The bottom line: Do you trust Microsoft? That's ultimately what this is all about.
I don't understand what it is about these technologies and their evangelists that makes it so easy for them to wooll over listeners and analysts eyes. I mean, the author of the article quotes Stallman's and Sulzberger's comments, but they seem to go in one ear and out the rest.
This isn't about whether one trusts Microsoft. People who dislike Palladium and TC are not tinfoil hatters who think that once it is deployed Microsoft will use it to take over the world, or whatever. The bottom line is exactly what Sulzberger says: How much control should users have over their own systems.
Microsoft's representative covers this up in invented technical terms, and talks about "security" and "trust" because those words sound good to the uninitiated, but that is just a smokescreen for the true neature (not a lie - they are upfront about what the system includes, they just spin it so people Chris Preimesberger will miss the point).
The point is this: every piece of "security" and "trust" that can be gained from Palladium is gained by palladium taking away from the user control of his own computer. Once that control is removed, ISPs can "secure" and "trust" that the user has his system configured as they mandate (see the Cisco router story). Microsoft can "secure" and "trust" that their software is licensed and registered. The record companies can "secure" and "trust" that their songs cannot be copied, ALL BECAUSE ULTIMATELY THE COMPUTER, NOT THE USER, IS IN CONTROL!
The question he asked "Does Microsoft have a back door" is stupid. Nobody serious believes that Palladium contains a backdoor so that MS can take over the computer. They believe the point with Palladium's design is that software can be installed with restrictions that the user cannot circumvent, and that people will be forced into installing such software, hostile to themselves, on their own PCs, in order to exchange data and connect to the Internet.
The reported responses from the MS representative give us absolutely no reason to answer "no" to either of Sulzberger's questions, even though the article claims so. In fact, when MS say things like, "We are building a scalable, distributed credential-based security model here," and list features of "attestations with authenticated code that is affiliated with only that particular process" - that is exactly what Sulzberger and Stallman are talking about. The Palladium computer will attest - BEYOND THE USERS CONTROL - whether the computer is running software that is "trusted" by the counterpart and hostile to the user, exactly so that the counterpart can mandate the use of such software (read DRM).
The fact that Microsoft tell us that the code will be open for review gives absolutely no confort. It is not the code, but the very concept of Palladium that is frightening beyond belief. Apparently Microsoft have nothing to fear regarding being open about it, as for some reason so many people cannot seem the grasp the point that Stallman, Sulzberger, and myself scream into the void!
Please make this comment visible. There is no story here!
Re:30,000 km/s can do a lot of damage
on
ISS Fender Bender
·
· Score: 1
The earths radius is about 6,360 km. The ISS altitude varies, but is around 340-380 km mean most of the time. This gives an orbital distance of ca. 42,500 km (I know, orbits aren't circles, but I'm not being precise).
If the ISS were traveling 30,000 km/s it would circle the earth in 1.5 seconds.
The objects you use and have own you just as much as you own them. Don't believe me? Try giving away your beloved (I'm as guilty as the rest of us on this) computer for a month. It's hard.
Bullshit logic: How does the latter imply the former? Try going a month without shoes. That is hard too, does it mean that your shoes own you? Try going without a fork. Try going without a watch. Your computer doesn't own you just because it sucks to not have access to it. It is just a useful tool that makes life better (like shoes, forks, and watches).
These people, while extremely knowledgeable, don't relate to the average user. They don't see the problem with having to track down DLL files and source files needed for compilation of EVERY interesting program--they're used to it. They don't understand that most people don't want to deal with the command line unless they have to. They can't see why people would want to be able to just copy and paste files of a program from a computer with one linux distribution and have it work on another distribution.
Bullshit generalisation! People understand these things very well, I think you would have to search long and hard to find a programmer with the views you ascribe to everybody. The reason that the GUIs on many Linux programs aren't perfect is simply that coding good GUIs takes time and effort, and the people who wrote the software (often driven by their own needs) didn't invest that time and effort. It's not like understanding that people want easy to learn GUIs suddenly makes every program you write newbie-friendly.
The problem lies deeper than just ease of use like most people think. The problem is, while C++, PERL, bash scripting, security patching, and command line usage, are commonplace in the geek community, they are (relatively, when including those people of the "real world") a rather obscure and silly set of skills that are--unlike Windows--NECESSARY for proper usage of linux.
