The problem is, of course, that you are entirely wrong.
Yes, you need to know every one of those things, except maybe the difference between magnetic and true north (unless you're close to the north pole, anyway) because GPSes are, get this, sometimes wrong. Or worse, they don't work at all.
I was driving the other day on an Ontario highway. I don't live in Ontario, so I was using my GPS. Guess what: the GPS had an outdated map, so it got lost. I didn't because I know how to use a map besides a GPS, and because I can read road signs.
What Joel says is exactly right - nobody gets to be a good programmer without having good knowledge of the underlying systems. Oh, you can be passable and earn a decent living, but all the really good ones know exactly how the CPU is going to execute their code and how all the different parts fit together. The same thing goes for any other part of life. You can get around fine without worrying too much about how your GPS works or about the finer details of cartography, but when it comes down to it, anyone who can use a map is automatically better at navigating than someone who can't because there are going to be situations where a GPS just isn't going to be good enough, and even if you never come across those, knowing how to use a map will make you better at following directions, anyway, much like knowing how the CPU works can help you even when you're writing in a high level language.
What I don't get is the comment in the summary about the economics of doing this. So what if it's 40x as expensive. Let's look at the breakdown. Your standard 100W light bulb, at $0.10/kwhr will, if used for 4 hours a day, will cost $0.04/day to run. This new bulb will cost $0.028 to run if it's 30% more efficient. That's a savings of 1.2 cents per day. The price difference is $4.75, so that means it would take 396 days, or about 13 months to pay for itself. However, it also claims to last three times longer, so I'd easily expect the bulb to last at least 13 months (a normal incandescent bulb can last this long with this kind of load). So it should pay for itself rather easily.
Sigh. I said "close". I also double counted a lot more things if it uses fork without exec, or if it mmaps a file with MAP_SHARED with PROT_WRITE. What I gave was a simple way to get something reasonable. If you want something even closer, you need more sophisticated methods than that.
Actually, I think you're on the verge of something here. Let me make the step for you. Repeat after me... "Nobody will choose security over convenience."
You can say you know what you're doing, but the only real difference between you and the "convenience users" you mention is that you draw the line in a different place. There are still plenty of things that are probably too inconvenient for you to do, even though they'd make your computer more secure.
Really, the most secure OS is not the one that is off, nor the one that can be used in a secure fashion if you know what you're doing. The most secure OS is the one that makes security convenient.
It is for sure not easy, and saying "someone competent" indicates how much you have thought about it. Which is not much at all.
Try not to betray your ignorance so readily. On Linux, you can get something pretty close by summing all the rw sections of/proc/self/maps and then counting the r sections once. External programs don't matter, obviously, because they're all the same for all browsers, so if you count them or not, it's the same result, just so long as you do things consistently.
Someone with more Windows knowledge can probably tell you how to do this with Windows. I'm pretty sure you can get the loaded module list with Windows, at least, and maybe even some information about how much memory each is using.
Again, someone competent would know this. Just because it's hard doesn't mean someone competent can't figure it out. I've done this kind of thing plenty of times, though on Unix OSes, so save your condescension for the peanut gallery.
Do you really do 'which sudo' before running it and entering your password.
Do you think it's so hard to replace "which" as well?
Sha1 your.profile each time your open a new terminal?
Or sha1sum?
Check your.desktop files to see if the system administration menu items actually run those programs?
That could be a little harder, but I'm sure you get my point by now.
However, there are several good reasons not to run your browser as root. First, it can do a lot more damage if it misbehaves. Second, there's greater security exposure since it means that other users who don't have sudo access on the local machine might be able to get root by exploiting the browser.
E.g. if someone exploits sendmail, but it's run as a non-privileged user, then that user might be able to find a hole exploitable only via localhost to gain root access via the browser. This is why you should run as few things as root as possible, even if they don't access remote machines.
Actually, you should learn at least a little about the page cache and such before saying such things. You definitely do end up with sharing, even between processes. More so on Linux than Windows, but Windows still shares memory.
For 30 tabs, you can actually get a lot more than that. If the base libraries and the shared code for chrome itself are counted 30 times, then that can easily double the amount of memory, or more. I also looked for that in the article and the author states he summed the memory for all processes, which is to say that the stats for Chrome are wrong. This would also apply to IE, had he been successful at collecting any.
Also, the Firefox memory, and most likely all the others, are wrong, too, because Firefox ends up using up memory that never gets released normally when you use JavaScript applications. Simply opening tabs and summing memory usage is an idiotic way to measure memory usage of a browser.
In short, you should try to find someone competent to run your memory benchmarks.
Chrome is doing shit like reporting back all the websites I visit to a centralized database.
