> Trusted CAs aren't the epitome of web safety. In fact, they are LESS safe than one of those > "Invalid" (to use Mozilla's ill-chosen words) self-signed certificates under some circumstances.
If you have a secure channel to verify that the self-signed cert is correct, you can add it to your approved list and treat it as valid. Actually the ideal way to do things like internal self-signed certs for an intranet site is to set up your own private CA (preferably on a physically-secured, non-networked system), add it to the trusted list in the clients, and use it to sign the server certs.
And yes, the second choice would be to do an out-of-band verification (as you describe) of each cert before accepting it.
But in the real world this is all a WHOLE lot more trouble than most ordinary end-users are willing to go to in order to verify the cert for each and every public https site they are visiting.
So we have trusted CAs, because the other options are A) forget about https entirely and do everything with http, B) give the user a stupid pointless annoying "accept this cert" button to mindlessly click on (no they will NOT actually read the message before clicking the Accept button), or C) just trust all certs no matter what. These options are all considerably worse than trusting the certificate authorities.
Of course, those are the options for what the browsers can do by *default*. If you are a knowledgeable user with concerns about security, you can always remove the CA certs from your trusted list and inspect every cert manually. It's a lot of trouble, but good security isn't always convenient.
There's also the small matter of the extensive overuse of https. It seems almost every site that lets you log in wants to protect the login with https, even the login doesn't protect anything more important than your preferences for some of the site's features and maybe an "about me" blurb that other users of the site can see by clicking on your name. Every webmaster wants to feel that his site is so uber-important that it would be a disaster if somebody's login got compromised, but with a few exceptions (e.g., banks) this is mostly just arrogance. I suppose an argument could be made that users tend to use the same username/password pair for many services, and one of the *other* services the user uses with the same login credentials could actually matter. But on the whole I think this is outweighed by the fact that having tens of billions of https sites on the internet fundamentally makes the PKI vastly more difficult to oversee and protect. IMO, using https on sites where the user's account fundamentally will never ever matter (e.g., slashdot) is silly, unnecessary, and collectively (if a lot of sites do it) bad for the security of the whole internet.
> The problem is... the inability of these providers to peer with each other. > AT&T may have been down, but what about T-Mobile... When a major failure... locks > out only some... users in a given area, the problem is not technology but politics.
I agree with you this far. However...
> Why, given how critical cell phones are during an emergency, this is allowed to continue is beyond me.
Give me a break. Cell phones are a luxury item that just a few years ago almost nobody could afford, and now suddenly they're a critical service that can't be allowed to go down for a day, something homeland security should be concerned with keeping running in a natural-disaster scenario? Why, just because a lot of people have them now?
Get real. Normal phone lines to my knowledge have never been seen as critical to national security or emergency management, so why would cell phones? If robotic maids become cheap enough and popular enough that everybody and his dog has one, will those also then be critical to national security?
In the real world, if the power goes out for a few minutes during a storm we shrug, get out the candles, forget about the television and the internet for a couple of hours, and read a book or play a game or talk to one another (you know, face-to-face, verbally). You might want to hang up your cellphone and join us out here some time, just for a change of pace. It may be a bit different from what you're used to, but the video is pretty high-res.
> Rubiks cube is seldom encountered, as to place it in the game would probably result > in at least 2 points being knocked off the games score owing to frustration
Actually, I think it's kinda cool when a game includes an _optional_ really-really-hard puzzle, one that you can complete the entire game without solving, but which provides some easter egg or another if you do manage to solve it.
What I don't like is when there's a fairly difficult puzzle in the opening scene and you can't get *anywhere* in the game until you solve it. The flashlight battery in Curses, was almost like this, but it wasn't quite early enough in the game to be really bad, since you could explore a *fair* amount of territory, and several easier puzzles, before solving it.
> You'll get IE (cause you'll need something provided to go grab the stuff) > and you'll get a pretty plain OS otherwise. I'm a huge fan of that.
I would be, if getting the stuff you need consisted of ticking a few boxes in a package management tool and hitting the "install" button once. Think in terms of Synaptic or rpmdrake or cetera.
I realize it wouldn't be practical to include a full assortment of third-party software in such a scheme, nor probably stuff like Word (because of licensing issues), but anything Microsoft makes that's freely downloadable anyway *should* be able to be quickly and easily installed this way. It would be a major improvement over separately tracking down each item and running a separate installer for each one.
The uninstall UI is already pretty well present, in the Add/Remove Programs section of the control panel. But last I checked you can't really download and install stuff from there, except for things that MS considers to be OS components.
> It is called "sudo" and if your theoretical linux games would need root access > to install mods as well. Or do you run your linux box as root all the time?
Firefox on Linux doesn't need root access to install add-ons. Neither does Battle for Wesnoth. If the software is designed correctly, add-on content can be installed on a per-user basis, and you only need admin privs to install it system-wide. This should be true on Windows as well as on Unix systems.
