So if AT&T can get an app banned (as Gruber says is the case), what happens later on when the iPhone is not tied to any one phone company in the U.S.? Carrier specific stores?
That's perfectly feasible. In fact, Google currently does that with their Android Market (equivalent of app store). They can and do filter certain apps based on the carrier, country, and phone hardware and software.
Google Voice for phone calls uses at&t minutes, which don't cost Apple.
Yes but using Google voice to make international calls would be way cheaper than making a phone call on your cell phone with AT&T. At the moment, the iPhone isn't just the device, it's also the infrastructure that supports the iPhone (which you pay gobs for). Google voice offers services that compete with AT&T and the iPhone infrastructure in ways big enough to hurt the bottom line of AT&T, which as you can see from other comments at the least, made this app go pouf disappear.
So, why didn't they pull Skype for those same reasons?
I don't dislike Microsoft because of their business practices; I dislike Microsoft because I don't like the way they design most of their products.
And Apple designs their products such that they are owned and controlled completely by Apple even after you've bought them from Apple. You consider that a good design? I consider that just like Microsoft.
I'd say Apple kicks it up a couple of notches. Microsoft doesn't usually exhibit this kind of control over the devices running their software. As far as I've been aware, you could freely install pretty much any application on Windows desktop/server, Windows mobile, and their other operating systems that accommodate 3rd party applications.
They do a fantastic job of luring in the mindless masses who don't have the wherewithal to actually think through the consequences of their purchases. When I buy something I want to control it.
Not everybody wants to exhibit full control over the devices they purchase. Sure auto enthusiasts would like to have full access to their cars' engines; geeks would like to know what their computers are doing and tweak them to their content. But most people just want to get from point A to point B safely and reliably; and most people run Windows and as long as they can browse the web, check e-mail and play their favorite game they don't care the details of the underlying OS.
I'd say most iPhone owners are perfectly content with Apple making most decisions for them - people are used to their carrier and manufacturer decide for them what features/programs go their cell phones anyway. In fact most of them probably haven't even heard of Google Voice, and if Apple and AT&T have their way, they won't ever hear about it.
This is not something new. Since you read the comment to TFA, I assume you also read TFA itself - the condition is called Electromagnetic Hypersensitivity and it appears some people do indeed suffer from it. While wifi is not the exclusive contributor to this, it amplifies the radiation, and it is meant to cover more areas and be turned on most of the time as opposed to other bluetooth devices, or baby monitors.
On the other hand, it's hard to tell whether this guy is telling the truth as being a DJ, he is probably exposed to different equipment emitting radiation at work throughout the day. And, why would he carry the wifi detector to avoid wifi specifically is not clear.
Maybe Slashdot isn't representative of the Linux community, but if it's not-- what is?
Nothing./. represents nothing more than the/. community. Not the "Linux community" (if there is such a thing), not the OSS community, not even the virgin community, etc.
Quite frankly, Linux is so big and so spread out already, there is no single "community" that covers everyone and everything. There are a lot of communities. But, hey, generalizations are fun and poking fun at "loons" is cute, so why not - pick your labels and hit the gas.
More to the point of the original article - that's the typical Linus position on these things. But like many other times, he sidesteps the issue (like a politician, ironically) and argues against a modified point. It's kind of like when politicians come on TV and start asking themselves and answering their own questions, completely ignoring the issue at hand.
But on the other hand, we already know what his positions on the issues are, so it doesn't really matter. Linus wants to support Linux, not the principle of Free software. Linus doesn't care if a thief steals his expensive TV and then sells the remote control back to him at a really good price. He'll buy the remote control and tell you not to hate the thief.
A better analogy would be if you bought a used car, then the dealership came back to you and said, "it turns out the car we sold you was stolen, and we had no right to sell it to you in the first place. Here's your money back." Yeah, that would suck, but I don't see any alternative (under the current legal regime).
Under UCC, which almost all (or all?) states have adopted, you can keep the car as long as you conducted the transaction in "good faith" - for example, you had no knowledge or couldn't have possibly known the merchandise was stolen.
My point is that this man has tremendous influence in the US. He's not an intellectual lightweight. Unfortunately, I can't read what he wrote since the blog entry seems to be down now.
Exactly my thoughts as well. Of course, it is very possible the cancer had eventually appeared in his liver as well - I am no medical guru, but liver being the center of the chemical balance of the whole body, it's pretty grim. A liver transplant buys you a little time, but the cancer is very likely to reappear very soon. For this reason, I didn't think they even allowed you access to the liver in this case - at least not in the USA.
