Well, actually he got elected largely on small donations. So he isn't as beholden to those interests as you think. The main worry would be that if he comes out strongly against them, they might fight to get him out of office in 2012. More likely, he's worried about congressional elections in 2010.
The fact is that for the average American, this simply isn't a deciding issue. It's not even on the radar. It's because of this that we are at the mercy of copyright and patent interests - their voices are the loudest, and to a large degree the only, voices that are speaking to these issues.
Figure out a way to get the average American to really care about these issues, and it won't matter what the lobbyists want. The only reason it looks like it matters is because there are so few things that we (as a group!) really care about enough to let it change our vote.
Firefox appears to run on it. I can't get the networking to work under VMware because (I think) VMware is choosing the wrong network adapter. There don't seem to be any preference panes or anything like that.
It happens quite a bit, both because people doing bug-fixes have to read the code to understand how to change it, and because people who do commercial work with Linux pay for code reviews.
When I was still maintaining the open-source ISC DHCP distribution, on several occasions I got email from people who were doing line-by-line audits of my code, and found some things that they thought were issues.
It's possible that they don't catch everything, but based on the bug reports I got, I'd have to say that they were very thorough.
Anyway, if you're concerned about security, buy your distro from someone who will give you some assurances about the security of the code. You can't afford to give those assurances out yourself. That's what companies like Red Hat and SuSE are for.
I dunno, calling on bills like this seems like a good idea to me. Why did King bring it up in the first place? Because he got more than one call from someone asking for this, and he hasn't heard from anyone opposing bills like this. Give him a whiff of a constituency that would support him being reasonable instead of an asshole, and he might not even advance a bill like this.
And anyway, this is no less insane than COPA. That took almost a decade and many millions of dollars, plus a lot of precious court time, to kill. Nipping this shit in the bud is extremely cost effective compared to litigating it after it passes. And we are the ones picking up the tab.
...now's your chance. It's been referred to the House Committee on Energy and Commerce. Please check the membership list to see if your representative is on it. If so, please call them and ask them not to support this bill when it is considered by the committee. Be polite. Try to have a good reason prepared before you call.
Well, it's kind of my point that talent and hard work aren't enough. And that backstabbing isn't any more useful. To a large degree you do rely on luck. But really what helps most is when everyone's pulling for you. And the only way to get that in the long run is to pull for them. Which is the opposite of backstabbing.
Hunh. That must explain why I'm living on the streets instead of in a nice house in a nice neighborhood. Oh, wait! No! I am living in the nice house. And I didn't stab anybody in the back to get it. Nor, for that matter, is my business acumen the reason I'm in the nice house - in fact, it's basically just good fortune.
I'm not saying that there's no value in hard work, or in any of the other things we do on the job. But I'm sufficiently ancient at this point to have seen a lot of comings and goings, and the fact is that prosperity and [insert name of business tactic here] are largely orthogonal. If you don't have any talent, sure, maybe being an asshole is your only hope. Or maybe you should just go do what you really want to do and stop screwing around in a job you aren't suited for.
Of course you can always find an anecdotal counterexample, but the one time I decided I wanted to get someone out of a management position that was interfering with my job, it wound up backfiring hugely (the situation was *worse* after I succeeded) and on a personal level it's something I regret to this day.
On the other hand, every time I've come into a job situation and behaved with honesty and integrity, it's worked out well for me. And I get to sleep at night.
It's a pretty safe bet that the good reviews are going to be astroturfed to some degree. If you don't assume that, you're living in a dream world. If you look at the bad reviews, you can see what pissed people off about the product. If what they say resonates for you, don't buy it. Sometimes what they say just indicates that they don't know what they're doing. But you can be pretty sure that they weren't astroturfed.
Although I suppose at some point manufacturers might start astroturfing the bad reviews too...
You can't make a problem go away by redefining it. When you try to open a new market niche, you do have competitors: the status quo. You can't say "oh, we're an educational project" and that way the status quo no longer exists - what you call yourself doesn't change who you are competing with.