But the big difference between UAC and sudo is that Unix systems have had privileges and security from day one, and it has been the normal expected thing that the user does not log in with administrative privileges for normal day-to-day activities, since the seventies. So most software is designed so that it doesn't *need* root privs. So you only get gksudo prompts when you try to do actual system administration stuff; the rest of the time it leaves you alone. Stuff that needs to happen in the background with root privileges is run via a mechanism that can provide that (e.g., cron), *not* out of the logged-in user's desktop session.
With Vista the notion that the user is normally unprivileged is a totally new way of thinking for many software developers. They were *supposed* to build software for Windows 2000 and for Windows XP under the assumption that the logged-in user might not be an administrator, but even quite major developers like Symantec (and, indeed, even Microsoft) never really got this completely right, and there's a lot of software out there that gets it very BADLY wrong, to the extent that some software simply CANNOT be run from a limited account at all. (Ostensibly educational games are particularly bad for this.)
This was a problem before, including under XP, but Vista forces the issue because the user *can't* work around it by just logging in as Administrator all the time (which they *shouldn't* have been doing all along, but people were). This is what needed to be done, but it'll take a while for all the third-party Windows-based software developers to update their whole way of thinking and finally get this right.
UAC, in the long term, is good. It's necessary, and it's the way Windows *desperately* needed to go. But yes, it's going to cause some pain for a few years until all the third-party software gets redesigned under the new way of thinking. Seven isn't really going to change this one way or another, because it's the application software that needs to be updated to fix the too-many-UAC-prompts problem.
> Traditionally, when you buy something, you pay one time or one total, and it > becomes yours. This rake is $5+tax. It can be yours for that much.
The one-time nature of this traditional arrangement has been eroding for a while now. When I was about eight years old, my dad bought two leaf rakes, a blue one and a green one. They eventually wore out, but by that time I was in college. Meanwhile my grandfather was still using one that he bought when my mom was in gradeschool.
The last leaf rake we bought lasted about two years.
Have you noticed that you can't buy cheap wire shirt hangars that last forever any more? All they sell are the plastic ones that break.
I don't think there's exactly a deliberate conspiracy to do this. I think it's more a result of the market-for-lemons phenomenon combined with a shift in our culture away from placing *value* on durability and permanence, toward caring more about having the latest and greatest and trendiest stuff.
> It was either buy food or rent MS Word for three hours, and I didn't want to starve.
With that line the teacher can offer you food, which leaves you with no excuse next time. What you want to say is, "We were all out of computer time, and we couldn't buy more until mom gets paid Friday." This one can be used every week (well, until the teacher hands you a Knoppix CD).
> When I was a kid, we were assigned ~400 hours of homework a year. From what I hear, it's more now.
Well, there's what they're assigned, which varies by grade and teacher, and then there's what they actually do, which varies by student. I'm not convinced there's any correlation between the two.
They're *the* local telephone company across multiple states. They have no direct competitors in that market.
I suppose cellular providers and cable providers will try to take advantage of this, but cellphonscht kshcht bzsakt shchtkischt rural kschischt bzczoscht, and cable providers only offer internet and maybe VOIP (err, and television if you're into that), which in the general case are not necessarily very good substitutes for an actual phone line.
Having said that, the fact that Fairpoint is a spinoff from Verizon makes me very glad I stopped doing business with Verizon a couple of years ago. This is a cut and dry case of abusing a monopoly in one market (phone lines) to elbow your way into (and competitors out of) another market (email), so hopefully they'll get the book thrown at them good and hard. Full-blown net neutrality may not even be necessary in this particular case; standard antitrust regulation should be good enough, I would think. (But IANAL,ATINLA.)
> a drooling, knuckle dragging luddite... a step away from churning your own butter.
Well, I've never been to a barn raising, and I don't churn butter. I do make home-made ice cream, though, does that count? We also make and can our own spaghetti sauce, and applesauce. I don't have a cell phone because it's bad enough we have a landline; one of my life goals is to someday live in a house with no phones whatsoever. And I prefer the 80-key XT keyboard over the newer 101-key layout. And I still use Perl, haven't bothered to learn Ruby yet.
But yeah, I tried last.fm, and xmms still scrobbles what I listen to, although I've become pretty well disabused of the notion that I will ever discover any music there that I would actually like. Neighbor Radio mostly plays boring non-contrapunctal drivel, Brahms and Mozart and so forth, even though I've scrobbled mostly Bach and Scarlatti. I get significantly better results with the "people who shopped for this also bought" feature on Amazon.
> I wonder if RedHat would be justified in a trademark lawsuit
Not likely. Of all the kinds of words that it's normal to see in the names of multiple companies and products, stars and constellations are probably third on the list. (The top slot on the list, of course, goes to common surnames, and second is probably common given names.) The only other major contender I can think of that might beat out stars and constellations would be animals, but of course a cygnus is an animal as _well_ as a constellation, so there's no help there.
> And then immediately lawsuits are pressed against Canonical, Debian,...
