It's not artificially-inspired confidence. I have to brake suddenly less often, I glance around me and am unsure if I've accidentally missed my exit more often, I accidentally switch lanes with another car in my blind spot less often, I notice deer on the side of the road before they charge in front of me (especially in some seasons there are as many as 2-3 new deer roadkill along my drive each day). I'm more aware of the road in ways which are demonstrable and not a product of my imagination as you suggest it could only be.
On long drives, an audiobook keeps my brain aware, I don't get as glazed over, I don't find myself fighting to stay awake, and I don't just drive on autopilot. An audiobook doesn't require nearly as much attention as personal conversation, and it never requires you to take your eyes off the road such as in your example.
I'll agree that if you're on an unfamiliar course, you could put those brain cycles to better use - if I had to do something which required a higher level of brain faculty, such as drive through a detour, then I'd pause the book. But when you're doing the same path you've done literally hundreds of times before, and you're at it for three hours each day, your attention wanders and it increases your risks. When your brain doesn't have anything to focus on, it shuts down and you enter a bit of a stupor which requires a short but possibly fatal amount of time to recover from.
No, what I mean is that I'd be startled by the car in front of me braking less frequently, and not need to slam on my brakes, or realizing I was drifting over the center line, or suddenly looking around me and not being certain if I had missed my exit because I didn't immediately recognize my surroundings.
Things which I couldn't have helped noticing before because they would have made themselves known to me eventually if I had missed them.
I have to disagree with at least one point for serious reasons.
I drove 3 hours a day for 4 years. About 6 months into this I started listening to books on tape, and I found my alertness level while driving was improved significantly. When I was just listening to the radio or my ipod, and it was the same stuff I've heard a thousand times before, my mind drifted. When I started keeping my mind awake and aware with audiobooks, I found I was surprised by traffic around me much less often.
I touted this to several coworkers who also had long drives, and collectively we all agreed: audiobooks keep your mind more active, and increase your overall awareness of arising traffic situations, we found ourselves in fewer close calls and surprised by things around us less often.
Because of big business. Big business demands to run fully-patched software, but also demands that software not be upgraded before every application the business users use has been tested with it.
Most big business upgrade cycle is 2-5 years behind cutting edge. We got XP last year.
I keep IE 6 running in my VirtualBox instance of XP. If a site works in IE 6, it almost certainly works in IE 7. The converse is not true, and since I have testing to do, that's a safer platform to test on.
I bought Bioshock, it installed securom, and I couldn't play it because I had too many developer hooks in my system. I fought with support who essentially said the only recourse was to reinstall Windows.
So I got a pirated copy, and all was well.
Game copy restrictions don't do anything to stop pirates, they delay the cracked version by a matter of a week, maybe days, perhaps even only hours. They only punish legitimate customers.
People who have money in their fist and want to give it to the game company are being negatively conditioned away from doing so. People realize that crackers do a better job of providing a seamless and error free customer interface than the company who takes their money for failing to provide this service.
When you decide one day that you want to play a certain game, now the decision comes down to: doing it legitimately with a reasonable chance that it's also unsuccessful, and you're out the cost of the game (since it's now open, the store won't take it back), or doing it illegitimately with an almost certain chance that it's successful, and even if it's not, at least you're not out any money.
But they have to prove that you didn't buy that music. An affirmative defense is that your music CD's were stolen. They would have a very, very hard time busting you for music you merely have a copy of, the burden of proof is too high.
Even if they could show that for example the media had "Downloaded from MuzicWarez - Your place for pirated music," in its metadata, they'd have a hard time convincing a jury to convict you if your defense was that you downloaded that track after you lost the original album.
However, distribution is something they can bust you on, because unless you're an online music retailer, you don't have any license to be distributing music electronically.
Sneaker net ping times are terrible, and the search engine sucks, but once a transfer starts, the bandwidth can't be beat. In a school setting, it's not too hard to find someone with your music taste to combine resources with.
I believe the relevant quote is: Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway. - Tanenbaum, Andrew S. (1996).
No way, they know exactly why they made it, and they do in fact need it very badly.
Flex + Flash's ubiquity + Adobe Air = obviated operating system. It doesn't matter what OS you run if you can create a single application which runs on mobile phones, from a web browser on all major OS's, or as a desktop application on all major OS's.
