and include in your game. If you're supporting it, that effectively is a port, and you can then selectively port anything you need to make it feel more "native".
Then again, a game is probably one of the easiest things to port. Anyone who's already made a Mac port already has an OpenGL renderer, so most of the work for a Linux port is done.
Then there are hundreds of different Linux distros and configurations
That's a conservative estimate, but there are just as many possible configurations of Windows. As others have pointed out, we've got games like Quake3 -- the original Quake3 binaries still work after 10 years, on pretty much any distro.
Also, just imagine the outcry about DRM and Valve not open sourcing Steam or it's games.
You mean, like the outcry about not getting an open-source World of Goo, or UT2004, or...? There are now four or five posts on this thread, all modded insightful, which have pointed out the hypothetical outcry. Is there a single person on Slashdot who's going to be more annoyed at a Linux port which still has DRM and isn't open source than no Linux port at all?
I mean, earth to odies, nearly all of Oracle's shit runs on Linux, and it's more proprietary and more expensive than anything Steam has. Maya runs on Linux. Pretty much everything Google does runs on Linux, and it's almost all proprietary, including their own proprietary extensions to Linux, and that's before we start talking about things like Android.
So I have to wonder where you're getting this...
The whole open source and everything-must-be-free mentality
...Stallman doesn't speak for me, and he doesn't speak for pretty much anyone running Ubuntu -- or they wouldn't be running Ubuntu, they'd be running GNewSense.
You can already read here on slashdot how some people refuse to use Steam because it might go down in 50 years.
Or in five. Or tomorrow. But these people exist on every platform, and I know plenty of Linux people who use things like GMail.
This thinking is 100x worse with Linux users.
Really? Can you point me to any posts to that effect on any of the other stories speculating about a Linux port of Steam? Can you give me a citation for the 100x number?
I don't have a problem with proprietary games. I would much rather have open source games, but I'd rather play a proprietary game on Linux than have to reboot and play an open source one. I prefer DRM-free, but I like Steam, and I own several Steam games on Windows -- I'd probably own several times as many if there was a native Linux port.
It's not a surprise, but it is disappointing, and I really, truly hope it was because of something boring like market research, and not because of delusions like yours.
I don't mind proprietary things, but nvidia is a special case -- it's a proprietary binary blob in my kernel which has been known to cause instability and crashes, and which the kernel devs can't do anything about.
But sure, I'll buy a proprietary game.
I dual-boot, but there are very few new games that I'll buy for Windows anymore. I have a limited amount of time and money to spend on games, so it makes sense to focus on games with native Linux versions, as well as good solid indie titles in general.
It may not be a priority for Linux, but I'd guess I'm not alone in this.
Nothing can completely prevent piracy forever. That's not the point.
The point is, Steam at least presents a scenario where if we ignore all moral, legal, and financial reasons, and reduce it to the raw functionality, I'd prefer Steam.
By contrast, almost all other DRM schemes are exactly the opposite. I prefer not to have physical media which can be scratched, so even in its most refined form (console games) where a game can pretty much be treated as a physical object to be bought, sold, lended, rented, etc, I'd still rather have something I can download immediately, back up, and otherwise save from physical harm.
I don't know how many people think like me, but if you add the relative convenience of Steam (click, buy, download faster than a torrent, sometimes start playing before it's even done downloading, no searching for cracks, no worrying about viruses), I would guess there's a large swath of gamers who might consider piracy, but would rather use Steam.
"Alternative to DRM"? No, this is just another form of DRM.
I like what Steam offers. I think it's a fair trade. I'm still not going to call it something other than DRM.
You know what the "alternative" to DRM is? Not putting fucking DRM on your products!
It has been shown many times and it has been shown again: Web 1.0, with all of the glorious unreadable Perl stuff, neatly and cleanly defeats all this Ruby on Rails, gradients-and-rounded-corners, Twitter-compatible, "beta" Web 2.0 nonsense!
I can write that script much quicker and cleaner in Ruby. In nine lines, I might even be able to tweet the results, just to annoy you...
...or maybe Web 2.0 people should stop designing RESTful asynchronous JavaScript-compatible social-media APIs that are too easily abused. It's not that hard!
Agreed. It's actually quite easy to create a RESTFUL AJAX-compatible social-media API which isn't so easily abused.
(This was supposed to be a humorous post, but it's not really working today, is it?)
Nope.
Unfortunately, aside from being "cute" for a beer or something, it could conceivably be used as evidence to show that you were in a certain place at a certain time. Exploits like these have to become pretty common before we can be reasonably sure a court will throw out the "evidence" that I checked in at the scene of the crime...
It shows what you can expect from their so-called free and open source browser.
What's "so-called" about a browser which is actually free, and actually open source?
In fact, everytime they do it, they prove GNU/FSF "nitpicking" about the FOSS and plain "open source".