Debian would never allow something like this into the distro in the first place. They're _very_ picky about such things. (You can't even get Firefox as such from an official Debian repository, because of obscure trademark licensing issues somehow related to the fact that the Debian security team releases security updates. Although what you do get is functionally equivalent until you run into some stupid websites that engages in egregious sniffing, e.g., Hotmail. Not that the average Debian user is tremendously likely to use Hotmail anwyay...)
So if you wanted to install something like that on a Debian system, you'd have to use an unofficial repository that the Debian folks don't have control over -- presumably operated by someone in a jurisdiction that doesn't enforce software patents. (China seems like an obvious place for someone to run such a repository...) You'd be in violation of the patent if you use such a repository and install anything from it, but for an individual user wanting icon previews on a single workstation this would probably not be perceived as a very large danger. (Deploying it across an enterprise would, of course, be potentially a whole different thing, and probably unwise.)
> Will VHS be missed? Not... with videos being brittle, clunky, and rather user-unfriendly
What's the basis of comparison? I would argue that at this point the major alternative is DVD, which is a lot *MORE* brittle and *extremely* user-unfriendly. And if there's anything clunkier than the assortment of different kinds of not-so-good cases that DVDs come in, I'm sure I don't know what it is. Really VHS would be a huge improvement over DVD.
In terms of user-friendliness, VHS is pretty hard to beat. You put the thing in and it *just plays*. No fiddling, no menus to navigate, no figuring out which buttons do what (which actually varies from one DVD to another), and if you want to fast forward or rewind you *can*, but if you don't want to mess with buttons the tape will play to the end and, in most VCRs, automatically rewind and eject when it's done, so you literally don't have to push a single button. You just put the tape in the player and that's *all* you have to do. Okay, so you do have to manage to not put the tape in backwards, which I guess might be hard if you have the same level of intelligence as a garden slug.
The brittleness claim would have more merit, if the alternative weren't DVD. Sure, if you beat up on a VHS tape too much it'll break. But at least you don't have to be too worried about what happens if you get a *scratch* on it, for crying out loud.
I suppose "clunky" can mean a lot of different things...
> I hope our far-left overlords are more benevolent than the far-right ones were.
Oh, man, if you think Bush is far-right, you really have no idea how far *right* the right can go.
Far right would be someone who wants to put the dollar back on the gold standard, put retirement planning back in the hands of individual workers, decouple health insurance from employment so that people can make their own choices about what coverage to buy, require the jurors on malpractice suits to be licensed medical doctors (the defendant's peers), deregulate the automobile industry (specifically, stop telling the manufacturers what kinds of vehicles they can and can't build and sell), outlaw closed-shop union contracts so that workers can make their own individual choice whether to unionize or not, and cut the federal budget across the board 30% the first year, and then turn around and cut it 30% again the next year, and again the next, using the resulting surplus to pay down the national debt and cut income taxes. Oh, and do I need to even mention that he'd want to make the income tax rate the same for all income brackets?
No, we're not in any danger of getting someone like that as President. He could get the vote in the conservative states, but he'd never win in the moderate swing states, and the GOP knows this and would never give such a person the party nomination, not with the electoral college system in place in its current form.
But that's where the far right is. And those are just the economic issues. Do you even want to think about where the far right is on social issues? Just knowing where the far right stands on the issue of divorce could make a liberal's blood boil.
> I believe all of the other posts are omitting steps (3) and (5),
Where I come from step 3 is called "speeding". It's not okay to speed just because you're passing someone. People do it all the time, of course... and complain when they get caught and ticketed. I have no sympathy at all.
If the guy ahead of you is going so close to the speed limit that you can't pass without speeding, then you do not need to pass him. Even if you have to follow the guy for an entire hour, those 3mph are only going to cost you two and a half minutes. Get over it. If that kind of impatience doesn't get you killed, it will at least give you an ulcer. Chill out a little.
> > a deer suddenly jumps from the trees beside the road in front of your car, > > or an idiot in the right lane decides to move into your lane.
> Both of those are foreseeable possibilities based on conditions. > Bad conditions obviously mean you need to be more careful.
If things like there being trees along the side of the road or drivers in another lane qualify as "bad conditions", it's probably best to just figure on bad conditions pretty much all the time. Those aren't exactly unusual circumstances, and there are plenty more similarly dangerous and similarly common kinds of situations where they came from. Night, going over a hill (you never know what's just over the top until you're pretty much there), rain, fog, night, the sun in your eyes, potholes, kids on bikes, dogs, cats, coons, possums, flat tires, small pieces of road debris (e.g., gravel) flying up from the car ahead and hitting the windshield,...
Stuff happens, in other words.
If you leave a bit more space between your car and the next one, you'll have that much more time to react and deal with whatever situation may arise. It's just safer. And the amount of time it costs you, even on a quite long trip, is generally going to be measured in seconds.
20mph on the highway is still stupid and dangerous.