It's quick and easy to create a single application which runs just about anywhere - much simpler than creating a standard desktop application. So as a developer, as long as you don't need something high-end enough to rule out Flex as a platform (ie, 3d games, etc), there's really very little reason to not currently be developing for Flex.
Microsoft knows this, Flash crept up on them and turned into a serious threat to their monopoly. They're probably really kicking themselves for having distributed it for a while, cinching the install base.
Silverlight is Microsoft's knee-jerk reaction to the realization that this sleeping giant is waking up. It's their attempt to maintain the lock-in they currently own. This is why they're now shoving Silverlight down your throat. For example, my Microsoft Office 2008 on OS X had a "Critical Update," whose description was vague, and did not contain a link to the full details. Installing it because of course that's what you do with critical updates, it turns out to have simply been an install of Silverlight, even though there was no way for me to have known this in advance.
I went to a Silverlight developer conference, and I saw Microsoft employees showing off example applications, including walking us through the creation of these applications. I can say without a doubt that Flex is substantially easier to work with; in the time and lines of code they created a simple slideshow with fading transitions that reads filenames out of a CSV, at the Adobe conference, they'd made a slideshow with thumbnails, transitions, varying timers, pause, manual navigation, and even a carousel mode, which read data from a CSV, SOAP, WSDL, or REST web service.
Like many things, Microsoft is putting just enough effort into Silverlight to make it look competitive.
FWIW, I asked during the Silverlight developer conference a few months ago what the current install base of Silverlight was, and the only response they were willing to give is, "If you don't use it, nobody will install it." That means practically nobody.
They don't do this because it's a vector for non-Apple-approved applications to run on the iPhone. It's the same reason they refuse to allow Java to run on it. They want to control what people run on the phone so they can charge for services which free (speech/beer) software could enable for... well, for free (beer).
That isn't the defense being taken here. They're not saying, "We'd love to have that feature, but it's too hard to reliably test," they're essentially saying, "I like it better this way, so should everyone else."
Such an isolated feature as user-resizable vs auto-resizign text boxes would not introduce much testing complexity, and should not be the sort of thing which is likely to get broken by, or break other features very frequently unless it's specifically a feature dealing with how text boxes are resized. Besides, the GTK engine pretty much handles this for them.
This is an example of a trivial feature with significant usability and preference implications; the developers refusing to listen to their users is simple stubbornness.
Under this analogy, they already had a blue bikeshed, but built a new pink bikeshed, and instead of giving users an option which shed they want to store their bike in, they tore down the blue one.
MS's UI's are dictated by HCI staff. Developers write code, HCI staff interact with people. UI design and software development are extremely different skills.
Some developers may be decent at UI design, and some UI designers may know how to write code, but likewise some mechanics are good at playing piano, and some pianists may know how to tune up their engine. That doesn't mean they have much to do with each other.
Say what you will about Microsoft's UI (I'm not a big fan myself), but they make many of their UI decisions based on extensive and costly research which involves real people interacting with interfaces. When users say, "I would have found this task easier if X," they try that idea on other users, and if it doesn't introduce some new usability problem, it becomes a feature.
In this case, if this were Microsoft, users would be saying, "I would find it easier to IM if I could resize my text box," the UI designers would try this idea on some other folks, and as long as that didn't make their task harder, it would at least become an option.
It's GAIM's (sorry, Pidgin's) horrible interface which keeps me from being able to use it myself. It sounds as if they're making negative progress on this front, unfortunately.
The project really could use a fork, someone who will strip out the communications core, and put a completely redesigned front end on it. Frankly, Pidgin ought to be two closely related projects - one which provides IM libraries, and one which provides a front end to those libraries.
I agree, this is why I always remove gnome-screensaver and install xscreensaver.
Gnome's developers think you should like every screen saver to use only default settings. They even go on to suggest that these apps should be good enough to be able to run without any user-specified options like, say, what RSS feeds do you want to pull?
The university might assist since they'll know he's actually dead. The agencies which have no form of personal relationship with him though won't know this, and should treat all such inquiries as if they are social engineering attacks until lawyers become involved.
There was a deal a few years ago where a soldier who had been killed in Iraq had a Yahoo email account. Yahoo refused to give the family access, and I believe it was found that they didn't have any obligation to do so.