Bullshit lies. I couldn't write C++ to save my life, hardly know any perl, and know just enough bash to write a loop. I patch security updates by starting synaptic and pressing the upgrade button, which is easier than Windows Update. The only one of these skills I guess I have is CLI usage, but from what I have seen one could use (though possibly not configure completely) a Linux box without ever touching the CLI.
The command line is a big no-no for the general public. If you tell an ordinary user they got to go to the command line for anything... even to type the word "install", they're just gonna laugh at you.
But then, programs like synaptic and aptitude mean you don't have to use the command line to install software from apt. I mean, compared to packaging problem, how difficult is a GUI fronend?
The real problem with apt is that it depends on a single centrally managed collection of software. If you keep adding new apt sources for every package you install, things get screwed up very fast.
But note that since no actually distribution needs to have been made, it will be sufficient to have downloaded a copy of SW Ep III two days before it is released, and have it end up in the shared folder of the P2P client (or not even that as many clients like eDonkey and Overnet share files while they are being downloaded).
Anyone want to see the first ten year old prosecuted under this law?
If I say, "motorcycles are ready for the road," will people think that means that everybody who uses a car should be able to get on a motorcycle and start riding around? Does it mean that people who have delicate haircuts, or who use their cars to transport people and cargo, will become angry and dissapointed.
Linux can compete with Windows in it's market. I used to be in windows market, but am not any more because Linux provided a better desktop option for me. I bet 9/10 of the Linux users out there were taken from the Windows market.
"Linux is ready for desktop use. It may or may not be able to provide your desktop needs." I don't think most people will have any trouble understanding that.
I read this and found out that X has a completely adequate clipboard - it is just that applications too often clobber it. Apparently the latest version of pure desktops do it right though.
(I always thought that ctrl-c ctrl-v in X apps was just an ugly hack using the same selection as select and middle-click (apparently because old KDE did it wrong).)
That may be how windows explorer handles cut and paste of file icons, but AFAIK data copied by highlighting the actual text, image, etc is actually copied into a clipboard (how else can it survive after the application closes?)
In the case of my proposal, you simply define a "file location" type of data in.clipboard directory, for use by file managers, where the information written in.clipboard is the identity of the file, not the file to copy itself.
I am using Suse 8.2 with KDE and I haven't noticed and copy/past issues. I can easily copy and paste between Opera, blufish, kwrite, openoffice, etc...
But you can't copy between for example OpenOffice and mozilla composer and keep the font and layout settings. Nor can you (most of the time) copy images or sounds or other multimedia between different programs. Closing applications typically means loosing what is copied, and it is easy to exidentally replace the buffer by highlighting something inadvertly.
I use X exclusively on the Desktop (Linux at home, Solaris at work) and I'm completely happy with highlight and middle click: but it does not match the cut-n-paste expectations of people moving from Windows and Mac.
It needs to be networked, like X, but have a standard widget set and clipboard.
I'm not so sure that the clipboard should be a part of X, or it's equivalent. Wouldn't it be even nice if cut-n-paste was a system wide thing?
I think one of the biggest mistakes in Linux usability has been the mistaking of X's built in text drag-n-drop capability (which is what the highlight and middleclick is, not cut-n-paste) with a clipboard in the Windows or Mac sense.
Why couldn't we make a.clipboard directory in the user directory instead? (or maybe/var/clipboard/userfoo/ ?) Cutting or copying something from an application would move the data into a file in this directory, and it would be possible to have more than one thing in the clipboard at once (possibly this should be a power-user feature). The files in the.clipboard should be in standardized fromats for each type of data, so we could support cut-n-paste between applications of more than just text. Cut and paste would be possible in the console, and the clipboard could survive crashes and reboots.
Since we have good free file-fromats for most types of data, and libraries to handle it, implementing this doesn't seem so difficult. Is there any movement to have something like this happen?
1) It needs better CD burning software (any opinions on which is best?) It may have this already and I overlooked it...
K3B. gToaster. Burn.app.
2) It needs A pre-configured link to download a p2p system like Kazaa so kids can download music if they want. (it's what the consumers want, lets face it.)
But do the parents want a $2000 settlement with the RIAA?
3) It needs plug-n-play. I heard rumors about mandrake having it, anyone tried it though? If I plug in my USB-pen, it better find it, and open it as a folder! 2000/XP is the shit in this respect.
Mandrake, and especially Knoppix have had next to perfect PnP for years. I have used Knoppix on windows machines to identify hardware it couldn't recognize.
When people say that linux is not ready for the desktop, they mean that the average user will not be able to use it.