Yes, it takes longer to change speed or direction on snow (although if your braking "does nothing", you're doing it wrong; snow is not asphalt, and treating as though it is does not yield good results). You need to leave yourself more room to stop, which means more room between vehicles, so there's more room to stop. (Yes, there's room on the highway when it's snowy to leave more room between vehicles than usual, because of all the people who aren't on the road because of the snow, and if NOT then the heavy traffic will have eradicated the snow and the pavement will be dry, like the interstates typically are even when the snow is still coming down.) You also need to be careful and have some experience on snow. People who have not walked and run and played on snow should NOT attempt to drive on it. If you just moved up from Georgia and have never seen so much white stuff before in your life, please don't drive on it until you've done enough walking and running on it to be comfortable with the physics. This generally takes at least a couple of months, even if you're fairly bright.
When turning around while running on a thin layer of wet snow (on a smooth surface like pavement) is something easy that you do without thinking, and without looking at your feet, then maybe you're ready to try driving on it.
> This rule will not allow sufficient traffic to move at posted speeds with the > number of lanes available in most US highway configurations.
I don't know about "most", but you are certainly correct that there are sometimes real-world traffic conditions do not allow the rule to be followed. So let me add a second rule of thumb for those situations:
If you are repeatedly cut off by people pulling into the space ahead of you and driving there (not just an idiot driver being foolish, but a pattern driven by the number of cars on the road), then you need to leave a little less space, even if it's not ideal, because it's safer than being repeatedly cut off like that.
> So while the rule is a great guide for safety purposes, it is effectively useless
It's useless when traffic conditions will not allow you to follow it. In that case, use the second rule.
But when you *can* follow the carlength-per-10mph rule, you generally should.
> I'm not sure what reasons you may have for passing someone, but generally when I > pass someone, it is because they are traveling at a lower speed than I would prefer.
There may also be times when people might pass a vehicle because they don't want to be behind it. My dad doesn't like to be behind semi trucks, for instance, partly because they obstruct his view of the road ahead (yes, he's a tailgater something aweful) but also partly because they tend not to maintain a constant speed, since they slow down uphill and speed up downhill. He doesn't like that.
> In passing them, I have already accelerated to a higher speed than they are traveling. > When I merge back over, I am still moving faster than they are- because of this, there > is no reason for them to unexpectedly slow down.
The problem is what happens if *you* have to suddenly slow down, and they are behind you. You don't *expect* that to happen (or you wouldn't have passed them), but sometimes things occur that you did not anticipate. Safe drivers leave a bit of margin to give people time to react to unexpected events that may occur.
> The only way I can see this becoming an issue is if you're accelerating to pass somebody, > then slowing down to the same or a slower speed than that person is traveling. If that's > the case, you're just being an asshole.
I will agree that pulling out in front of somebody and then slowing down to their speed or less is at best extremely rude and at worst very dangerous. (FWIW, though there are things I don't like about my dad's driving, he does not commit this particular transgression. When he passes, he invariably continues afterward to move faster than the vehicle he has just passed until he is a considerable distance ahead of them.)
On the contrary, the carlength-per-10mph rule is a pretty good rule of thumb, although if your brakes and/or reaction time are bad it can be inadequate.
> You don't have to stop in the distance between you and the car ahead of you
No, but...
> You don't have to stop in the distance between you and the car ahead of you. > You have to stop in the distance between you plus the distance it takes him to stop.
That's also wrong, and an extremely dangerous notion to keep in your head if you're a driver.
In fact, you have to stop in the distance between you and the car ahead of you, plus the distance the car ahead of you travels while stopping, *less* the distance you travel during your reaction time. The distance you travel during your reaction time is about a carlength per 10mph, more if you're slow on the uptake.
So the rule of thumb actually assumes your brakes are at least as good as the ones on the car ahead of you. If your brakes are bad, you should theoretically try to leave *more* distance if you can. Although in practice sometimes traffic conditions are such that you simply cannot leave much more than 4-5 carlengths, because somebody will promptly pull into that space in front of you and drive there; if that's happening repeatedly, it's better to leave a little less space, even though it's not ideal, because it's safer than getting repeatedly cut off like that.
> If someone else can zip around you, you had plenty of space to safely move to the right.
This is absolutely untrue. At 65mph the minimum safe distance is about six carlengths. If you can't see their heighlights in your rearview mirror, you should not be pulling over in front of them.
> and there's no reason for you to slow down or stop in the near future
You're supposed to drive as if you don't know the future. Because you don't.
> (easily determined by looking ahead)
Apparently you live in a bizarro universe where things like deer never run out onto the road suddenly?
Well, over here in the real world we don't always know ahead of time when we're going to need to slow down.
> especially if you're doing a slow pass instead of finishing the damn maneuver and getting over to the right
By "slow pass" I assume you mean passing without speeding -- which is the only legally acceptable way to do it.
> Trusted CAs aren't the epitome of web safety. In fact, they are LESS safe than one of those
> "Invalid" (to use Mozilla's ill-chosen words) self-signed certificates under some circumstances.
