I've heard numerous times that for the same power output, a nuclear reactor generates less radioactive material than, say, a coal fired plant. The problem is that the nuclear waste is in a big chunk, and must be stored somewhere. My question is, why not pulverize said nuclear waste and pump it into the atmosphere? At worst, we'd be doing slightly better than coal plants right? And we'd have solved the waste storage problem... right? I'm sure there's something I'm missing (other than the obvious: that's just insidiously stupid).
Encrypted file system I guess... then they'd at least have to subpoena the key from the data owner, even if they can get the hardware through other avenues. Of course, this may be cumbersome on a remotely hosted website.
I've long been hearing about alternate voting methods (e.g. instant-runoff) and could not figure out why any rational being would not consider such a system surperior to the current system in the US. Especially given the 2000 presidential election, where many felt that Gore would have won had Nader not ran and "stolen" votes from Gore. If that were the case, then cleary, Gore should have been president and a voting system that could measure that should be instated.
I was watching a poli-sci professor on C-SPAN last night talk about direct election (favored by him) vs. our electoral system. At the end, he stated that one common concern was that having direct election would require a runnoff vote. He dismissed this concern (and I don't understand why the question even came up...), but he spoke of a runnoff vote as if it were the spawn of satan and would "destabilze the two party system". This is the first time I've heard anyone give a real reason in defence of the current voting system, and I was quite shocked to be honest. Apparently many people believe a two party system (as enforced by the vote scoring method) is superior to any other method (which to me would seem to more accurately measure the desires of the public and hence be better).
So, I leave confused and wondering what exactly is so great about a two party system. Can anyone point me to some discussion of this?
Well, technically perhaps but not really in spirit. You could certainly say I'm biased toward a certain view of how markup language *should* be used... But take a look at the slashdot html: Let's see, the Slashdot logo at the top, maybe call it a toplevel heading? Nope, it's table data. The headline above each article, perhaps we should call it a second level heading using? Nope, table data. The paragraph summary of each article? Table data.
It may be technically correct, but lying to my browser about what is contained within those tags annoys me somewhat. In my opinion, and in the opinion of many proponents of [X]HTML+CSS, Slashdot (like most other websites) abuses the markup language, defeating the entire purpose. For some websites, they might as well serve up pdfs.
There are also significant bandwith savings to be had as show by A List Apart which has been posted numerous times before.
I'm by no means an expert, but I was under the impression that a cluster of yesterday's computers would easily be outperformed by a single top of the line computer. So, except for fun and learning, clustering with old computers is just a waste of time.
Absolutely. But see, there's this little problem: education takes time and effort. If I can't decide who to vote for based on sound bites from TV, then hell, I'll vote for Kerry because Bush looks like an ape. Yes, I am the problem with America today. I watch TV and never hear mention of any concepts covered in, say entry level economics or history courses. I hear tax cuts this, free trade that, but have no concept of the long or short term effects of these policies. I don't know what my senator has done; I don't know the name of my current House representative.
Everything else follows quickly and easily from that.
Well, I'm not sure I agree with that, but since telling free software developers what to work on (aside from being wrong) is like herding cats, I guess we're stuck waiting to find out if you're right.
I just don't see the whole interface inconsistency being solved. The more I learn about interface design, the more I realize how different my mindset was before I decided usability is the most important feature of any software program. In free software, I've mostly seen a lone coder or three working on some technically interesting project only to neglect the interface, or use some oddball toolkit, or whatever*. And no one wants to be told how they should code their project. Though, there are steps being made to improve this situation (Gnome usability guidlines and freedesktop.org come to mind) which are, of course, due to developers doing what makes them happy. So perhaps you are right after all.
* Why is it that in shell, there is only one interface (stdin, stdout, commandline args), but in "GUI" there are a multitude of interfaces. Would it not be a fascinating and purely technical challenge to create a gui system that didn't insist upon or enforce a single interface (in the way gnome insists on gtk) but merely was this way by design? From a purely technical standpoint, this would reduce code redundancy (how many GUI toolkits do we have? how many applications have coded a "file->save" function? in which text widgets can I use bold? why does app A have "undo" while app B does not?), and it would be a huge win for usability.
...doesn't have to pay me to swing by every month to run a spyware/virus scan.
Of course, as some would tell it, if and when significant migration to Linux occurs, so will viruses and other malware migrate to Linux, though this doesn't diminish the other reasons for migration.
Let's assume that you need at least one eye to successfully poke the eye of another. After poking a single eye, the eye-poker is now one-eyed, having given an eye for justice, but can still poke. The eye-poker's eye won't be removed until he is caught. This is the key point, and it allows the one-eyed (or two-eyed) eye-poker to poke as many eyes as she can until someone stops her. So it is possible for a clever eye poker to poke many more eyes than her two eye quota that she would at first appear to be limited to. (Let's ignore for a moment the question of punishing a multi-poker who has but two eyes to lose). Thus, it is certainly possible for everyone to go blind in an eye for an eye world if we allow for an eye-poking rampage.
Or maybe you just shouldn't take everything so literally.