Of course, there could be a different disease unrelated (or only indirectly related) to the previous cancer treatment that had damaged the liver, and it was being legitimately being replaced.
System-wide-installed extensions fundamentally can't be uninstalled directly by a user without some sort of privilege escalation, which Firefox doesn't support. MS didn't explicitly disable uninstallation, it's just a side effect of being a system-wide installation.
That doesn't make any sense. If a user is able to install the systemwide extension, that same user should be able to uninstall it also. There are 2 scenarios:
1. If an extension was installed through Firefox, delete it and remove all its configuration.
2. If it's a file contained outside of Firefox - i.e. not installed directly via Firefox - and resides in some external.net directory, then the "uninstall" functionality should simply be able to remove all references to that extension from Firefox. It would be similar to disable functionality, but take it out of the list altogether and remove all its references, settings, etc.
In either case, uninstall should never be disabled unless the user doesn't actually have rights/privileges to uninstall.
"It's like you're dreaming about Gorgonzola cheese when it's clearly Brie time baby!"
I just read the comments to this story to find this reference; and if not, to add one. That's classic. If you actually listen to that clip, it's relevant to this patent in more than one way. Classic.
They could go after proprietary vendors I suppose but I find it far more likely that these are defensive patents.
Sure they are probably not going to use the patents against open source projects themselves. But the possibility to use them against "proprietary vendors" part opens up a whole new can of worms. If there is an [unrelated] disagreement - e.g. licensing, agreements or whatever - between Red Hat and, say, Novell/Cisco/IBM/Oracle/etc. Red Hat lawyers can quickly bring these patents into their countersuit. Red Hat could argue - "you are distributing free software library XYZ which violates our patent 123."
As far as losing the ability to redistribute GPL software if they do bring up patents against it, that's not so straightforward as you describe. GPL is a licensing agreement between 2 parties. First, as far as I know, courts have yet to have a case claiming that and rule in that direction. Second, you probably would need help from the original author(s) of the library/software in question and a separate lawsuit against Red Hat. This lawsuit would still be unrelated to Red Hat's patent claims and could go on for as long as Red Hat could drag it out, or until the other party runs out of cash/time.
I believe the issue is that Google isn't allowed to provide such service with Google Maps. Google has licensed maps from different sources, which in turn have re-licensed it or are providing their own paid services such as turn-by-turn navigation, re-routing, etc. I believe the license agreement Google has limits the use of maps to not allow these services specifically. If you use Google Maps API, they make you agree to such terms too.
Want to do all that stuff? Get your own maps and host the service on your own servers. Something like andnav2.
A company that's hostile to all companies in your country is probably not going to do well, regardless who is or is not right.
True, country controls its own court system, and is free to favor local companies. However, if said country has also made international commitments and is part of such trade organizations, all things being equal, other countries would expect said country to treat legal entities from other countries fairly.
Otherwise they risk ruining relationships with the trade organizations and/or its member countries, and losing credibility in having legal entities from said country being treated equally fairly abroad.
I'm not saying that Windows contains back doors and switches but once you stir DRM into the equation, a tap of a key in a far off country could cripple your economy...
Very true. But I wouldn't call it DRM, as that term relates to a [usually] failed attempt at keeping Alice from Alice. Microsoft, Intel and others have announced their intentions to deliver a platform capable of producing a full true end-to-end encryption (see TCPA).
It would not be unimaginable for the OS to have certain applications/processes run in an encrypted mode where the instructions on any kind of storage device (HD, RAM, etc.) are encrypted; and where only the hardware executing the instructions is able to decrypt them.
So, given a collaboration between, say CIA/NSA and MS in the name of terrorism (or piracy), what may look like a normal encrypted channel to windowsupdate.microsoft.com sending/receiving information about automatic online updates may in fact be executing unwanted operations on the system, as well as transferring information - all without the user's knowledge, or any real power to find out.
Other browsers won't be able to do that with most FF plugins, unless they also implement XUL. This means they either have to re-use gecko, or write their own XUL implementation from scratch. I don't have anything against XUL, but I don't see any browsers taking it up.
In other news, anything you "hear" from anyone on/. that doesn't gel well with you, you can safely attribute to anyone else whom you disagree with. That way, you can label everyone you have an argument or a disagreement with with the same stroke of the brush chosen at your convenience. Whew! What a relief!
Exactly. If you are comparing the iPhone handset vs. the Blackberry handset you are absolutely right - they are completely different "levels of quality" - as in iPhone does it and Blackberry doesn't.