The FSF wouldn't have a problem with this, as far as I can tell -- Chromium still is, and always was, Free as in Freedom. Paying Google $5 to list/host your extension hardly counts as making it proprietary -- if you like, you can still release source, and still allow anyone else to either pay the $5 or host it themselves.
I don't want to make anyone feel guilty or cheap but I personally know some open source developers who can't even afford a $30 external disk and development stops until they get some kind of donation.
Where in my post did I say anything about this being "cheap"? I don't care if it's $5, $50, or $5000 -- it's just a listing. If there's a really good extension which can't afford or doesn't want this service, they can always host it elsewhere. It's not as though that's entirely unprecedented -- PuTTY pretty much refuses to get a real domain, and they haven't suffered because of it.
I also have never once called GNU or the FSF "nitpicking" -- I understand why they do what they do, and while I don't agree that all software should be Free, I have no problem with their definition of Free. It seems you do, however -- remember, it's got nothing to do with price.
So, just who did you think you were responding to?
you don't need a WWDC subscription to write safari browser extensions and publish.
Nor do you need this $5 fee -- it's entirely possible to host a Chrome extension yourself, have it auto-update yourself, and all the mechanisms to do this are already in place.
I don't see any benefit to integrating email into the Wave system - I wouldn't want to interactively create an email message,
That's not the only thing you could do with it. Since it had extensions, you could easily embed, say, a map, a calendar meeting, or a survey into a Wave. Tools to embed these things into email are cumbersome, nonstandard, and not necessarily secure. Having the concept built-in has some advantages.
It's also useful in that if someone's not online, it can behave like email, much better than IM offline messages for the same purpose. But when someone is online, it simply and naturally flips to IM. It's nice that Google Talk is in Gmail, but it's not truly integrated -- I can't immediately continue an email conversation as IM, or vice versa.
One of the purposes of Wave was to unify the various means we have of communicating. You wouldn't need forums, Facebook, email, IM, IRC, mailing lists, etc, because Wave does it all.
Unfortunately, the implementation bogged down when we had too many messages, it wasn't nearly streamlined enough, and -- not entirely Google's fault, since there was enough of an API for people to do this -- but since it didn't wire into any of those systems, and since everyone wasn't trying it all at once (partly because of the semi-closed beta), you now had forums, Facebook, email, IM, IRC, mailing lists, Wave, etc, which isn't an improvement.
IPs work even when the DNS server is down/unreachable.
In this case, since the DNS server I'm using is in the device acting as my router, if it's down, it needs to come back up right now.
Oh, and you have to enter the IP of the DNS server.
I don't -- on IPv4, that's what DHCP is for. On IPv6, DHCP is still available, but it's got a stateless autoconf built in, which does provide DNS.
Of course, if I'm doing either of these, that's an additional point of failure. Currently, the same server -- even the same process (dnsmasq) -- provides both of these, and it's been stable and reliable, but again, if it's down, that's an immediate priority.
Since this is a home network, I don't have terribly many things which need to connect to each other which shouldn't be configured dynamically -- in particular, the convenience of not having to reconfigure my laptop based on which network it's plugged into is worth the risk of having to deal with DHCP not working, which hasn't happened yet.
I do still assign sane IPs, and I do still have some memorized, but I also have DNS configured to where I'm not sure I would notice or care if I flipped over to IPv6.
A typical university student needs a word processor, maybe a spreadsheet.
And a web browser, at the very least. The ability to watch videos online, while not strictly a requirement, could definitely be helpful. All of these bump the specs up a bit -- in particular, I don't care how efficient your browser is, that 200 mhz Pentium is going to be useless on the modern Internet.
A cheap Linux Netbook is fine for most students, but won't run Portal.
It depends on the netbook. It's not going to be fast, but it could work. It's certainly more likely to than most modern, commercial games they could've picked.
A lot of students these days have Macs, but I think there is now a Mac port of Portal - was there back when this decision was made?
I don't know, but as part of my computer science program, I get a Windows license. Boot Camp is free, if I had a Mac.
There isn't a Linux port of Portal, yet, but one is in the works. I don't really see it being more of an issue here than in other disciplines, where, for example, LabVIEW is still required. It's certainly better than if they had required an iPad game.
It is still not possible to buy Portal without invasive DRM (you need either Steam or an XBox),
For some value of "invasive". I'd prefer no DRM at all, but Steam is a fair compromise.
Add to all these considerations the fact that most schools have computer labs of some sort, and I don't really see this being a problem.
I hope it's an optional module...
From TFA, the course is required, but not all sections of it require Portal yet.
Should be <AOE if Slashdot didn't eat that.
Not everyone has a computer, but I think it's fair to say that everyone who has a computer capable of the kinds of things needed for college these days, probably has a computer capable of playing Portal. Just crank the settings down.
I use Dvorak, you insensitive clod!