These uninvolved companies don't know anything about their user or the user's wishes. Just because I'm deceased doesn't mean I want my estranged wife having a chance to dig through all my emails (not that I'm either deceased or have an estranged wife). For example, there could be evidence in the email which permits her to challenge the will I'd written, etc.
I put florescent zip ties on my luggage to hold the zipper shut. If they want in, they can snip it easily enough, but I can tell that they've been in there (unless they're keeping a supply of florescent zip ties around to replace them with). This way my luggage isn't snoopable by regular handlers, and won't by happenstance come unzipped in transit.
I use florescent because there's less chance they have that just sitting around to replace mine with. It also aids in locating your luggage when you go to pick it up at the baggage claim; unless you want some ugly luggage, chances are there's a few other people on your flight with fairly similar bags.
With a PGP encrypted message, you have communications whose capture is almost guaranteed, and whose access may be compromised at a much later date if the client uses a compromised terminal to read one of these messages months later.
With a physical meeting, you can sweep for bugs. You can know that unless the communications are captured in real time, they cannot be intercepted later. You can bring eavesdropping defeating devices (eg, a recording of a crowded room played loudly while you and your client converse quietly). You can require that some form of conspiracy must be made to eavesdrop on each occasion (installing listening devices requires physical access and knowledge of the meeting room in advance).
With a physical meeting, there really are fewer points of attack than with electronic communications.
Any other charitable organization is fully capable of picking up where the OLPC left off should the OLPC organization instantly vanish tomorrow and leave no support infrastructure. Individual governments would able to hire some programmers to pick up the pieces and continue on. Individual people would be able to extend the core and continue to introduce new features.
If Microsoft goes bankrupt, or even simply loses interest in this project, nobody has the background resources necessary to support it unless Microsoft completely open sourced it first.
That's funny, it doesn't make me feel very good at all.
Er... wait, maybe someone is reading this. I LOVE THIS LIST, IT PROTEXTS US FROM THA BADGUYZ.
Whew, almost found myself blacklisted from the world! Which is my own funny way of saying the government agent currently reading this is quite attractive and intelligent in every way, and should have no reason to blacklist me.
It also provides yet another black list which a person can get on without any sort of due process if they annoy the wrong bureaucrat.
These lists are a form of punishment without trial. Someone can end up unable to fly, do business, or get a job simply by having their name appear on a list - that they weren't aware they were on, may have done no wrong, and have absolutely no way of challenging or even confronting their accusers.
Yes, you can have KDE and Gnome both installed at the same time. sudo apt-get install kde-desktop. This'll change the bootup logo to be the Kubuntu version, but on the login screen you'll be able to choose your session, and choose KDE or Gnome depending on what you feel like getting into.
I go back and forth every few months personally. I like the simplicity of Gnome, they make the features I want most easily available to me.
But then I'll find some corner case which I simply can't do with Gnome (eg Gnome mis-detecting a file type and refusing to let you open it because the file contents don't match the extension), then I fire up KDE for a while, revel in the many, many options to tweak, dork around with those for a while before discovering this stuff actually gets in the way of my productivity when I'm not just tooling around, and end up going back to Gnome again.
No matter what desktop I'm using, I end up using programs from the alternate environment. For example, I always use Kate as my text editor, Amarok as my media library, and Kopete as my IM client (all KDE apps).
It's a good point. Even the most thoroughly inspected terminal, if it's in a public location, may still capture your details.
For the most die-hard paranoid, one-time passwords are the only real security you can offer against capture (such that even if they're captured, they're useless), presuming you have a way to look up the one-time password without exposing subsequent passwords (ie, you can't just have them written all on the same sheet of paper, or the pinhole camera could capture the next ones).
Even RSA SecureID is vulnerable if its information is captured by a system able to exploit it in real time, unless the RSA system only permits one successful login per account per minute (the interval it takes the SecureID to refresh its numbers). If the information is precaptured (such as by a camera before you've typed the digits), they could beat you to the authentication. The public terminal could be written to capture your authentication credentials, changed your entered credentials to false ones (to make you think you'd typed it wrong), and perform the real authentication in the background to gain access.
So it comes down to: there's theoretically no way to completely secure yourself when using a public terminal or when using even a trusted terminal in a public space if you are hyper paranoid.