On the other hand, if you are comparing the Exchange integration between the iPhone handset vs. the Blackberry enterprise server, then you are comparing two different animals - and you at least have to briefly mention the fact.
You could also attach a paper calendar with a velcro ® at the back of the Blackberry and call it the "100% Exchange integration" feature. Hey, it's "paid" and "professional grade."
You are right - RIM gives access to their telephony functions to carriers, not to the 3rd party developers. That's what I meant. That is why you don't have Skype, Gizmo, or any other true VoIP clients on Blackberries, even via WiFi.
Here's their API. You can't program a phone with it.
Don't get me wrong, I have a Blackberry and I like it; but let's not start giving each other blowjobs just yet:
no restrictions on what you do with it
Can't do VoIP apps - restricted by RIM.
outstanding integration with Exchange...
Purchase/licensing and maintenance of a separate Blackberry enterprise server required. Note that iPhone integrates w/Exchange without requiring you to license/maintain this component.
Asus: Hey, MS - we are making 8-9" netbook PCs that are going to be small, hip and cheap; wanna give us a good rate on XP - say $X? MS: Hmmphhh... dream on - you guys pay what everyone else pays - $Y! No exceptions! Oh, and by the way, we'll be phasing out the XP, so plan on running and paying for Vista - $Z! Asus: OK then, we'll use Linux. MS: We've heard that before, good luck!
Later:
MS: Hey Asus, your netbook thingies really took off - they have to run XP - we want you to phase out your Linux version. Asus: Ummm, can't really do that, sorry. MS: We are giving you (and everyone else) XP licensing deal for $X for netbooks! You'd better do it, or we'll screw with your other products - you know how that goes. Asus: OK, master, you are the boss, will do. MS: Good boy.
My thoughts exactly. What I thought I was seeing was Asus making a hit in the sub-10" netbook and leading the pack. Everyone else was playing catch-up. All other manufacturers' offerings were being judged and rated against Asus' offerings and kept coming up short one way or another (price, features, SSD, camera, etc.). The last thing they (Asus) needed to accomplish with their Eee line was to get the cellular providers' support and place the GSM slot on the motherboard. Their hardware already supports all this!
Now, they are going to stop doing what they did right and get into the same business with everyone else. Asus, in my eyes, is not a leader in the 10"+ market. They are going to have to play catch-up with big boys and will probably lose. Bad decision, IMO.
In fact, I own a 900 (Linux version), and I was planning to get 901 and give the 900 to my wife who has been begging me for it for awhile. I may not get it anymore if the whole line is going to be phased out in the next few months. Hopefully, other manufacturers will fill the void Asus is choosing to abandon. There is a big demand for sub-10" netbooks (fast boot times, long battery life, good performance, etc.) and hopefully it gets filled and they do it right.
I still don't understand why Asus would abandon the market where they are doing so good! Did I say it was a bad decision? Is this a manufactured story by El Reg again?
Under the DMCA, providers are required by law to remove the "offending" material upon receipt of an infringement notice. If they don't, they become liable for infringement as well. No real proof of ownership is required, the author of the notice simply has to say it belongs to them.
Don't forget that the claim in the DMCA notice by the copyright holder is made under the penalty of perjury. In other words, assuming this was a DMCA request, if IOC claimed to have copyright on X and they don't they have just perjured themselves. Although, I don't know what processes exist to prevent DMCA abuses, or who checks the validity of all notices sent, and most importantly, who does anything when the claims in them are false. AFAIK, most DMCA notices are never forwarded to the alleged violators, they are just executed and thrown away by providers. Maybe we do need copyright cops after all, just for a different purpose than the entertainment industry has in mind.
I still use my "ancient" Netgear RT311 (has both web and telnet interfaces), has run for years at a time without rebooting. None of the setting changes require it to restart.
Since RT311 is just a router (no access point, or a hub/switch), I have a separate switch and a Cisco access point (which also has rock solid stability).
I recently bought a combo device (wireless ap, router, hub) D-Link from Best Buy for my internal subnet and I have to reset that puppy every few days. Sometimes it just gets weird latency issues as well out of the blue.
So I'm dumb or a troll for feeling let down when a stable release (which a 4.0.X release should be) of a major open source project tends to crash on me constantly?