In the case of OLPC, the two main competitors are the computer industry as it was, and doing nothing. Both of these are highly viable, defensible options. So if your marketing strategy doesn't take them into account, you have no hope of winning.
This is where OLPC really fell down, IMHO. They didn't come out of the gate with their guns blazing. If you wanted an OLPC, you had to jump through hoops to get it. If you wanted to try one, mostly that was too bad. If you wanted to do development that wasn't on the roadmap, too bad.
So what we wound up with coming out of the starting gate was a machine that didn't actually do most of the things it was supposed to do. There's no decent book reader application. Surfing the web is chancy at best. When Sugar is up and running, there is almost no free memory, so nearly anything you do is going to hit the wall on memory.
But holy shit, what a sweet piece of hardware.
So instead of grousing about what OLPC did wrong, I think the right thing to do is to just keep plugging. Will OLPC survive the economic crisis? Who knows? Who owns the OLPC hardware designs? Who knows? Can we do anything to help? Sure. Will our work be wasted if OLPC ultimately fails? No.
So instead of grousing, what I'd like to see is:
- Let's figure out whether or not we can get more OLPCs even if the OLPC project fails. And how much it will cost.
- Let's make the software better.
- Let's explore new ways of accomplishing the very laudable goals of the OLPC project, if we don't think the way they are doing it is right.
Nope, it's like that when you live there too. I used to park my car down at Pier 49 (ish) and then wait for the train at the Houston St. 1/9 stop. Usually a couple of express trains would go by before my train showed up. Just seeing so many people blip by in a few seconds was like drinking a shot of espresso.
Back before the WTC collapsed, I used to go to the mall in the basement there every morning for coffee and a pastry. I'd usually get there right at the peak of the rush, when the commuters were coming up out of the PATH trains in waves like a waterfall. It felt a lot like I imagine a gazelle would feel in the middle of a huge herd. Even though everyone was going in different directions, there were no collisions - just currents of people moving in and out of each other.
That's exactly what I thought about New Yorkers until I (wait for it) lived there for a while.
The problem with New York is that it's not something you can really grok by visiting for a week. If you live there, you can become part of a social web that doesn't really exist anywhere else I've lived - not in San Francisco, not in Boston, certainly not in Tucson.
There's something about the incredibly high population density that does it.
And it's not that all pizza in New York is better - frankly most of it sucks. I used to think that New York had the worst coffee ever, until recently. What New York really shines at is iterations.
What I mean is that in a typical city, there will be one or two coffee shops or pizza places that are in range for you. You're pretty much stuck with them if you don't want to go farther afield. So they are able to sustain themselves not by being good, but by being there.
In New York, there will be a dozen pizza places within walking distance of where you are living. So the ones that suck go out of business quickly. Survival of the fittest. You still have plenty of pizza places that neither suck nor excel, but if you really care about pizza, you can get outstanding pizza.
Likewise coffee. Most people don't give a shit, and they go to Dunkin Donuts or *$$ and are perfectly happy. More power to them. But there are a few insanely good coffee joints in New York, if that's what you care about. And because competition is harsh, the prices tend to be reasonable.
So yeah, New Yorkers and ex-New Yorkers do tend to trumpet the virtues of the city, and it is kind of annoying, but there's a reason for it.
First of all, you're right - China can't monitor its populace. It can try, and like the RIAA it can catch a fair number of violations, some of which may be real, and punish them all strongly in order set an example that it hopes the people it can't catch will follow anyway.
Secondly, China is much better organized than India. I don't always agree with the Chinese government (I don't always disagree, either), but they are a lot more effective at getting things done.
In India, almost any project is going to require a huge amount of baksheesh to get going, and the culture of baksheesh is going to prevent it from making any real progress - instead of resulting in enforcement, this program will simply enrich the people who are supposed to do the enforcement.
Actually, the evidence would suggest that the reason for the Mumbai attacks was not to establish a pretext for creating a panopticon state. Rather, it was a strategic move on the part of the Taliban in Pakistan to get Pakistani troops moved to the border with India and away from the Afghanistan border, so that the Taliban could act with impunity there. And that is precisely what has happened.