Slashdot Mirror
It's not artificially-inspired confidence. I have to brake suddenly less often, I glance around me and am unsure if I've accidentally missed my exit more often, I accidentally switch lanes with another car in my blind spot less often, I notice deer on the side of the road before they charge in front of me (especially in some seasons there are as many as 2-3 new deer roadkill along my drive each day). I'm more aware of the road in ways which are demonstrable and not a product of my imagination as you suggest it could only be.
On long drives, an audiobook keeps my brain aware, I don't get as glazed over, I don't find myself fighting to stay awake, and I don't just drive on autopilot. An audiobook doesn't require nearly as much attention as personal conversation, and it never requires you to take your eyes off the road such as in your example.
I'll agree that if you're on an unfamiliar course, you could put those brain cycles to better use - if I had to do something which required a higher level of brain faculty, such as drive through a detour, then I'd pause the book. But when you're doing the same path you've done literally hundreds of times before, and you're at it for three hours each day, your attention wanders and it increases your risks. When your brain doesn't have anything to focus on, it shuts down and you enter a bit of a stupor which requires a short but possibly fatal amount of time to recover from.
No, what I mean is that I'd be startled by the car in front of me braking less frequently, and not need to slam on my brakes, or realizing I was drifting over the center line, or suddenly looking around me and not being certain if I had missed my exit because I didn't immediately recognize my surroundings.
Things which I couldn't have helped noticing before because they would have made themselves known to me eventually if I had missed them.
I have to disagree with at least one point for serious reasons.
I drove 3 hours a day for 4 years. About 6 months into this I started listening to books on tape, and I found my alertness level while driving was improved significantly. When I was just listening to the radio or my ipod, and it was the same stuff I've heard a thousand times before, my mind drifted. When I started keeping my mind awake and aware with audiobooks, I found I was surprised by traffic around me much less often.
I touted this to several coworkers who also had long drives, and collectively we all agreed: audiobooks keep your mind more active, and increase your overall awareness of arising traffic situations, we found ourselves in fewer close calls and surprised by things around us less often.
Because of big business. Big business demands to run fully-patched software, but also demands that software not be upgraded before every application the business users use has been tested with it.
Most big business upgrade cycle is 2-5 years behind cutting edge. We got XP last year.
I keep IE 6 running in my VirtualBox instance of XP. If a site works in IE 6, it almost certainly works in IE 7. The converse is not true, and since I have testing to do, that's a safer platform to test on.
I bought Bioshock, it installed securom, and I couldn't play it because I had too many developer hooks in my system. I fought with support who essentially said the only recourse was to reinstall Windows.
So I got a pirated copy, and all was well.
Game copy restrictions don't do anything to stop pirates, they delay the cracked version by a matter of a week, maybe days, perhaps even only hours. They only punish legitimate customers.
People who have money in their fist and want to give it to the game company are being negatively conditioned away from doing so. People realize that crackers do a better job of providing a seamless and error free customer interface than the company who takes their money for failing to provide this service.
When you decide one day that you want to play a certain game, now the decision comes down to: doing it legitimately with a reasonable chance that it's also unsuccessful, and you're out the cost of the game (since it's now open, the store won't take it back), or doing it illegitimately with an almost certain chance that it's successful, and even if it's not, at least you're not out any money.
But they have to prove that you didn't buy that music. An affirmative defense is that your music CD's were stolen. They would have a very, very hard time busting you for music you merely have a copy of, the burden of proof is too high.
Even if they could show that for example the media had "Downloaded from MuzicWarez - Your place for pirated music," in its metadata, they'd have a hard time convincing a jury to convict you if your defense was that you downloaded that track after you lost the original album.
However, distribution is something they can bust you on, because unless you're an online music retailer, you don't have any license to be distributing music electronically.
Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway. - Tanenbaum, Andrew S. (1996).
No way, they know exactly why they made it, and they do in fact need it very badly.
Flex + Flash's ubiquity + Adobe Air = obviated operating system. It doesn't matter what OS you run if you can create a single application which runs on mobile phones, from a web browser on all major OS's, or as a desktop application on all major OS's.
It's quick and easy to create a single application which runs just about anywhere - much simpler than creating a standard desktop application. So as a developer, as long as you don't need something high-end enough to rule out Flex as a platform (ie, 3d games, etc), there's really very little reason to not currently be developing for Flex.