KDE 4.0.x does NOT crash "constantly" - make sure you use the distribution that cares about the desktop and their users, or packages that are stable. I have been running KDE 4.0.x since it came out on OpenSUSE 10.3 with 16 virtual desktops with at least 20 applications ( don't like to close apps and tend to leave them open for reuse) open at a time, and handful of plasmoids - the desktop is on for weeks sometimes (until I restart it for a patch) with everyday work load and hasn't crashed on me ONCE. And, this is with compositing and certain effects turned on too.
If your 4.0 branch has scathing architectural deficiencies that make it unusable for production environments
It does NOT have any "architectural deficiencies" - do you even know what you are talking about? KDE4, plasma, the new compositing window manager, Qt4, Akonadi provide vast architectural foundation for future application development. It is the best open source (if not all of software) effort for desktop architecture.
Slashdot Mirror
So if AT&T can get an app banned (as Gruber says is the case), what happens later on when the iPhone is not tied to any one phone company in the U.S.? Carrier specific stores?
That's perfectly feasible. In fact, Google currently does that with their Android Market (equivalent of app store). They can and do filter certain apps based on the carrier, country, and phone hardware and software.
Google Voice for phone calls uses at&t minutes, which don't cost Apple.
Yes but using Google voice to make international calls would be way cheaper than making a phone call on your cell phone with AT&T. At the moment, the iPhone isn't just the device, it's also the infrastructure that supports the iPhone (which you pay gobs for). Google voice offers services that compete with AT&T and the iPhone infrastructure in ways big enough to hurt the bottom line of AT&T, which as you can see from other comments at the least, made this app go pouf disappear.
So, why didn't they pull Skype for those same reasons?
I don't dislike Microsoft because of their business practices; I dislike Microsoft because I don't like the way they design most of their products.
And Apple designs their products such that they are owned and controlled completely by Apple even after you've bought them from Apple. You consider that a good design? I consider that just like Microsoft.
I'd say Apple kicks it up a couple of notches. Microsoft doesn't usually exhibit this kind of control over the devices running their software. As far as I've been aware, you could freely install pretty much any application on Windows desktop/server, Windows mobile, and their other operating systems that accommodate 3rd party applications.
They do a fantastic job of luring in the mindless masses who don't have the wherewithal to actually think through the consequences of their purchases. When I buy something I want to control it.
Not everybody wants to exhibit full control over the devices they purchase. Sure auto enthusiasts would like to have full access to their cars' engines; geeks would like to know what their computers are doing and tweak them to their content. But most people just want to get from point A to point B safely and reliably; and most people run Windows and as long as they can browse the web, check e-mail and play their favorite game they don't care the details of the underlying OS.
I'd say most iPhone owners are perfectly content with Apple making most decisions for them - people are used to their carrier and manufacturer decide for them what features/programs go their cell phones anyway. In fact most of them probably haven't even heard of Google Voice, and if Apple and AT&T have their way, they won't ever hear about it.
This is not something new. Since you read the comment to TFA, I assume you also read TFA itself - the condition is called Electromagnetic Hypersensitivity and it appears some people do indeed suffer from it. While wifi is not the exclusive contributor to this, it amplifies the radiation, and it is meant to cover more areas and be turned on most of the time as opposed to other bluetooth devices, or baby monitors.
On the other hand, it's hard to tell whether this guy is telling the truth as being a DJ, he is probably exposed to different equipment emitting radiation at work throughout the day. And, why would he carry the wifi detector to avoid wifi specifically is not clear.
Nothing. /. represents nothing more than the /. community. Not the "Linux community" (if there is such a thing), not the OSS community, not even the virgin community, etc.
Quite frankly, Linux is so big and so spread out already, there is no single "community" that covers everyone and everything. There are a lot of communities. But, hey, generalizations are fun and poking fun at "loons" is cute, so why not - pick your labels and hit the gas.
More to the point of the original article - that's the typical Linus position on these things. But like many other times, he sidesteps the issue (like a politician, ironically) and argues against a modified point. It's kind of like when politicians come on TV and start asking themselves and answering their own questions, completely ignoring the issue at hand.
But on the other hand, we already know what his positions on the issues are, so it doesn't really matter. Linus wants to support Linux, not the principle of Free software. Linus doesn't care if a thief steals his expensive TV and then sells the remote control back to him at a really good price. He'll buy the remote control and tell you not to hate the thief.
Under UCC, which almost all (or all?) states have adopted, you can keep the car as long as you conducted the transaction in "good faith" - for example, you had no knowledge or couldn't have possibly known the merchandise was stolen.
Relevant quote from Article 2, 2-403
A person with voidable title has power to transfer a good title to a good faith purchaser for value.
Read that link for more legalese if you wish.