Next phase? Get rid of all the non-madrassa schools. Those are the ones that allow girls to attend. Then the entire region becomes a recruiting zone for more suicidal terrorists.
Meanwhile, back in India, this sounds like a typical piece of crap from the legislature, which often overreacts when bad things happen and writes legislation like this. Then there's a big public cry of outrage, and the legislation is withdrawn.
Anyway, India is the last place for a panopticon. Do you have any idea how many people there are there? It's simply not feasible.
Actually, a properly-designed recumbent can be much more efficient than a pedal bike, because you don't have the leverage on a pedal bike to really take advantage of the power of your leg muscles - you are really limited by the force you can bring to bear on the pedal, which comes from a combination of your body weight and whatever leverage you apply to the handlebars.
Back in the day, when people were still doing serious bicycle development, someone came up with a device called the Velocipede which set a land-speed record. The big innovation was the pedaling system, which operated more like a leg press in a weight lifting station than it did like a regular bicycle pedal, and thus allowed you to get a great deal more energy to the wheel with a lot less work.
The other big innovation was a two-rear-wheel steering system that allowed you to lie down almost flat on the velocipede, and thus reduced your wind resistance very substantially.
Unfortunately, the velocipede was very difficult to operate, and was so low to the ground that nobody could see you, and was never mass-produced, so quite expensive.
However, it was a very interesting advance in human-powered vehicle design, and it's disappointing to me that nobody's built on the work they did.
The problem right now is that when you produce bicycles, you get sued. And the more different your bicycle is than the average bicycle, the more likely you are to get sued. So most innovation is done by people with no assets, in small shops or in their own garages, and the bicycle that the average person actually gets to use hasn't changed in any substantial way in my lifetime.
Suppose you're smart. Suppose you can predict that oil prices will rise. Suppose you see that people are attracted to SUVs. But because you can see the future, because you can see that oil prices will rise, you see that if you build a company on the basis of what customers prefer now, it will work well for you in the short term, and so as you take advantage of what customers want now, you also work to try to sell more and more of them on what they are going to want when gas prices double.
This is the marketing strategy of a winner. A company that doesn't want to go off the deep end when the bottom falls out of a doomed market.
What the U.S. car industry did instead was to try to encourage people to want more and bigger SUVs, even though they must have known that the market was driving right off a cliff. They figured that when they drove off the cliff, it wouldn't be a problem, because their industry cannot fail. They would be bailed out. And so there was no risk involved in reaping the immediate rewards of an advertising-driven, artificially bloated market for SUVs.
And here we are, out over the edge of the cliff, looking at the bottom that's fallen out of the market. What are the car companies saying? Help us! Help us! We can't fail! Please forgive us for lying to you all these years. Please forgive us for pandering to your worst impulses. Please forgive us for lobbying for lower fuel standards (which, by the way, we are still doing)! Please dig deep into your pockets and rescue us, because see, we have three million hostages that we will shoot in the head if you don't.
So please, please, let us have another dozen years of insanely high bonuses at the expense of the rank and file workers, because if you don't, we will just kick them off the cliff and retire to Aspen with the bonuses we were collecting back when you guys were still buying SUVs.
Blame the workers all you want. The unions certainly helped to create the situation we're in now. But the bottom line is that things are so badly broken in the car industry right now that trusting those bastards to build any kind of national infrastructure would be like building a house with lard and rice paper.
You make changes to httpd.conf and don't restart apache? Dude, that's just nuts. Anyway, as long as your home machine with a stable address is in the file, you can still get in from there.
I've had KDE crash. I usually get sick of Gnome too quickly, but you're right, it does seem stable.
The problem is that most of the knobs don't do anything. Most of the hardware-related control panels don't work consistently or dependably. The GUI gets out of sync with the underlying system state. Sure, it doesn't crash, and if you're on a desktop machine with USB keyboard and mouse it will work, but go beyond that and you'd better know how to wield the command line.
Tabbing of windows: who cares? Seriously. I've tried this, and it's just not a big win. It's not bad, don't get me wrong, but it's not a big win. Tabs in the browser *are* a win, but I've yet to find another application where they're useful.