If you have a secure channel to verify that the self-signed cert is correct, you can add it to your approved list and treat it as valid. Actually the ideal way to do things like internal self-signed certs for an intranet site is to set up your own private CA (preferably on a physically-secured, non-networked system), add it to the trusted list in the clients, and use it to sign the server certs.
And yes, the second choice would be to do an out-of-band verification (as you describe) of each cert before accepting it.
But in the real world this is all a WHOLE lot more trouble than most ordinary end-users are willing to go to in order to verify the cert for each and every public https site they are visiting.
So we have trusted CAs, because the other options are A) forget about https entirely and do everything with http, B) give the user a stupid pointless annoying "accept this cert" button to mindlessly click on (no they will NOT actually read the message before clicking the Accept button), or C) just trust all certs no matter what. These options are all considerably worse than trusting the certificate authorities.
Of course, those are the options for what the browsers can do by *default*. If you are a knowledgeable user with concerns about security, you can always remove the CA certs from your trusted list and inspect every cert manually. It's a lot of trouble, but good security isn't always convenient.
There's also the small matter of the extensive overuse of https. It seems almost every site that lets you log in wants to protect the login with https, even the login doesn't protect anything more important than your preferences for some of the site's features and maybe an "about me" blurb that other users of the site can see by clicking on your name. Every webmaster wants to feel that his site is so uber-important that it would be a disaster if somebody's login got compromised, but with a few exceptions (e.g., banks) this is mostly just arrogance. I suppose an argument could be made that users tend to use the same username/password pair for many services, and one of the *other* services the user uses with the same login credentials could actually matter. But on the whole I think this is outweighed by the fact that having tens of billions of https sites on the internet fundamentally makes the PKI vastly more difficult to oversee and protect. IMO, using https on sites where the user's account fundamentally will never ever matter (e.g., slashdot) is silly, unnecessary, and collectively (if a lot of sites do it) bad for the security of the whole internet.
Don't you read kerneltrap? What kind of geek are you? I suppose next you're going to say you don't know Damian Conway?
> The problem is... the inability of these providers to peer with each other.
> AT&T may have been down, but what about T-Mobile... When a major failure... locks
> out only some... users in a given area, the problem is not technology but politics.
I agree with you this far. However...
> Why, given how critical cell phones are during an emergency, this is allowed to continue is beyond me.
Give me a break. Cell phones are a luxury item that just a few years ago almost nobody could afford, and now suddenly they're a critical service that can't be allowed to go down for a day, something homeland security should be concerned with keeping running in a natural-disaster scenario? Why, just because a lot of people have them now?
Get real. Normal phone lines to my knowledge have never been seen as critical to national security or emergency management, so why would cell phones? If robotic maids become cheap enough and popular enough that everybody and his dog has one, will those also then be critical to national security?
In the real world, if the power goes out for a few minutes during a storm we shrug, get out the candles, forget about the television and the internet for a couple of hours, and read a book or play a game or talk to one another (you know, face-to-face, verbally). You might want to hang up your cellphone and join us out here some time, just for a change of pace. It may be a bit different from what you're used to, but the video is pretty high-res.
> Rubiks cube is seldom encountered, as to place it in the game would probably result
> in at least 2 points being knocked off the games score owing to frustration
Actually, I think it's kinda cool when a game includes an _optional_ really-really-hard puzzle, one that you can complete the entire game without solving, but which provides some easter egg or another if you do manage to solve it.
What I don't like is when there's a fairly difficult puzzle in the opening scene and you can't get *anywhere* in the game until you solve it. The flashlight battery in Curses, was almost like this, but it wasn't quite early enough in the game to be really bad, since you could explore a *fair* amount of territory, and several easier puzzles, before solving it.
> I suppose .NET was always supposed to be the way out for Microsoft... ... but .NET is catching on about as fast as IPv6.
> You'll get IE (cause you'll need something provided to go grab the stuff)
> and you'll get a pretty plain OS otherwise. I'm a huge fan of that.
I would be, if getting the stuff you need consisted of ticking a few boxes in a package management tool and hitting the "install" button once. Think in terms of Synaptic or rpmdrake or cetera.
I realize it wouldn't be practical to include a full assortment of third-party software in such a scheme, nor probably stuff like Word (because of licensing issues), but anything Microsoft makes that's freely downloadable anyway *should* be able to be quickly and easily installed this way. It would be a major improvement over separately tracking down each item and running a separate installer for each one.
The uninstall UI is already pretty well present, in the Add/Remove Programs section of the control panel. But last I checked you can't really download and install stuff from there, except for things that MS considers to be OS components.
> It is called "sudo" and if your theoretical linux games would need root access
> to install mods as well. Or do you run your linux box as root all the time?
Firefox on Linux doesn't need root access to install add-ons. Neither does Battle for Wesnoth. If the software is designed correctly, add-on content can be installed on a per-user basis, and you only need admin privs to install it system-wide. This should be true on Windows as well as on Unix systems.