Come on, do I get some love for quoting the law?
Now do you see the irony?
Exactly my thoughts as well. Of course, it is very possible the cancer had eventually appeared in his liver as well - I am no medical guru, but liver being the center of the chemical balance of the whole body, it's pretty grim. A liver transplant buys you a little time, but the cancer is very likely to reappear very soon. For this reason, I didn't think they even allowed you access to the liver in this case - at least not in the USA.
Of course, there could be a different disease unrelated (or only indirectly related) to the previous cancer treatment that had damaged the liver, and it was being legitimately being replaced.
That doesn't make any sense. If a user is able to install the systemwide extension, that same user should be able to uninstall it also. There are 2 scenarios:
1. If an extension was installed through Firefox, delete it and remove all its configuration.
2. If it's a file contained outside of Firefox - i.e. not installed directly via Firefox - and resides in some external .net directory, then the "uninstall" functionality should simply be able to remove all references to that extension from Firefox. It would be similar to disable functionality, but take it out of the list altogether and remove all its references, settings, etc.
In either case, uninstall should never be disabled unless the user doesn't actually have rights/privileges to uninstall.
"It's like you're dreaming about Gorgonzola cheese when it's clearly Brie time baby!"
I just read the comments to this story to find this reference; and if not, to add one. That's classic. If you actually listen to that clip, it's relevant to this patent in more than one way. Classic.
Sure they are probably not going to use the patents against open source projects themselves. But the possibility to use them against "proprietary vendors" part opens up a whole new can of worms. If there is an [unrelated] disagreement - e.g. licensing, agreements or whatever - between Red Hat and, say, Novell/Cisco/IBM/Oracle/etc. Red Hat lawyers can quickly bring these patents into their countersuit. Red Hat could argue - "you are distributing free software library XYZ which violates our patent 123."
As far as losing the ability to redistribute GPL software if they do bring up patents against it, that's not so straightforward as you describe. GPL is a licensing agreement between 2 parties. First, as far as I know, courts have yet to have a case claiming that and rule in that direction. Second, you probably would need help from the original author(s) of the library/software in question and a separate lawsuit against Red Hat. This lawsuit would still be unrelated to Red Hat's patent claims and could go on for as long as Red Hat could drag it out, or until the other party runs out of cash/time.
I believe the issue is that Google isn't allowed to provide such service with Google Maps. Google has licensed maps from different sources, which in turn have re-licensed it or are providing their own paid services such as turn-by-turn navigation, re-routing, etc. I believe the license agreement Google has limits the use of maps to not allow these services specifically. If you use Google Maps API, they make you agree to such terms too.
Want to do all that stuff? Get your own maps and host the service on your own servers. Something like andnav2.
True, country controls its own court system, and is free to favor local companies. However, if said country has also made international commitments and is part of such trade organizations, all things being equal, other countries would expect said country to treat legal entities from other countries fairly.
Otherwise they risk ruining relationships with the trade organizations and/or its member countries, and losing credibility in having legal entities from said country being treated equally fairly abroad.
Very true. But I wouldn't call it DRM, as that term relates to a [usually] failed attempt at keeping Alice from Alice. Microsoft, Intel and others have announced their intentions to deliver a platform capable of producing a full true end-to-end encryption (see TCPA).
It would not be unimaginable for the OS to have certain applications/processes run in an encrypted mode where the instructions on any kind of storage device (HD, RAM, etc.) are encrypted; and where only the hardware executing the instructions is able to decrypt them.
So, given a collaboration between, say CIA/NSA and MS in the name of terrorism (or piracy), what may look like a normal encrypted channel to windowsupdate.microsoft.com sending/receiving information about automatic online updates may in fact be executing unwanted operations on the system, as well as transferring information - all without the user's knowledge, or any real power to find out.
Other browsers won't be able to do that with most FF plugins, unless they also implement XUL. This means they either have to re-use gecko, or write their own XUL implementation from scratch. I don't have anything against XUL, but I don't see any browsers taking it up.
In other news, anything you "hear" from anyone on /. that doesn't gel well with you, you can safely attribute to anyone else whom you disagree with. That way, you can label everyone you have an argument or a disagreement with with the same stroke of the brush chosen at your convenience. Whew! What a relief!
Exactly. If you are comparing the iPhone handset vs. the Blackberry handset you are absolutely right - they are completely different "levels of quality" - as in iPhone does it and Blackberry doesn't.
On the other hand, if you are comparing the Exchange integration between the iPhone handset vs. the Blackberry enterprise server, then you are comparing two different animals - and you at least have to briefly mention the fact.