Tagging of windows: who cares? This one I say out of pure ignorance - maybe I'd like it if I learned how it worked. Is this anything like the ability in Mac to switch between windows that belong to the same app, rather than amongst all windows? It would be nice if you could do that on LInux - it's one of the big UI glitches I trip over on Linux.
Desktop activities: who cares? I don't use those on Windows or Mac, where they also exist. I don't know anyone who does use them.
D-Bus: yes, it's an open standard. It also completely reinvents the wheel in a very arcane way. The support libraries are version 0.1-ish. Last time I tried to do a d-bus app, I wound up writing my own library because none of the existing libraries had any documentation. Every tool I know of that's based on d-bus has a strong tendency to get wedged because there is no clear tracking of internal state in the system.
In principle I think d-bus is a great idea, and I'd really like it if it worked, but it's not 100% reliable, and it needs to be before it's useful. And I don't see a path to that point, because I don't think many people really clearly understand how it all fits together, and there's no quick way to learn.
Theming: this is a completely useless waste of time. That's harsh, I know, but it's true. I have no idea why people waste so much time on this nonsense. Make the damned computer work reliably. *Then* put some lipstick on it.
Multiple workspaces. I've tried these. They don't seem all that useful. When OS X came out with spaces, I tried using it for about a week and then just disabled it, because I wasted more time getting things into the right workspace than I saved having them there. On Linux I always disable the extra workspaces. Maybe I'm a luddite, but so is the average user.
User actions? Nobody but a serious geek is going to use stuff like this. That doesn't mean it's bad, but it's not a selling point for the average user.
What the average user cares about is that they can make the system reliably do what they need it to do. If it does that, they won't complain about it. If it doesn't do that, they'll reinstall the OEM copy of Windows over it in a heartbeat.
Slashdot Mirror
Ooh, good point!
Well, actually he got elected largely on small donations. So he isn't as beholden to those interests as you think. The main worry would be that if he comes out strongly against them, they might fight to get him out of office in 2012. More likely, he's worried about congressional elections in 2010.
The fact is that for the average American, this simply isn't a deciding issue. It's not even on the radar. It's because of this that we are at the mercy of copyright and patent interests - their voices are the loudest, and to a large degree the only, voices that are speaking to these issues.
Figure out a way to get the average American to really care about these issues, and it won't matter what the lobbyists want. The only reason it looks like it matters is because there are so few things that we (as a group!) really care about enough to let it change our vote.
No, Neo. It means you won't *have* to.
Firefox appears to run on it. I can't get the networking to work under VMware because (I think) VMware is choosing the wrong network adapter. There don't seem to be any preference panes or anything like that.
It happens quite a bit, both because people doing bug-fixes have to read the code to understand how to change it, and because people who do commercial work with Linux pay for code reviews.
When I was still maintaining the open-source ISC DHCP distribution, on several occasions I got email from people who were doing line-by-line audits of my code, and found some things that they thought were issues.
It's possible that they don't catch everything, but based on the bug reports I got, I'd have to say that they were very thorough.
Anyway, if you're concerned about security, buy your distro from someone who will give you some assurances about the security of the code. You can't afford to give those assurances out yourself. That's what companies like Red Hat and SuSE are for.
Does anybody still take CNN seriously as a news source?
I dunno, calling on bills like this seems like a good idea to me. Why did King bring it up in the first place? Because he got more than one call from someone asking for this, and he hasn't heard from anyone opposing bills like this. Give him a whiff of a constituency that would support him being reasonable instead of an asshole, and he might not even advance a bill like this.
And anyway, this is no less insane than COPA. That took almost a decade and many millions of dollars, plus a lot of precious court time, to kill. Nipping this shit in the bud is extremely cost effective compared to litigating it after it passes. And we are the ones picking up the tab.
...now's your chance. It's been referred to the House Committee on Energy and Commerce. Please check the membership list to see if your representative is on it. If so, please call them and ask them not to support this bill when it is considered by the committee. Be polite. Try to have a good reason prepared before you call.