So is Firefox if you have phishing detection turned on.
Anyway, I expect that many of these things will be in Chrome soon enough, once extension support is done.
BTW, one point I should mention is that although Chrome doesn't support NoScript right now, there's less of a need for it because of its improved security model. At least in my case, that's a lot better because I tried NoScript and it drove me insane within a week, so I removed it. YMMV
Finally, Chrome has better privacy than Firefox in some ways because it has an anonymous browsing mode, and it's more secure because of the process and sandbox model they use. Firefox is working on that, but they're a long way behind.
Really, Firefox is falling pretty far behind many of the other browsers at this point. Don't get me wrong - it's still a good browser, and way better than IE, but all that it's got going for it now is the extension framework. Safari and Chrome are both way faster than Firefox. Even IE has a better process model (Firefox runs everything in a single thread, which is why it gets really sluggish with a lot of open tabs, or when one tab is really CPU heavy). Firefox is also a huge memory hog and, at least on Linux, is pretty unstable.
Again, most of these things will be fixed if they can fix their process model, but I expect that's a long ways off. They've got a lot of catching up to do.
Yes, it's a range. Think about it. 14 GHz - 7 GHz is still 7 GHz, so it can be a particular frequency, but the range of frequencies uses the same unit. It's like how you can say something is 10 feet away or how something is 10 feet long, between 10 feet away and 20 feet away.
I've been kind of expecting/hoping for this for some time. Ever since the economy got so bad, I figured we might come to this. I'm actually amazed this didn't have before. You can't go that long with such incompetent management of a project and expect to be allowed to keep going, especially when you can't even be bothered to produce an actual time estimate for completion anymore except "When it's done".
I didn't say the Android's UI was bad. I know reading carefully and recognizing nuance is a dying art, especially around here, but please try to pay at least some attnetion.
What I did say was that the Android was less polished than the iPhone. I've yet to see any other phone come close to the iPhone in that way. But I do like the Android quite a lot, TYVM.
You clearly are a programmer, but you're not very good at it. At least if your use of O notation is any indication. I've lost count of the number of times I had to jailbreak my iphone. I've lost several hours of my life doing all the nasty things required to make it work after it's eaten itself for no apparent reason. I've spent quite a bit of time reading poorly written blogs to see if they have finally figured out how to jailbreak the most recent update so I can install it, and whether I need to run some piece of software first pre-upgrade or else I might lose my data. And then I just stopped upgrading. Wasn't worth the hassle. Then I got a Google Android and it's unlocked, I can put anything I want on it, and I can use the shell or ssh or whatever without having to do anything except find an app in the marketplace.
Overall, I'd say I've spent a lot less time getting the Android to work than the iPhone. Of course, if you feel your time is not as valuable as mine, you're welcome to continue doing whatever you want.
You have to first jailbreak the phone if you want to unlock it. But I recently switched to Google Android so I don't have to deal with this. It's a less nice experience, but I imagine a lot of people who are willing to go through the trouble of jailbreaking a phone are also willing to put up with the less polished UI.
Obama is pretty bright as far as he goes. Read his books or something. He's a pretty thoughtful guy.
But saying the democrats are smarter than the republicans seems kinda pointless. Arguing about which party is smarter is kind of like arguing about who's the valedictorian of the special ed class.
TomTom has, apparently, been pursuing MS for a year. We have no idea what they've been saying or asking for. TomTom probably felt as you say - that it shouldn't file a lawsuit because this very thing would happen. So maybe they tried to bluff and MS called it. Or maybe they didn't think their terms were unreasonable but MS did.
We don't know what went on in these negotiations, but we do know it's been going on for a year with no license signed. Maybe it's one huge coincidence that MS was asked to pay royalties and then sued the very same company asking for them for patent violations when they have no prior history of using patents in a first-strike manner.
At the very least, if this is the start of something bigger, then MS went after TomTom first because they already were threatening to sue. But my money is on this just being an old fashioned throw-everything-at-the-wall-and-see-what-sticks lawsuit to force a settlement on terms more favorable to MS.
The funny thing is that, if the summary is true, it could have been TomTom, not MS, that shot first. Maybe MS suing TomTom was just retaliation for TomTom trying to collect royalties.
There is one click buying where you preconfigure everything like which address to use, the shipping method, and the credit card. You have to explicitly enable it in your account with them.
Slashdot Mirror
The problem is, of course, that you are entirely wrong.
Yes, you need to know every one of those things, except maybe the difference between magnetic and true north (unless you're close to the north pole, anyway) because GPSes are, get this, sometimes wrong. Or worse, they don't work at all.