You could also attach a paper calendar with a velcro ® at the back of the Blackberry and call it the "100% Exchange integration" feature. Hey, it's "paid" and "professional grade."
You are right - RIM gives access to their telephony functions to carriers, not to the 3rd party developers. That's what I meant. That is why you don't have Skype, Gizmo, or any other true VoIP clients on Blackberries, even via WiFi.
Here's their API. You can't program a phone with it.
Don't get me wrong, I have a Blackberry and I like it; but let's not start giving each other blowjobs just yet:
Can't do VoIP apps - restricted by RIM.
Purchase/licensing and maintenance of a separate Blackberry enterprise server required. Note that iPhone integrates w/Exchange without requiring you to license/maintain this component.
Here's what I think happened:
Several years ago:
Asus: Hey, MS - we are making 8-9" netbook PCs that are going to be small, hip and cheap; wanna give us a good rate on XP - say $X?
MS: Hmmphhh... dream on - you guys pay what everyone else pays - $Y! No exceptions! Oh, and by the way, we'll be phasing out the XP, so plan on running and paying for Vista - $Z!
Asus: OK then, we'll use Linux.
MS: We've heard that before, good luck!
Later:
MS: Hey Asus, your netbook thingies really took off - they have to run XP - we want you to phase out your Linux version.
Asus: Ummm, can't really do that, sorry.
MS: We are giving you (and everyone else) XP licensing deal for $X for netbooks! You'd better do it, or we'll screw with your other products - you know how that goes.
Asus: OK, master, you are the boss, will do.
MS: Good boy.
My thoughts exactly. What I thought I was seeing was Asus making a hit in the sub-10" netbook and leading the pack. Everyone else was playing catch-up. All other manufacturers' offerings were being judged and rated against Asus' offerings and kept coming up short one way or another (price, features, SSD, camera, etc.). The last thing they (Asus) needed to accomplish with their Eee line was to get the cellular providers' support and place the GSM slot on the motherboard. Their hardware already supports all this!
Now, they are going to stop doing what they did right and get into the same business with everyone else. Asus, in my eyes, is not a leader in the 10"+ market. They are going to have to play catch-up with big boys and will probably lose. Bad decision, IMO.
In fact, I own a 900 (Linux version), and I was planning to get 901 and give the 900 to my wife who has been begging me for it for awhile. I may not get it anymore if the whole line is going to be phased out in the next few months. Hopefully, other manufacturers will fill the void Asus is choosing to abandon. There is a big demand for sub-10" netbooks (fast boot times, long battery life, good performance, etc.) and hopefully it gets filled and they do it right.
I still don't understand why Asus would abandon the market where they are doing so good! Did I say it was a bad decision? Is this a manufactured story by El Reg again?
Don't forget that the claim in the DMCA notice by the copyright holder is made under the penalty of perjury. In other words, assuming this was a DMCA request, if IOC claimed to have copyright on X and they don't they have just perjured themselves. Although, I don't know what processes exist to prevent DMCA abuses, or who checks the validity of all notices sent, and most importantly, who does anything when the claims in them are false. AFAIK, most DMCA notices are never forwarded to the alleged violators, they are just executed and thrown away by providers. Maybe we do need copyright cops after all, just for a different purpose than the entertainment industry has in mind.
I still use my "ancient" Netgear RT311 (has both web and telnet interfaces), has run for years at a time without rebooting. None of the setting changes require it to restart.
Since RT311 is just a router (no access point, or a hub/switch), I have a separate switch and a Cisco access point (which also has rock solid stability).
I recently bought a combo device (wireless ap, router, hub) D-Link from Best Buy for my internal subnet and I have to reset that puppy every few days. Sometimes it just gets weird latency issues as well out of the blue.
KDE 4.0.x does NOT crash "constantly" - make sure you use the distribution that cares about the desktop and their users, or packages that are stable. I have been running KDE 4.0.x since it came out on OpenSUSE 10.3 with 16 virtual desktops with at least 20 applications ( don't like to close apps and tend to leave them open for reuse) open at a time, and handful of plasmoids - the desktop is on for weeks sometimes (until I restart it for a patch) with everyday work load and hasn't crashed on me ONCE. And, this is with compositing and certain effects turned on too.
It does NOT have any "architectural deficiencies" - do you even know what you are talking about? KDE4, plasma, the new compositing window manager, Qt4, Akonadi provide vast architectural foundation for future application development. It is the best open source (if not all of software) effort for desktop architecture.