Yeah, I used to work at a bank too. Fortunately, I was able to find work elsewhere eventually...
Well, it's kind of my point that talent and hard work aren't enough. And that backstabbing isn't any more useful. To a large degree you do rely on luck. But really what helps most is when everyone's pulling for you. And the only way to get that in the long run is to pull for them. Which is the opposite of backstabbing.
Correct.
That never works on women. They assume it's a frameup. Don't ask how I know...
(just kidding, of course)
(no, really!)
Hunh. That must explain why I'm living on the streets instead of in a nice house in a nice neighborhood. Oh, wait! No! I am living in the nice house. And I didn't stab anybody in the back to get it. Nor, for that matter, is my business acumen the reason I'm in the nice house - in fact, it's basically just good fortune.
I'm not saying that there's no value in hard work, or in any of the other things we do on the job. But I'm sufficiently ancient at this point to have seen a lot of comings and goings, and the fact is that prosperity and [insert name of business tactic here] are largely orthogonal. If you don't have any talent, sure, maybe being an asshole is your only hope. Or maybe you should just go do what you really want to do and stop screwing around in a job you aren't suited for.
Of course you can always find an anecdotal counterexample, but the one time I decided I wanted to get someone out of a management position that was interfering with my job, it wound up backfiring hugely (the situation was *worse* after I succeeded) and on a personal level it's something I regret to this day.
On the other hand, every time I've come into a job situation and behaved with honesty and integrity, it's worked out well for me. And I get to sleep at night.
So take your pick.
It's a pretty safe bet that the good reviews are going to be astroturfed to some degree. If you don't assume that, you're living in a dream world. If you look at the bad reviews, you can see what pissed people off about the product. If what they say resonates for you, don't buy it. Sometimes what they say just indicates that they don't know what they're doing. But you can be pretty sure that they weren't astroturfed.
Although I suppose at some point manufacturers might start astroturfing the bad reviews too...
You can't make a problem go away by redefining it. When you try to open a new market niche, you do have competitors: the status quo. You can't say "oh, we're an educational project" and that way the status quo no longer exists - what you call yourself doesn't change who you are competing with.
In the case of OLPC, the two main competitors are the computer industry as it was, and doing nothing. Both of these are highly viable, defensible options. So if your marketing strategy doesn't take them into account, you have no hope of winning.
This is where OLPC really fell down, IMHO. They didn't come out of the gate with their guns blazing. If you wanted an OLPC, you had to jump through hoops to get it. If you wanted to try one, mostly that was too bad. If you wanted to do development that wasn't on the roadmap, too bad.
So what we wound up with coming out of the starting gate was a machine that didn't actually do most of the things it was supposed to do. There's no decent book reader application. Surfing the web is chancy at best. When Sugar is up and running, there is almost no free memory, so nearly anything you do is going to hit the wall on memory.
But holy shit, what a sweet piece of hardware.
So instead of grousing about what OLPC did wrong, I think the right thing to do is to just keep plugging. Will OLPC survive the economic crisis? Who knows? Who owns the OLPC hardware designs? Who knows? Can we do anything to help? Sure. Will our work be wasted if OLPC ultimately fails? No.
So instead of grousing, what I'd like to see is:
- Let's figure out whether or not we can get more OLPCs even if the OLPC project fails. And how much it will cost.
- Let's make the software better.
- Let's explore new ways of accomplishing the very laudable goals of the OLPC project, if we don't think the way they are doing it is right.
- Let's not waste time crying in our beer.
Nope, it's like that when you live there too. I used to park my car down at Pier 49 (ish) and then wait for the train at the Houston St. 1/9 stop. Usually a couple of express trains would go by before my train showed up. Just seeing so many people blip by in a few seconds was like drinking a shot of espresso.
Back before the WTC collapsed, I used to go to the mall in the basement there every morning for coffee and a pastry. I'd usually get there right at the peak of the rush, when the commuters were coming up out of the PATH trains in waves like a waterfall. It felt a lot like I imagine a gazelle would feel in the middle of a huge herd. Even though everyone was going in different directions, there were no collisions - just currents of people moving in and out of each other.