Microsoft knows this, Flash crept up on them and turned into a serious threat to their monopoly. They're probably really kicking themselves for having distributed it for a while, cinching the install base.
Silverlight is Microsoft's knee-jerk reaction to the realization that this sleeping giant is waking up. It's their attempt to maintain the lock-in they currently own. This is why they're now shoving Silverlight down your throat. For example, my Microsoft Office 2008 on OS X had a "Critical Update," whose description was vague, and did not contain a link to the full details. Installing it because of course that's what you do with critical updates, it turns out to have simply been an install of Silverlight, even though there was no way for me to have known this in advance.
I went to a Silverlight developer conference, and I saw Microsoft employees showing off example applications, including walking us through the creation of these applications. I can say without a doubt that Flex is substantially easier to work with; in the time and lines of code they created a simple slideshow with fading transitions that reads filenames out of a CSV, at the Adobe conference, they'd made a slideshow with thumbnails, transitions, varying timers, pause, manual navigation, and even a carousel mode, which read data from a CSV, SOAP, WSDL, or REST web service.
Like many things, Microsoft is putting just enough effort into Silverlight to make it look competitive.
FWIW, I asked during the Silverlight developer conference a few months ago what the current install base of Silverlight was, and the only response they were willing to give is, "If you don't use it, nobody will install it." That means practically nobody.
They don't do this because it's a vector for non-Apple-approved applications to run on the iPhone. It's the same reason they refuse to allow Java to run on it. They want to control what people run on the phone so they can charge for services which free (speech/beer) software could enable for... well, for free (beer).
That isn't the defense being taken here. They're not saying, "We'd love to have that feature, but it's too hard to reliably test," they're essentially saying, "I like it better this way, so should everyone else."
Such an isolated feature as user-resizable vs auto-resizign text boxes would not introduce much testing complexity, and should not be the sort of thing which is likely to get broken by, or break other features very frequently unless it's specifically a feature dealing with how text boxes are resized. Besides, the GTK engine pretty much handles this for them.
This is an example of a trivial feature with significant usability and preference implications; the developers refusing to listen to their users is simple stubbornness.
Under this analogy, they already had a blue bikeshed, but built a new pink bikeshed, and instead of giving users an option which shed they want to store their bike in, they tore down the blue one.
"Internet Elephant" - I thought we were supposed to stay away from things abbreviated IE.
An internet is a type of thing. The Internet is a specific one of these internets. It's like god vs God.
We have a company intranet. An intranet is a type of internet, but it is not the Internet.
Internet2 is an internet, but it is not the Internet, it is the Internet2.
Enough examples? =)
MS's UI's are dictated by HCI staff. Developers write code, HCI staff interact with people. UI design and software development are extremely different skills.
Some developers may be decent at UI design, and some UI designers may know how to write code, but likewise some mechanics are good at playing piano, and some pianists may know how to tune up their engine. That doesn't mean they have much to do with each other.
Say what you will about Microsoft's UI (I'm not a big fan myself), but they make many of their UI decisions based on extensive and costly research which involves real people interacting with interfaces. When users say, "I would have found this task easier if X," they try that idea on other users, and if it doesn't introduce some new usability problem, it becomes a feature.
In this case, if this were Microsoft, users would be saying, "I would find it easier to IM if I could resize my text box," the UI designers would try this idea on some other folks, and as long as that didn't make their task harder, it would at least become an option.
It's GAIM's (sorry, Pidgin's) horrible interface which keeps me from being able to use it myself. It sounds as if they're making negative progress on this front, unfortunately.
The project really could use a fork, someone who will strip out the communications core, and put a completely redesigned front end on it. Frankly, Pidgin ought to be two closely related projects - one which provides IM libraries, and one which provides a front end to those libraries.
I agree, this is why I always remove gnome-screensaver and install xscreensaver.
Gnome's developers think you should like every screen saver to use only default settings. They even go on to suggest that these apps should be good enough to be able to run without any user-specified options like, say, what RSS feeds do you want to pull?
The university might assist since they'll know he's actually dead. The agencies which have no form of personal relationship with him though won't know this, and should treat all such inquiries as if they are social engineering attacks until lawyers become involved.
There was a deal a few years ago where a soldier who had been killed in Iraq had a Yahoo email account. Yahoo refused to give the family access, and I believe it was found that they didn't have any obligation to do so.