But the big difference between UAC and sudo is that Unix systems have had privileges and security from day one, and it has been the normal expected thing that the user does not log in with administrative privileges for normal day-to-day activities, since the seventies. So most software is designed so that it doesn't *need* root privs. So you only get gksudo prompts when you try to do actual system administration stuff; the rest of the time it leaves you alone. Stuff that needs to happen in the background with root privileges is run via a mechanism that can provide that (e.g., cron), *not* out of the logged-in user's desktop session.
With Vista the notion that the user is normally unprivileged is a totally new way of thinking for many software developers. They were *supposed* to build software for Windows 2000 and for Windows XP under the assumption that the logged-in user might not be an administrator, but even quite major developers like Symantec (and, indeed, even Microsoft) never really got this completely right, and there's a lot of software out there that gets it very BADLY wrong, to the extent that some software simply CANNOT be run from a limited account at all. (Ostensibly educational games are particularly bad for this.)
This was a problem before, including under XP, but Vista forces the issue because the user *can't* work around it by just logging in as Administrator all the time (which they *shouldn't* have been doing all along, but people were). This is what needed to be done, but it'll take a while for all the third-party Windows-based software developers to update their whole way of thinking and finally get this right.
UAC, in the long term, is good. It's necessary, and it's the way Windows *desperately* needed to go. But yes, it's going to cause some pain for a few years until all the third-party software gets redesigned under the new way of thinking. Seven isn't really going to change this one way or another, because it's the application software that needs to be updated to fix the too-many-UAC-prompts problem.
> Traditionally, when you buy something, you pay one time or one total, and it
> becomes yours. This rake is $5+tax. It can be yours for that much.
The one-time nature of this traditional arrangement has been eroding for a while now. When I was about eight years old, my dad bought two leaf rakes, a blue one and a green one. They eventually wore out, but by that time I was in college. Meanwhile my grandfather was still using one that he bought when my mom was in gradeschool.
The last leaf rake we bought lasted about two years.
Have you noticed that you can't buy cheap wire shirt hangars that last forever any more? All they sell are the plastic ones that break.
I don't think there's exactly a deliberate conspiracy to do this. I think it's more a result of the market-for-lemons phenomenon combined with a shift in our culture away from placing *value* on durability and permanence, toward caring more about having the latest and greatest and trendiest stuff.
But I do think it's a bad thing.
> It was either buy food or rent MS Word for three hours, and I didn't want to starve.
With that line the teacher can offer you food, which leaves you with no excuse next time. What you want to say is, "We were all out of computer time, and we couldn't buy more until mom gets paid Friday." This one can be used every week (well, until the teacher hands you a Knoppix CD).
> When I was a kid, we were assigned ~400 hours of homework a year. From what I hear, it's more now.
Well, there's what they're assigned, which varies by grade and teacher, and then there's what they actually do, which varies by student. I'm not convinced there's any correlation between the two.
> The competitors should be advertising
They're *the* local telephone company across multiple states. They have no direct competitors in that market.
I suppose cellular providers and cable providers will try to take advantage of this, but cellphonscht kshcht bzsakt shchtkischt rural kschischt bzczoscht, and cable providers only offer internet and maybe VOIP (err, and television if you're into that), which in the general case are not necessarily very good substitutes for an actual phone line.
Having said that, the fact that Fairpoint is a spinoff from Verizon makes me very glad I stopped doing business with Verizon a couple of years ago. This is a cut and dry case of abusing a monopoly in one market (phone lines) to elbow your way into (and competitors out of) another market (email), so hopefully they'll get the book thrown at them good and hard. Full-blown net neutrality may not even be necessary in this particular case; standard antitrust regulation should be good enough, I would think. (But IANAL,ATINLA.)
> a drooling, knuckle dragging luddite... a step away from churning your own butter.
Well, I've never been to a barn raising, and I don't churn butter. I do make home-made ice cream, though, does that count? We also make and can our own spaghetti sauce, and applesauce. I don't have a cell phone because it's bad enough we have a landline; one of my life goals is to someday live in a house with no phones whatsoever. And I prefer the 80-key XT keyboard over the newer 101-key layout. And I still use Perl, haven't bothered to learn Ruby yet.
But yeah, I tried last.fm, and xmms still scrobbles what I listen to, although I've become pretty well disabused of the notion that I will ever discover any music there that I would actually like. Neighbor Radio mostly plays boring non-contrapunctal drivel, Brahms and Mozart and so forth, even though I've scrobbled mostly Bach and Scarlatti. I get significantly better results with the "people who shopped for this also bought" feature on Amazon.
IMO, Web TV doesn't count as a Microsoft failure. It had pretty much already failed before Microsoft bought it.
> I live in Canada, we are not far away from this problem here also, except that our movies/music sucks even more.
Wait... Canada produces music and movies?
> I wonder if RedHat would be justified in a trademark lawsuit
Not likely. Of all the kinds of words that it's normal to see in the names of multiple companies and products, stars and constellations are probably third on the list. (The top slot on the list, of course, goes to common surnames, and second is probably common given names.) The only other major contender I can think of that might beat out stars and constellations would be animals, but of course a cygnus is an animal as _well_ as a constellation, so there's no help there.