I was driving the other day on an Ontario highway. I don't live in Ontario, so I was using my GPS. Guess what: the GPS had an outdated map, so it got lost. I didn't because I know how to use a map besides a GPS, and because I can read road signs.
What Joel says is exactly right - nobody gets to be a good programmer without having good knowledge of the underlying systems. Oh, you can be passable and earn a decent living, but all the really good ones know exactly how the CPU is going to execute their code and how all the different parts fit together. The same thing goes for any other part of life. You can get around fine without worrying too much about how your GPS works or about the finer details of cartography, but when it comes down to it, anyone who can use a map is automatically better at navigating than someone who can't because there are going to be situations where a GPS just isn't going to be good enough, and even if you never come across those, knowing how to use a map will make you better at following directions, anyway, much like knowing how the CPU works can help you even when you're writing in a high level language.
What I don't get is the comment in the summary about the economics of doing this. So what if it's 40x as expensive. Let's look at the breakdown. Your standard 100W light bulb, at $0.10/kwhr will, if used for 4 hours a day, will cost $0.04/day to run. This new bulb will cost $0.028 to run if it's 30% more efficient. That's a savings of 1.2 cents per day. The price difference is $4.75, so that means it would take 396 days, or about 13 months to pay for itself. However, it also claims to last three times longer, so I'd easily expect the bulb to last at least 13 months (a normal incandescent bulb can last this long with this kind of load). So it should pay for itself rather easily.
Sigh. I said "close". I also double counted a lot more things if it uses fork without exec, or if it mmaps a file with MAP_SHARED with PROT_WRITE. What I gave was a simple way to get something reasonable. If you want something even closer, you need more sophisticated methods than that.
Actually, I think you're on the verge of something here. Let me make the step for you. Repeat after me... "Nobody will choose security over convenience."
You can say you know what you're doing, but the only real difference between you and the "convenience users" you mention is that you draw the line in a different place. There are still plenty of things that are probably too inconvenient for you to do, even though they'd make your computer more secure.
Really, the most secure OS is not the one that is off, nor the one that can be used in a secure fashion if you know what you're doing. The most secure OS is the one that makes security convenient.
Try not to betray your ignorance so readily. On Linux, you can get something pretty close by summing all the rw sections of /proc/self/maps and then counting the r sections once. External programs don't matter, obviously, because they're all the same for all browsers, so if you count them or not, it's the same result, just so long as you do things consistently.
Someone with more Windows knowledge can probably tell you how to do this with Windows. I'm pretty sure you can get the loaded module list with Windows, at least, and maybe even some information about how much memory each is using.
Again, someone competent would know this. Just because it's hard doesn't mean someone competent can't figure it out. I've done this kind of thing plenty of times, though on Unix OSes, so save your condescension for the peanut gallery.
Do you think it's so hard to replace "which" as well?
Or sha1sum?
That could be a little harder, but I'm sure you get my point by now.
However, there are several good reasons not to run your browser as root. First, it can do a lot more damage if it misbehaves. Second, there's greater security exposure since it means that other users who don't have sudo access on the local machine might be able to get root by exploiting the browser.
E.g. if someone exploits sendmail, but it's run as a non-privileged user, then that user might be able to find a hole exploitable only via localhost to gain root access via the browser. This is why you should run as few things as root as possible, even if they don't access remote machines.
Actually, you should learn at least a little about the page cache and such before saying such things. You definitely do end up with sharing, even between processes. More so on Linux than Windows, but Windows still shares memory.
For 30 tabs, you can actually get a lot more than that. If the base libraries and the shared code for chrome itself are counted 30 times, then that can easily double the amount of memory, or more. I also looked for that in the article and the author states he summed the memory for all processes, which is to say that the stats for Chrome are wrong. This would also apply to IE, had he been successful at collecting any.
Also, the Firefox memory, and most likely all the others, are wrong, too, because Firefox ends up using up memory that never gets released normally when you use JavaScript applications. Simply opening tabs and summing memory usage is an idiotic way to measure memory usage of a browser.
In short, you should try to find someone competent to run your memory benchmarks.
GAA TAC ATC GCA CAT TAG TAT ATT GAG ACT
(Yes, I know that's DNA and not RNA, but it's really hard to make words with a U and no T).
So is Firefox if you have phishing detection turned on.
Anyway, I expect that many of these things will be in Chrome soon enough, once extension support is done.
BTW, one point I should mention is that although Chrome doesn't support NoScript right now, there's less of a need for it because of its improved security model. At least in my case, that's a lot better because I tried NoScript and it drove me insane within a week, so I removed it. YMMV
Why? It's either a troll or satire.
The "individualized web toolkit" is webkit, which is used by KDE, Nokia, and Safari.