Some people have never been exposed to WASD, but everyone knows how to read a book. Will people be expected to game to be culturally literate these days?
I'm not sure if that would be a bad thing, but it would be different.
That would not change with IPv6 because now only the VPN service is exposed (a.k.a. anyone can try to hack it), but the internal services can be unencrypted because if someone hacks the VPN then he most likely would be able to hack whatever services there are.
Yeah, I thought so, too, but you at least want authentication working properly. That goes doubly for localhost -- it amazes me how many people assume something's "secure" just because it's running on localhost, or behind a firewall -- hello local escalation.
I use a VPN currently, but I do so for two main reasons. First, it's nice to be able to route all traffic through a trusted network -- or at least, one relatively more trusted than whoever happens to be listening at my local coffee shop. And second, because it's way more convenient than forwarding ports, thanks to ipv4 and NAT.
Still, it should be possible to mandate ipsec, and I suspect that would solve this issue. Even better, some higher-end network cards can implement ipsec in hardware.
It looks like the designers of SIP were actually trying to make it impossible to use with NAT.
It's possible NAT wasn't a consideration at the time. Same goes for FTP -- again, there are kludges, but compare this to a protocol like Skype, which has to deliberately punch holes through NAT "firewalls" to connect two cooperating parties.
I would still try to use NAT with IPv6 (if iptables or other sotware supported it), if not for anything else then to have my internal IPs constant and independent of the ISP
What? Why are you using IPs directly? That's what DNS is for.
Currently, with IPv4, I create a dnsmasq rule for the mac address of each machine I want to connect, giving it a fixed IP (though dynamic) and a DNS entry. I'm not sure exactly how I'd do this with v6, but it should be possible.
One ISP I know, offers 3 IPs with its fastest plan. I honestly would not be able to make use of them, well, I could assign separate IPs for 2 of my computers, but that would not become more useful to me than just forwarding ports.
For SSH alone, it'd be useful to me. Since I only use SSH keys, my SSH connection is at least as secure as my VPN connection, so I see no particular reason to tunnel it.
NAT works well enough and IPv6 is not (currently) much better than IPv4+NAT.
It is better in that, at the very least, I do not want ISPs getting the same idea. Some already are. Basically, think of your entire setup becoming no longer viable because your ISP ran out of IPs, and has put you behind a NAT -- now you need to pay extra for some third-party service which has IPs in order to deal with this (for example, I've got a VPS with its own IP, I can run a VPN through that), or you need to pay extra to your ISP for a real IP.
Or, again, all these problems go away with IPv6.
And yes, you can do NAT. I just can't see a good reason to do that, other than that it's what you're used to.
But other than the conceptual "a IP for each machine", having it is not any better than NAT. Well, unless you are at risk of running out of ports...
Running out? Probably not. But why should I have to connect on a non-standard port?
It's a minor annoyance and a potential security issue to have to run a reverse proxy or specify a port, and expose cookies to all subdomains behind a given NAT. It's a much bigger headache to have to deal with multiple SSH ports routing to multiple target machines (known_hosts doesn't seem to play well with this), and another security risk to use SSH forwarding instead of just connecting directly.
And then there are the systems which actually can't work on a non-standard port. To get around that, you'd need a VPN -- but if you're willing to expose a VPN, why not just expose the machines?
It also means that we can't have truly distributed systems, again, without manually forwarding ports. We can get close with hacks like Skype, but that by nature requires a centralized coordinating party. Compare that to something like SIP, which could be difficult to forward, but becomes simple and natural by simply giving each phone in the house a real, Internet-addressable IP. Hit that IP, the phone rings.
There are kludges, sure. There's also IP-over-DNS to avoid having to pay for wireless in Starbuck's, but aside from saving a few bucks, I can't imagine why anyone would prefer the kludge.
if it weren't for the fact that it eats all your memory, and then craps itself when it doesn't have any memory left to eat, and runs like a snail. Seriously, I love Firefox, except that it just keeps getting worse and worse at garbage collection....
Which is, again, a property of the Javascript runtime.
Of course, it could be that there's crap coding going on elsewhere, but again, this is a speed boost for the entire browser.
I'd also ask what version you're running, but I don't really care, haven't really run Firefox seriously for awhile.
couple places I really do need a working Greasemonkey.
Are these places where Chrome's user script support doesn't work?
If so, have you tried Chrome's extension support? It took me about an afternoon from knowing nothing about Chrome extensions (but having a working knowledge of web development) to having a working adblocker.
Jesus, Internet, can you please latch on to something else?
Like sex, boobs, sex, pics or it didn't happen, sex, really perverted fucked-up shit that should never exist, sex, and lolcats?
The Internet seems to "latch on" to anything and everything. It's an expression of our collective psyches -- so if humans have latched onto anything, anywhere, the Internet has, too.