What do you mean exactly by "the average user"? Clearly the average user, as in the user who is the average of all users in every respect doesn't exist, and people even closely aproximating him are pretty rare.
Do you mean the median user? (That is to say, when 50% of the people can use it.) Or do you mean a higher percentile? Or is 100% necessary? (Including illiterate people?)
You cant say, it is erady, but only for the geeks desktops. That kinda defeats the purpose.
Why not? Why does it have to black and white? Why can't an OS be ready for some peoples desktops and not for others? If Linux has 2.6% of the market today, would going to 15% not be an improvement? Would those 15% be using an OS that wasn't "ready" simply because it was unsuitable for the other 85%?
Your thinking makes no sense to me. Linux is ready for the desktop, and has been for some time. It is not ready for everybody's desktop, but it is ready for more users today than it was two years ago, and will be ready for even more in another two years. I see no contradiction.
I got scared by a couple of posts, especially the one I responded to, which was why responded harshly. If they were joking, then I apologize, IHBT.
You are right that my math skills are medicore (I haven't gotten anywhere on my dissertation in the last couple of months) but that doesn't really matter we are talking Teletubby level...
Well, you're already in bed with the evil phone company... so what's the difference?
Someone needs to just run fiber to everyone's house/business and put all these bozos out of business.
What makes you think that wouldn't end up being the evil fiber company?
Wiring peoples houses is conductive to natural monopolies. Some part of me can't help but think it might be better off as public infrastructure (a la roads), but then I think of how much I would be paying to wire all the people who have chosen to live in rural areas in that case...
Perhaps wireless is the ticket (there is a company two houses down from mine that sells 802.11b broadband - unfortunately they pointed there directional anntenna in the other direction...)
Don't post math comments until you pass pre-calculus. Please.
Fuck baby jesus. The level of this discussion is making me cry.
sqrt(x) is a monotonically increasing function. If a > b then sqrt(a) > sqrt(b). This isn't even highschool math, for Christs sake. That sqrt(1)=1 does not mean that the function isn't increasing at that point (consider the function f(x) = x).
And for the record sqrt(x) -> infinity as x -> infinity. This is also bloody fucking obvious (if it wasn't true, there would have to be numbers so large they had no square). 1/f(x) -> 0 as f(x) -> infinity, Xeno's paradox has not been a problem for at least 400 years.
Just responding to this thread has made me lose half my IQ...
In theory somebody out there could write a Back Orifice style program but register the port with UPNP. This will allow external attackers to tunnel through the firewall as if it wasn't there.
Please! If the trojan is already installed, then getting around the firewall is trivial. It can simply connect out, either to cracked server somewhere, to an IRC channel (this how most trojans are controlled), or even to a webpage (say for instance an empty discussion here on Slashdot). If you are blocking outgoing ports, then it just makes sure that it uses port 80. If you have a software firewall, then it communicates by using Windows scripting to invoke IE to open the pages on it's behalf.
Firewalls are good for keeping things out. Once something is already on your network, the firewall is 100% useless, UPNP or not.
Obviously UPNP is only available to hosts INSIDE the router.
This part of UPNP is a GOOD thing. It helps against the growing problem of the "the firewalled consumer" that thanks to NAT has created an A and B class of Internet citizens: those that can be connected to, and those that can't.
A good example of this is VoIP. VoIP phones are being sold that connect straight to the home router, and allow calling over the Internet. But 99% of all home routers do NAT, so without port forwards it will be impossible for two such phones to talk to one another. Since users can't really be expected to set this up manually, UPNP is a necessity. I think it is fantastic that routers are starting to support this!
I agree that it ought to be more configurable (which ports it will allow forwards to, etc) but in the end we cannot secure things by neutering the entire Internet. You shouldn't be running insecure services (consider that P2P applications like IM can be cracked straight through a NAT: they connect back to others at the request of the central server. and those back connections are just as vulnerable to buffer overflows and incoming (there is a worm for you!)).
"From these variations, he could calculate the frequencies of the sound waves propagating through the Universe during its first 760,000 years, when it was just 18 million light years across."
Right after the big bang in astronomical time, not human time.
Slashdot Mirror
I won't believe this. This story is FUD. FUD FUD FUD. FUD FUD FUD FUD FUD FUD FUD.
Calm down now, Steve. You know that the heart doctor told you about getting so worked up.
Yes, even about developers...
Sorry to be offtopic, but what did the deleted post contain? I must have missed this entirely.