Chrome is getting extension support (albeit not Firefox compatible) real soon now.
Furthermore, Chrome has Greasemonkey support to a degree already. See http://mashable.com/2008/12/15/google-chrome-greasemonkey-scripts/
Finally, Chrome has better privacy than Firefox in some ways because it has an anonymous browsing mode, and it's more secure because of the process and sandbox model they use. Firefox is working on that, but they're a long way behind.
Really, Firefox is falling pretty far behind many of the other browsers at this point. Don't get me wrong - it's still a good browser, and way better than IE, but all that it's got going for it now is the extension framework. Safari and Chrome are both way faster than Firefox. Even IE has a better process model (Firefox runs everything in a single thread, which is why it gets really sluggish with a lot of open tabs, or when one tab is really CPU heavy). Firefox is also a huge memory hog and, at least on Linux, is pretty unstable.
Again, most of these things will be fixed if they can fix their process model, but I expect that's a long ways off. They've got a lot of catching up to do.
There's always penis enlargement. Check your spam folder.
Yes, it's a range. Think about it. 14 GHz - 7 GHz is still 7 GHz, so it can be a particular frequency, but the range of frequencies uses the same unit. It's like how you can say something is 10 feet away or how something is 10 feet long, between 10 feet away and 20 feet away.
I've been kind of expecting/hoping for this for some time. Ever since the economy got so bad, I figured we might come to this. I'm actually amazed this didn't have before. You can't go that long with such incompetent management of a project and expect to be allowed to keep going, especially when you can't even be bothered to produce an actual time estimate for completion anymore except "When it's done".
Have you considered that:
and
might be related?
I didn't say the Android's UI was bad. I know reading carefully and recognizing nuance is a dying art, especially around here, but please try to pay at least some attnetion.
What I did say was that the Android was less polished than the iPhone. I've yet to see any other phone come close to the iPhone in that way. But I do like the Android quite a lot, TYVM.
You clearly are a programmer, but you're not very good at it. At least if your use of O notation is any indication. I've lost count of the number of times I had to jailbreak my iphone. I've lost several hours of my life doing all the nasty things required to make it work after it's eaten itself for no apparent reason. I've spent quite a bit of time reading poorly written blogs to see if they have finally figured out how to jailbreak the most recent update so I can install it, and whether I need to run some piece of software first pre-upgrade or else I might lose my data. And then I just stopped upgrading. Wasn't worth the hassle. Then I got a Google Android and it's unlocked, I can put anything I want on it, and I can use the shell or ssh or whatever without having to do anything except find an app in the marketplace.
Overall, I'd say I've spent a lot less time getting the Android to work than the iPhone. Of course, if you feel your time is not as valuable as mine, you're welcome to continue doing whatever you want.
You have to first jailbreak the phone if you want to unlock it. But I recently switched to Google Android so I don't have to deal with this. It's a less nice experience, but I imagine a lot of people who are willing to go through the trouble of jailbreaking a phone are also willing to put up with the less polished UI.
Have you stopped beating your wife yet?
Just a question. Seriously.
Very few people seem to realize Einstein was proved wrong there. Quantum mechanics does produce different results for the same action.
But it's amusing to watch stupid people quote Einstein for their own idiotic purposes, nonetheless.
Obama is pretty bright as far as he goes. Read his books or something. He's a pretty thoughtful guy.
But saying the democrats are smarter than the republicans seems kinda pointless. Arguing about which party is smarter is kind of like arguing about who's the valedictorian of the special ed class.
Really, now. This argument is a little silly.
TomTom has, apparently, been pursuing MS for a year. We have no idea what they've been saying or asking for. TomTom probably felt as you say - that it shouldn't file a lawsuit because this very thing would happen. So maybe they tried to bluff and MS called it. Or maybe they didn't think their terms were unreasonable but MS did.
We don't know what went on in these negotiations, but we do know it's been going on for a year with no license signed. Maybe it's one huge coincidence that MS was asked to pay royalties and then sued the very same company asking for them for patent violations when they have no prior history of using patents in a first-strike manner.
At the very least, if this is the start of something bigger, then MS went after TomTom first because they already were threatening to sue. But my money is on this just being an old fashioned throw-everything-at-the-wall-and-see-what-sticks lawsuit to force a settlement on terms more favorable to MS.
The funny thing is that, if the summary is true, it could have been TomTom, not MS, that shot first. Maybe MS suing TomTom was just retaliation for TomTom trying to collect royalties.
Q: How many bones are in the human body?
A: How does bones are in the human body make you feel?
There is one click buying where you preconfigure everything like which address to use, the shipping method, and the credit card. You have to explicitly enable it in your account with them.