> And then immediately lawsuits are pressed against Canonical, Debian, ...
Debian would never allow something like this into the distro in the first place. They're _very_ picky about such things. (You can't even get Firefox as such from an official Debian repository, because of obscure trademark licensing issues somehow related to the fact that the Debian security team releases security updates. Although what you do get is functionally equivalent until you run into some stupid websites that engages in egregious sniffing, e.g., Hotmail. Not that the average Debian user is tremendously likely to use Hotmail anwyay...)
So if you wanted to install something like that on a Debian system, you'd have to use an unofficial repository that the Debian folks don't have control over -- presumably operated by someone in a jurisdiction that doesn't enforce software patents. (China seems like an obvious place for someone to run such a repository...) You'd be in violation of the patent if you use such a repository and install anything from it, but for an individual user wanting icon previews on a single workstation this would probably not be perceived as a very large danger. (Deploying it across an enterprise would, of course, be potentially a whole different thing, and probably unwise.)
> Will VHS be missed? Not ... with videos being brittle, clunky, and rather user-unfriendly
What's the basis of comparison? I would argue that at this point the major alternative is DVD, which is a lot *MORE* brittle and *extremely* user-unfriendly. And if there's anything clunkier than the assortment of different kinds of not-so-good cases that DVDs come in, I'm sure I don't know what it is. Really VHS would be a huge improvement over DVD.
In terms of user-friendliness, VHS is pretty hard to beat. You put the thing in and it *just plays*. No fiddling, no menus to navigate, no figuring out which buttons do what (which actually varies from one DVD to another), and if you want to fast forward or rewind you *can*, but if you don't want to mess with buttons the tape will play to the end and, in most VCRs, automatically rewind and eject when it's done, so you literally don't have to push a single button. You just put the tape in the player and that's *all* you have to do. Okay, so you do have to manage to not put the tape in backwards, which I guess might be hard if you have the same level of intelligence as a garden slug.
The brittleness claim would have more merit, if the alternative weren't DVD. Sure, if you beat up on a VHS tape too much it'll break. But at least you don't have to be too worried about what happens if you get a *scratch* on it, for crying out loud.
I suppose "clunky" can mean a lot of different things...
> I hope our far-left overlords are more benevolent than the far-right ones were.
Oh, man, if you think Bush is far-right, you really have no idea how far *right* the right can go.
Far right would be someone who wants to put the dollar back on the gold standard, put retirement planning back in the hands of individual workers, decouple health insurance from employment so that people can make their own choices about what coverage to buy, require the jurors on malpractice suits to be licensed medical doctors (the defendant's peers), deregulate the automobile industry (specifically, stop telling the manufacturers what kinds of vehicles they can and can't build and sell), outlaw closed-shop union contracts so that workers can make their own individual choice whether to unionize or not, and cut the federal budget across the board 30% the first year, and then turn around and cut it 30% again the next year, and again the next, using the resulting surplus to pay down the national debt and cut income taxes. Oh, and do I need to even mention that he'd want to make the income tax rate the same for all income brackets?
No, we're not in any danger of getting someone like that as President. He could get the vote in the conservative states, but he'd never win in the moderate swing states, and the GOP knows this and would never give such a person the party nomination, not with the electoral college system in place in its current form.
But that's where the far right is. And those are just the economic issues. Do you even want to think about where the far right is on social issues? Just knowing where the far right stands on the issue of divorce could make a liberal's blood boil.
> I believe all of the other posts are omitting steps (3) and (5),
Where I come from step 3 is called "speeding". It's not okay to speed just because you're passing someone. People do it all the time, of course... and complain when they get caught and ticketed. I have no sympathy at all.
If the guy ahead of you is going so close to the speed limit that you can't pass without speeding, then you do not need to pass him. Even if you have to follow the guy for an entire hour, those 3mph are only going to cost you two and a half minutes. Get over it. If that kind of impatience doesn't get you killed, it will at least give you an ulcer. Chill out a little.
> > a deer suddenly jumps from the trees beside the road in front of your car,
...
> > or an idiot in the right lane decides to move into your lane.
> Both of those are foreseeable possibilities based on conditions.
> Bad conditions obviously mean you need to be more careful.
If things like there being trees along the side of the road or drivers in another lane qualify as "bad conditions", it's probably best to just figure on bad conditions pretty much all the time. Those aren't exactly unusual circumstances, and there are plenty more similarly dangerous and similarly common kinds of situations where they came from. Night, going over a hill (you never know what's just over the top until you're pretty much there), rain, fog, night, the sun in your eyes, potholes, kids on bikes, dogs, cats, coons, possums, flat tires, small pieces of road debris (e.g., gravel) flying up from the car ahead and hitting the windshield,
Stuff happens, in other words.
If you leave a bit more space between your car and the next one, you'll have that much more time to react and deal with whatever situation may arise. It's just safer. And the amount of time it costs you, even on a quite long trip, is generally going to be measured in seconds.