The bottom line: Do you trust Microsoft? That's ultimately what this is all about.
I don't understand what it is about these technologies and their evangelists that makes it so easy for them to wooll over listeners and analysts eyes. I mean, the author of the article quotes Stallman's and Sulzberger's comments, but they seem to go in one ear and out the rest.
This isn't about whether one trusts Microsoft. People who dislike Palladium and TC are not tinfoil hatters who think that once it is deployed Microsoft will use it to take over the world, or whatever. The bottom line is exactly what Sulzberger says: How much control should users have over their own systems.
Microsoft's representative covers this up in invented technical terms, and talks about "security" and "trust" because those words sound good to the uninitiated, but that is just a smokescreen for the true neature (not a lie - they are upfront about what the system includes, they just spin it so people Chris Preimesberger will miss the point).
The point is this: every piece of "security" and "trust" that can be gained from Palladium is gained by palladium taking away from the user control of his own computer. Once that control is removed, ISPs can "secure" and "trust" that the user has his system configured as they mandate (see the Cisco router story). Microsoft can "secure" and "trust" that their software is licensed and registered. The record companies can "secure" and "trust" that their songs cannot be copied, ALL BECAUSE ULTIMATELY THE COMPUTER, NOT THE USER, IS IN CONTROL!
The question he asked "Does Microsoft have a back door" is stupid. Nobody serious believes that Palladium contains a backdoor so that MS can take over the computer. They believe the point with Palladium's design is that software can be installed with restrictions that the user cannot circumvent, and that people will be forced into installing such software, hostile to themselves, on their own PCs, in order to exchange data and connect to the Internet.
The reported responses from the MS representative give us absolutely no reason to answer "no" to either of Sulzberger's questions, even though the article claims so. In fact, when MS say things like, "We are building a scalable, distributed credential-based security model here," and list features of "attestations with authenticated code that is affiliated with only that particular process" - that is exactly what Sulzberger and Stallman are talking about. The Palladium computer will attest - BEYOND THE USERS CONTROL - whether the computer is running software that is "trusted" by the counterpart and hostile to the user, exactly so that the counterpart can mandate the use of such software (read DRM).
The fact that Microsoft tell us that the code will be open for review gives absolutely no confort. It is not the code, but the very concept of Palladium that is frightening beyond belief. Apparently Microsoft have nothing to fear regarding being open about it, as for some reason so many people cannot seem the grasp the point that Stallman, Sulzberger, and myself scream into the void!
More accurately it means:
"People who don't trust you can trust your computer to control you."
Please make this comment visible. There is no story here!
The earths radius is about 6,360 km. The ISS altitude varies, but is around 340-380 km mean most of the time. This gives an orbital distance of ca. 42,500 km (I know, orbits aren't circles, but I'm not being precise).
If the ISS were traveling 30,000 km/s it would circle the earth in 1.5 seconds.
So yes: he is way off.
I'll get moderated down for this!
The objects you use and have own you just as much as you own them. Don't believe me? Try giving away your beloved (I'm as guilty as the rest of us on this) computer for a month. It's hard.
Bullshit logic: How does the latter imply the former? Try going a month without shoes. That is hard too, does it mean that your shoes own you? Try going without a fork. Try going without a watch. Your computer doesn't own you just because it sucks to not have access to it. It is just a useful tool that makes life better (like shoes, forks, and watches).
These people, while extremely knowledgeable, don't relate to the average user. They don't see the problem with having to track down DLL files and source files needed for compilation of EVERY interesting program--they're used to it. They don't understand that most people don't want to deal with the command line unless they have to. They can't see why people would want to be able to just copy and paste files of a program from a computer with one linux distribution and have it work on another distribution.
Bullshit generalisation! People understand these things very well, I think you would have to search long and hard to find a programmer with the views you ascribe to everybody. The reason that the GUIs on many Linux programs aren't perfect is simply that coding good GUIs takes time and effort, and the people who wrote the software (often driven by their own needs) didn't invest that time and effort. It's not like understanding that people want easy to learn GUIs suddenly makes every program you write newbie-friendly.
The problem lies deeper than just ease of use like most people think. The problem is, while C++, PERL, bash scripting, security patching, and command line usage, are commonplace in the geek community, they are (relatively, when including those people of the "real world") a rather obscure and silly set of skills that are--unlike Windows--NECESSARY for proper usage of linux.