That's exactly what I thought about New Yorkers until I (wait for it) lived there for a while.
The problem with New York is that it's not something you can really grok by visiting for a week. If you live there, you can become part of a social web that doesn't really exist anywhere else I've lived - not in San Francisco, not in Boston, certainly not in Tucson.
There's something about the incredibly high population density that does it.
And it's not that all pizza in New York is better - frankly most of it sucks. I used to think that New York had the worst coffee ever, until recently. What New York really shines at is iterations.
What I mean is that in a typical city, there will be one or two coffee shops or pizza places that are in range for you. You're pretty much stuck with them if you don't want to go farther afield. So they are able to sustain themselves not by being good, but by being there.
In New York, there will be a dozen pizza places within walking distance of where you are living. So the ones that suck go out of business quickly. Survival of the fittest. You still have plenty of pizza places that neither suck nor excel, but if you really care about pizza, you can get outstanding pizza.
Likewise coffee. Most people don't give a shit, and they go to Dunkin Donuts or *$$ and are perfectly happy. More power to them. But there are a few insanely good coffee joints in New York, if that's what you care about. And because competition is harsh, the prices tend to be reasonable.
So yeah, New Yorkers and ex-New Yorkers do tend to trumpet the virtues of the city, and it is kind of annoying, but there's a reason for it.
First of all, you're right - China can't monitor its populace. It can try, and like the RIAA it can catch a fair number of violations, some of which may be real, and punish them all strongly in order set an example that it hopes the people it can't catch will follow anyway.
Secondly, China is much better organized than India. I don't always agree with the Chinese government (I don't always disagree, either), but they are a lot more effective at getting things done.
In India, almost any project is going to require a huge amount of baksheesh to get going, and the culture of baksheesh is going to prevent it from making any real progress - instead of resulting in enforcement, this program will simply enrich the people who are supposed to do the enforcement.
Actually, the evidence would suggest that the reason for the Mumbai attacks was not to establish a pretext for creating a panopticon state. Rather, it was a strategic move on the part of the Taliban in Pakistan to get Pakistani troops moved to the border with India and away from the Afghanistan border, so that the Taliban could act with impunity there. And that is precisely what has happened.
Next phase? Get rid of all the non-madrassa schools. Those are the ones that allow girls to attend. Then the entire region becomes a recruiting zone for more suicidal terrorists.
Meanwhile, back in India, this sounds like a typical piece of crap from the legislature, which often overreacts when bad things happen and writes legislation like this. Then there's a big public cry of outrage, and the legislation is withdrawn.
Anyway, India is the last place for a panopticon. Do you have any idea how many people there are there? It's simply not feasible.
Actually, a properly-designed recumbent can be much more efficient than a pedal bike, because you don't have the leverage on a pedal bike to really take advantage of the power of your leg muscles - you are really limited by the force you can bring to bear on the pedal, which comes from a combination of your body weight and whatever leverage you apply to the handlebars.
Back in the day, when people were still doing serious bicycle development, someone came up with a device called the Velocipede which set a land-speed record. The big innovation was the pedaling system, which operated more like a leg press in a weight lifting station than it did like a regular bicycle pedal, and thus allowed you to get a great deal more energy to the wheel with a lot less work.
The other big innovation was a two-rear-wheel steering system that allowed you to lie down almost flat on the velocipede, and thus reduced your wind resistance very substantially.
Unfortunately, the velocipede was very difficult to operate, and was so low to the ground that nobody could see you, and was never mass-produced, so quite expensive.
However, it was a very interesting advance in human-powered vehicle design, and it's disappointing to me that nobody's built on the work they did.
The problem right now is that when you produce bicycles, you get sued. And the more different your bicycle is than the average bicycle, the more likely you are to get sued. So most innovation is done by people with no assets, in small shops or in their own garages, and the bicycle that the average person actually gets to use hasn't changed in any substantial way in my lifetime.