These uninvolved companies don't know anything about their user or the user's wishes. Just because I'm deceased doesn't mean I want my estranged wife having a chance to dig through all my emails (not that I'm either deceased or have an estranged wife). For example, there could be evidence in the email which permits her to challenge the will I'd written, etc.
I put florescent zip ties on my luggage to hold the zipper shut. If they want in, they can snip it easily enough, but I can tell that they've been in there (unless they're keeping a supply of florescent zip ties around to replace them with). This way my luggage isn't snoopable by regular handlers, and won't by happenstance come unzipped in transit.
I use florescent because there's less chance they have that just sitting around to replace mine with. It also aids in locating your luggage when you go to pick it up at the baggage claim; unless you want some ugly luggage, chances are there's a few other people on your flight with fairly similar bags.
With a PGP encrypted message, you have communications whose capture is almost guaranteed, and whose access may be compromised at a much later date if the client uses a compromised terminal to read one of these messages months later.
With a physical meeting, you can sweep for bugs. You can know that unless the communications are captured in real time, they cannot be intercepted later. You can bring eavesdropping defeating devices (eg, a recording of a crowded room played loudly while you and your client converse quietly). You can require that some form of conspiracy must be made to eavesdrop on each occasion (installing listening devices requires physical access and knowledge of the meeting room in advance).
With a physical meeting, there really are fewer points of attack than with electronic communications.
Any other charitable organization is fully capable of picking up where the OLPC left off should the OLPC organization instantly vanish tomorrow and leave no support infrastructure. Individual governments would able to hire some programmers to pick up the pieces and continue on. Individual people would be able to extend the core and continue to introduce new features.
If Microsoft goes bankrupt, or even simply loses interest in this project, nobody has the background resources necessary to support it unless Microsoft completely open sourced it first.
That's funny, it doesn't make me feel very good at all.
Er... wait, maybe someone is reading this. I LOVE THIS LIST, IT PROTEXTS US FROM THA BADGUYZ.
Whew, almost found myself blacklisted from the world! Which is my own funny way of saying the government agent currently reading this is quite attractive and intelligent in every way, and should have no reason to blacklist me.
It also provides yet another black list which a person can get on without any sort of due process if they annoy the wrong bureaucrat.
These lists are a form of punishment without trial. Someone can end up unable to fly, do business, or get a job simply by having their name appear on a list - that they weren't aware they were on, may have done no wrong, and have absolutely no way of challenging or even confronting their accusers.
Welcome to the Bushcarthy era.
Yes, you can have KDE and Gnome both installed at the same time. sudo apt-get install kde-desktop. This'll change the bootup logo to be the Kubuntu version, but on the login screen you'll be able to choose your session, and choose KDE or Gnome depending on what you feel like getting into.
I go back and forth every few months personally. I like the simplicity of Gnome, they make the features I want most easily available to me.
But then I'll find some corner case which I simply can't do with Gnome (eg Gnome mis-detecting a file type and refusing to let you open it because the file contents don't match the extension), then I fire up KDE for a while, revel in the many, many options to tweak, dork around with those for a while before discovering this stuff actually gets in the way of my productivity when I'm not just tooling around, and end up going back to Gnome again.
No matter what desktop I'm using, I end up using programs from the alternate environment. For example, I always use Kate as my text editor, Amarok as my media library, and Kopete as my IM client (all KDE apps).
It's a good point. Even the most thoroughly inspected terminal, if it's in a public location, may still capture your details.
For the most die-hard paranoid, one-time passwords are the only real security you can offer against capture (such that even if they're captured, they're useless), presuming you have a way to look up the one-time password without exposing subsequent passwords (ie, you can't just have them written all on the same sheet of paper, or the pinhole camera could capture the next ones).
Even RSA SecureID is vulnerable if its information is captured by a system able to exploit it in real time, unless the RSA system only permits one successful login per account per minute (the interval it takes the SecureID to refresh its numbers). If the information is precaptured (such as by a camera before you've typed the digits), they could beat you to the authentication. The public terminal could be written to capture your authentication credentials, changed your entered credentials to false ones (to make you think you'd typed it wrong), and perform the real authentication in the background to gain access.
So it comes down to: there's theoretically no way to completely secure yourself when using a public terminal or when using even a trusted terminal in a public space if you are hyper paranoid.