Bullshit lies. I couldn't write C++ to save my life, hardly know any perl, and know just enough bash to write a loop. I patch security updates by starting synaptic and pressing the upgrade button, which is easier than Windows Update. The only one of these skills I guess I have is CLI usage, but from what I have seen one could use (though possibly not configure completely) a Linux box without ever touching the CLI.
The command line is a big no-no for the general public. If you tell an ordinary user they got to go to the command line for anything... even to type the word "install", they're just gonna laugh at you.
But then, programs like synaptic and aptitude mean you don't have to use the command line to install software from apt. I mean, compared to packaging problem, how difficult is a GUI fronend?
The real problem with apt is that it depends on a single centrally managed collection of software. If you keep adding new apt sources for every package you install, things get screwed up very fast.
But note that since no actually distribution needs to have been made, it will be sufficient to have downloaded a copy of SW Ep III two days before it is released, and have it end up in the shared folder of the P2P client (or not even that as many clients like eDonkey and Overnet share files while they are being downloaded).
Anyone want to see the first ten year old prosecuted under this law?
If I say, "motorcycles are ready for the road," will people think that means that everybody who uses a car should be able to get on a motorcycle and start riding around? Does it mean that people who have delicate haircuts, or who use their cars to transport people and cargo, will become angry and dissapointed.
Linux can compete with Windows in it's market. I used to be in windows market, but am not any more because Linux provided a better desktop option for me. I bet 9/10 of the Linux users out there were taken from the Windows market.
"Linux is ready for desktop use. It may or may not be able to provide your desktop needs." I don't think most people will have any trouble understanding that.
I read this and found out that X has a completely adequate clipboard - it is just that applications too often clobber it. Apparently the latest version of pure desktops do it right though.
(I always thought that ctrl-c ctrl-v in X apps was just an ugly hack using the same selection as select and middle-click (apparently because old KDE did it wrong).)
That may be how windows explorer handles cut and paste of file icons, but AFAIK data copied by highlighting the actual text, image, etc is actually copied into a clipboard (how else can it survive after the application closes?)
.clipboard directory, for use by file managers, where the information written in .clipboard is the identity of the file, not the file to copy itself.
In the case of my proposal, you simply define a "file location" type of data in
and kudzu: http://rhlinux.redhat.com/kudzu/
I am using Suse 8.2 with KDE and I haven't noticed and copy/past issues. I can easily copy and paste between Opera, blufish, kwrite, openoffice, etc...
But you can't copy between for example OpenOffice and mozilla composer and keep the font and layout settings. Nor can you (most of the time) copy images or sounds or other multimedia between different programs. Closing applications typically means loosing what is copied, and it is easy to exidentally replace the buffer by highlighting something inadvertly.
I use X exclusively on the Desktop (Linux at home, Solaris at work) and I'm completely happy with highlight and middle click: but it does not match the cut-n-paste expectations of people moving from Windows and Mac.
It needs to be networked, like X, but have a standard widget set and clipboard.
.clipboard directory in the user directory instead? (or maybe /var/clipboard/userfoo/ ?) Cutting or copying something from an application would move the data into a file in this directory, and it would be possible to have more than one thing in the clipboard at once (possibly this should be a power-user feature). The files in the .clipboard should be in standardized fromats for each type of data, so we could support cut-n-paste between applications of more than just text. Cut and paste would be possible in the console, and the clipboard could survive crashes and reboots.
I'm not so sure that the clipboard should be a part of X, or it's equivalent. Wouldn't it be even nice if cut-n-paste was a system wide thing?
I think one of the biggest mistakes in Linux usability has been the mistaking of X's built in text drag-n-drop capability (which is what the highlight and middleclick is, not cut-n-paste) with a clipboard in the Windows or Mac sense.
Why couldn't we make a
Since we have good free file-fromats for most types of data, and libraries to handle it, implementing this doesn't seem so difficult. Is there any movement to have something like this happen?
1) It needs better CD burning software (any opinions on which is best?) It may have this already and I overlooked it...
K3B. gToaster. Burn.app.
2) It needs A pre-configured link to download a p2p system like Kazaa so kids can download music if they want. (it's what the consumers want, lets face it.)
But do the parents want a $2000 settlement with the RIAA?
3) It needs plug-n-play. I heard rumors about mandrake having it, anyone tried it though? If I plug in my USB-pen, it better find it, and open it as a folder! 2000/XP is the shit in this respect.
Mandrake, and especially Knoppix have had next to perfect PnP for years. I have used Knoppix on windows machines to identify hardware it couldn't recognize.