Suppose you're smart. Suppose you can predict that oil prices will rise. Suppose you see that people are attracted to SUVs. But because you can see the future, because you can see that oil prices will rise, you see that if you build a company on the basis of what customers prefer now, it will work well for you in the short term, and so as you take advantage of what customers want now, you also work to try to sell more and more of them on what they are going to want when gas prices double.
This is the marketing strategy of a winner. A company that doesn't want to go off the deep end when the bottom falls out of a doomed market.
What the U.S. car industry did instead was to try to encourage people to want more and bigger SUVs, even though they must have known that the market was driving right off a cliff. They figured that when they drove off the cliff, it wouldn't be a problem, because their industry cannot fail. They would be bailed out. And so there was no risk involved in reaping the immediate rewards of an advertising-driven, artificially bloated market for SUVs.
And here we are, out over the edge of the cliff, looking at the bottom that's fallen out of the market. What are the car companies saying? Help us! Help us! We can't fail! Please forgive us for lying to you all these years. Please forgive us for pandering to your worst impulses. Please forgive us for lobbying for lower fuel standards (which, by the way, we are still doing)! Please dig deep into your pockets and rescue us, because see, we have three million hostages that we will shoot in the head if you don't.
So please, please, let us have another dozen years of insanely high bonuses at the expense of the rank and file workers, because if you don't, we will just kick them off the cliff and retire to Aspen with the bonuses we were collecting back when you guys were still buying SUVs.
Blame the workers all you want. The unions certainly helped to create the situation we're in now. But the bottom line is that things are so badly broken in the car industry right now that trusting those bastards to build any kind of national infrastructure would be like building a house with lard and rice paper.
You make changes to httpd.conf and don't restart apache? Dude, that's just nuts. Anyway, as long as your home machine with a stable address is in the file, you can still get in from there.
I've had KDE crash. I usually get sick of Gnome too quickly, but you're right, it does seem stable.
The problem is that most of the knobs don't do anything. Most of the hardware-related control panels don't work consistently or dependably. The GUI gets out of sync with the underlying system state. Sure, it doesn't crash, and if you're on a desktop machine with USB keyboard and mouse it will work, but go beyond that and you'd better know how to wield the command line.
Tabbing of windows: who cares? Seriously. I've tried this, and it's just not a big win. It's not bad, don't get me wrong, but it's not a big win. Tabs in the browser *are* a win, but I've yet to find another application where they're useful.
Tagging of windows: who cares? This one I say out of pure ignorance - maybe I'd like it if I learned how it worked. Is this anything like the ability in Mac to switch between windows that belong to the same app, rather than amongst all windows? It would be nice if you could do that on LInux - it's one of the big UI glitches I trip over on Linux.
Desktop activities: who cares? I don't use those on Windows or Mac, where they also exist. I don't know anyone who does use them.
D-Bus: yes, it's an open standard. It also completely reinvents the wheel in a very arcane way. The support libraries are version 0.1-ish. Last time I tried to do a d-bus app, I wound up writing my own library because none of the existing libraries had any documentation. Every tool I know of that's based on d-bus has a strong tendency to get wedged because there is no clear tracking of internal state in the system.
In principle I think d-bus is a great idea, and I'd really like it if it worked, but it's not 100% reliable, and it needs to be before it's useful. And I don't see a path to that point, because I don't think many people really clearly understand how it all fits together, and there's no quick way to learn.
Theming: this is a completely useless waste of time. That's harsh, I know, but it's true. I have no idea why people waste so much time on this nonsense. Make the damned computer work reliably. *Then* put some lipstick on it.
Multiple workspaces. I've tried these. They don't seem all that useful. When OS X came out with spaces, I tried using it for about a week and then just disabled it, because I wasted more time getting things into the right workspace than I saved having them there. On Linux I always disable the extra workspaces. Maybe I'm a luddite, but so is the average user.
User actions? Nobody but a serious geek is going to use stuff like this. That doesn't mean it's bad, but it's not a selling point for the average user.
What the average user cares about is that they can make the system reliably do what they need it to do. If it does that, they won't complain about it. If it doesn't do that, they'll reinstall the OEM copy of Windows over it in a heartbeat.