20mph on the highway is still stupid and dangerous.
Yes, it takes longer to change speed or direction on snow (although if your braking "does nothing", you're doing it wrong; snow is not asphalt, and treating as though it is does not yield good results). You need to leave yourself more room to stop, which means more room between vehicles, so there's more room to stop. (Yes, there's room on the highway when it's snowy to leave more room between vehicles than usual, because of all the people who aren't on the road because of the snow, and if NOT then the heavy traffic will have eradicated the snow and the pavement will be dry, like the interstates typically are even when the snow is still coming down.) You also need to be careful and have some experience on snow. People who have not walked and run and played on snow should NOT attempt to drive on it. If you just moved up from Georgia and have never seen so much white stuff before in your life, please don't drive on it until you've done enough walking and running on it to be comfortable with the physics. This generally takes at least a couple of months, even if you're fairly bright.
When turning around while running on a thin layer of wet snow (on a smooth surface like pavement) is something easy that you do without thinking, and without looking at your feet, then maybe you're ready to try driving on it.
> This rule will not allow sufficient traffic to move at posted speeds with the
> number of lanes available in most US highway configurations.
I don't know about "most", but you are certainly correct that there are sometimes real-world traffic conditions do not allow the rule to be followed. So let me add a second rule of thumb for those situations:
If you are repeatedly cut off by people pulling into the space ahead of you and driving there (not just an idiot driver being foolish, but a pattern driven by the number of cars on the road), then you need to leave a little less space, even if it's not ideal, because it's safer than being repeatedly cut off like that.
> So while the rule is a great guide for safety purposes, it is effectively useless
It's useless when traffic conditions will not allow you to follow it. In that case, use the second rule.
But when you *can* follow the carlength-per-10mph rule, you generally should.
> I'm not sure what reasons you may have for passing someone, but generally when I
> pass someone, it is because they are traveling at a lower speed than I would prefer.
There may also be times when people might pass a vehicle because they don't want to be behind it. My dad doesn't like to be behind semi trucks, for instance, partly because they obstruct his view of the road ahead (yes, he's a tailgater something aweful) but also partly because they tend not to maintain a constant speed, since they slow down uphill and speed up downhill. He doesn't like that.
> In passing them, I have already accelerated to a higher speed than they are traveling.
> When I merge back over, I am still moving faster than they are- because of this, there
> is no reason for them to unexpectedly slow down.
The problem is what happens if *you* have to suddenly slow down, and they are behind you. You don't *expect* that to happen (or you wouldn't have passed them), but sometimes things occur that you did not anticipate. Safe drivers leave a bit of margin to give people time to react to unexpected events that may occur.
> The only way I can see this becoming an issue is if you're accelerating to pass somebody,
> then slowing down to the same or a slower speed than that person is traveling. If that's
> the case, you're just being an asshole.
I will agree that pulling out in front of somebody and then slowing down to their speed or less is at best extremely rude and at worst very dangerous. (FWIW, though there are things I don't like about my dad's driving, he does not commit this particular transgression. When he passes, he invariably continues afterward to move faster than the vehicle he has just passed until he is a considerable distance ahead of them.)
> That's a stupid rule.
On the contrary, the carlength-per-10mph rule is a pretty good rule of thumb, although if your brakes and/or reaction time are bad it can be inadequate.
> You don't have to stop in the distance between you and the car ahead of you
No, but...
> You don't have to stop in the distance between you and the car ahead of you.
> You have to stop in the distance between you plus the distance it takes him to stop.
That's also wrong, and an extremely dangerous notion to keep in your head if you're a driver.
In fact, you have to stop in the distance between you and the car ahead of you, plus the distance the car ahead of you travels while stopping, *less* the distance you travel during your reaction time. The distance you travel during your reaction time is about a carlength per 10mph, more if you're slow on the uptake.
So the rule of thumb actually assumes your brakes are at least as good as the ones on the car ahead of you. If your brakes are bad, you should theoretically try to leave *more* distance if you can. Although in practice sometimes traffic conditions are such that you simply cannot leave much more than 4-5 carlengths, because somebody will promptly pull into that space in front of you and drive there; if that's happening repeatedly, it's better to leave a little less space, even though it's not ideal, because it's safer than getting repeatedly cut off like that.
> If someone else can zip around you, you had plenty of space to safely move to the right.
This is absolutely untrue. At 65mph the minimum safe distance is about six carlengths. If you can't see their heighlights in your rearview mirror, you should not be pulling over in front of them.
> and there's no reason for you to slow down or stop in the near future
You're supposed to drive as if you don't know the future. Because you don't.
> (easily determined by looking ahead)
Apparently you live in a bizarro universe where things like deer never run out onto the road suddenly?
Well, over here in the real world we don't always know ahead of time when we're going to need to slow down.
> especially if you're doing a slow pass instead of finishing the damn maneuver and getting over to the right
By "slow pass" I assume you mean passing without speeding -- which is the only legally acceptable way to do it.