When people say that linux is not ready for the desktop, they mean that the average user will not be able to use it.
What do you mean exactly by "the average user"? Clearly the average user, as in the user who is the average of all users in every respect doesn't exist, and people even closely aproximating him are pretty rare.
Do you mean the median user? (That is to say, when 50% of the people can use it.) Or do you mean a higher percentile? Or is 100% necessary? (Including illiterate people?)
You cant say, it is erady, but only for the geeks desktops. That kinda defeats the purpose.
Why not? Why does it have to black and white? Why can't an OS be ready for some peoples desktops and not for others? If Linux has 2.6% of the market today, would going to 15% not be an improvement? Would those 15% be using an OS that wasn't "ready" simply because it was unsuitable for the other 85%?
Your thinking makes no sense to me. Linux is ready for the desktop, and has been for some time. It is not ready for everybody's desktop, but it is ready for more users today than it was two years ago, and will be ready for even more in another two years. I see no contradiction.
I got scared by a couple of posts, especially the one I responded to, which was why responded harshly. If they were joking, then I apologize, IHBT.
You are right that my math skills are medicore (I haven't gotten anywhere on my dissertation in the last couple of months) but that doesn't really matter we are talking Teletubby level...
Well, you're already in bed with the evil phone company... so what's the difference?
Someone needs to just run fiber to everyone's house/business and put all these bozos out of business.
What makes you think that wouldn't end up being the evil fiber company?
Wiring peoples houses is conductive to natural monopolies. Some part of me can't help but think it might be better off as public infrastructure (a la roads), but then I think of how much I would be paying to wire all the people who have chosen to live in rural areas in that case...
Perhaps wireless is the ticket (there is a company two houses down from mine that sells 802.11b broadband - unfortunately they pointed there directional anntenna in the other direction...)
Don't post math comments until you pass pre-calculus. Please.
Fuck baby jesus. The level of this discussion is making me cry.
sqrt(x) is a monotonically increasing function. If a > b then sqrt(a) > sqrt(b). This isn't even highschool math, for Christs sake. That sqrt(1)=1 does not mean that the function isn't increasing at that point (consider the function f(x) = x).
And for the record sqrt(x) -> infinity as x -> infinity. This is also bloody fucking obvious (if it wasn't true, there would have to be numbers so large they had no square). 1/f(x) -> 0 as f(x) -> infinity, Xeno's paradox has not been a problem for at least 400 years.
Just responding to this thread has made me lose half my IQ...
In theory somebody out there could write a Back Orifice style program but register the port with UPNP. This will allow external attackers to tunnel through the firewall as if it wasn't there.
Please! If the trojan is already installed, then getting around the firewall is trivial. It can simply connect out, either to cracked server somewhere, to an IRC channel (this how most trojans are controlled), or even to a webpage (say for instance an empty discussion here on Slashdot). If you are blocking outgoing ports, then it just makes sure that it uses port 80. If you have a software firewall, then it communicates by using Windows scripting to invoke IE to open the pages on it's behalf.
Firewalls are good for keeping things out. Once something is already on your network, the firewall is 100% useless, UPNP or not.
Obviously UPNP is only available to hosts INSIDE the router.
This part of UPNP is a GOOD thing. It helps against the growing problem of the "the firewalled consumer" that thanks to NAT has created an A and B class of Internet citizens: those that can be connected to, and those that can't.
A good example of this is VoIP. VoIP phones are being sold that connect straight to the home router, and allow calling over the Internet. But 99% of all home routers do NAT, so without port forwards it will be impossible for two such phones to talk to one another. Since users can't really be expected to set this up manually, UPNP is a necessity. I think it is fantastic that routers are starting to support this!
I agree that it ought to be more configurable (which ports it will allow forwards to, etc) but in the end we cannot secure things by neutering the entire Internet. You shouldn't be running insecure services (consider that P2P applications like IM can be cracked straight through a NAT: they connect back to others at the request of the central server. and those back connections are just as vulnerable to buffer overflows and incoming (there is a worm for you!)).
Hey if you wanna miss out on seeing Carrie-Anne Moss (as Trinity) gettin it on, fine by me... but that's definitelly not my chosen bathroom break :)
That depends on what he was going to do in the bathroom...
RTFA:
"From these variations, he could calculate the frequencies of the sound waves propagating through the Universe during its first 760,000 years, when it was just 18 million light years across."
Right after the big bang in astronomical time, not human time.