Re: your last paragraph. Usability is a hard sell. I remember an Apple commercial ("There's no step three") from a few years ago and it's really the only advertisement I really remember that mentioned usability. (Ok, I guess AOL is "so easy to use, no wonder it's number one"... And infomercials often discuss ease of use, but none of these ever mentions anything really concrete).
Anyway, when was the last time you shopped for a new stove / oven (yes, I know stove shopping isn't a weekly thing) and based your decision on whether or not the four stove control knobs were layed out in a square,thus having a direct mapping to the actual stove, or layed out in a single row, thus requiring some sort of symbol or instruction to map a knob to a stove burner (my grandmother started a fire because she turned the wrong burner on once). My guess is never, but this is a usability issue and, in my experience, a safety issue as well.
people have adapted to this new method of navigation and are quite comfortable with it.
People? Not my people. (or, not my parents.) They have been using the Web for about three or four years now, though on a highly irregular basis, and they find it extremely confusing at best. Most of the time they have trouble remembering how to get "here" or "there" (they don't quite get bookmarks, and bookmarks don't always work on highly dynamic websites anyway). I'd say much of this is due to poorly designed web sites with no clear cues hinting at the overall layout of the site, or worse, many conflicting cues (links across the top in several rows, side bar at the left, links in the body, etc.) combined with much glitz and glitter.
Jef is seeking to create an interface that works the way his brain works.
I must respectfully disagree here. I have not read the article, however I suspect you have not read Raskin's book The Humane Interface which makes it clear that he uses the limits of the [majority of] human minds to direct his designs. For example, the human short term memory makes it impossible to concentrate on two things at once. A human can do many things at once, but all but one of those things must be automatic (i.e performed without thinking). I can't really prove this, and some people surely can switch between concentrations more quickly than others, but imagine trying to solve a calculus problem while at the same time looking for an error in your checkbook...
An interface with a mouse and menus typically requires fairly precise mousing maneuvers to activate a menu item. It is impossible to [completely] automate these maneuvers, due to variations in the mouse cursor starting position, and the position of the window containing the menu (or button, etc.). So, when typing a document, any use of a menu item requires a context switch of the mind, interrupting the typing (which may not be a huge deal depending on what you are doing exactly.. it may have required a context switch anyway). This serves to create a cumbersome interaction experience (another unproven conjecture, but one I doubt most would disagree with).
Notice how many "experts", or those who need to use an interface for extensive periods, take the time to learn "keyboard shortcuts", which have the advantage that they may be automated once learned (unlike many mousing maneuvers)*. This is all fine and good, BUT, when beginning to learn an interface, one must make the choice to either use the easy to discover menus or the efficient keyboard shortcuts (which are not efficient until they become automatic). I'm not sure it is possible, but I believe it is a *very* worthy endevor to create an interface that combines the ease of learnability or occasional use of a graphical menu system with the ability of keyboard commands to be automated.
I'm sorry this turned out to be quite long... and perhaps not a strict response to the parent post. Reading the posts here I get the feeling that there is rampant misunderstanding of Raskin's goals, possibly due to Raskin's personality or whatever.
* I am aware of Bruce Toggnazini's tests that show that those using keyboard shortcuts believe they are faster than when using the mouse only to discover the stopwatch tells a different story. This only applies to certain cases, and these tests were performed on programs designed for mouse use with keyboard shortcuts added afterwards. You know as well as I do that "rm *.jpg" is faster than selecting all jpgs with the mouse and dragging to the trash (for an example of an interface designed for keyboard use vs. one designed for mouse use).
...but you've already got to scan through the headers to find out who it's really "from" anyway. Seems like From: and Reply-To: is a lost cause. And from a usability standpoint it just seems like a confusing disaster. "Wait, who is this from?" "I hit reply, but it wants to send it somewhere else?? why? oh yeah, Reply-To:."
And the MUA needs to display some sort of sender information. I'd rather have all messages from Joe show as joe@foo.com rather than joe@whatever.server.i.sent.this.from. which may require the display of an additional from address (the Reply-To:, which isn't a from address at all).
I suppose more intelligent minds than me have worked hard at this whole email mess... but there has to be some way to verify or trust the sender without imposing rules on the From: header.
One more ding against the nipple/joystick: You move the stick down to scroll down, and proceed to hold it still while the scrolling continues. With the clickwheel, your thumb is in motion with the scrolling motion, and held still to hold the scrolling still.
I haven't used an iPod for more than a few seconds, but I think "non-intuitive" is a rather pointless statement. Sure, you might not get it immediately, but once someone shows you, it's a piece of cake; you probably won't forget it. Remember all the jokes about people trying to use the computer mouse as a foot pedal, or putting it on the screen? The mouse is simple enough, but only after you know. Nothing is really "intuitive". The real goal is to be easy to learn and use, and difficult to forget once learned, which I think the clickwheel achieves.
Also, the scrolling speed of the clickwheel is directly proportional to the speed with which you move your thumb. The scrollwheel on a mouse also achieves this, but I find it more difficult with the rubber nipple style scroll stick. Finally, to continue scrolling, you just keep moving your thumb in the same circular direction. The rubber nipple also requires no change to continue scrolling, but a mouse's scrollwheel does (you must scroll a bit, then move your finger to the top of the wheel and scroll some more). This combination of the scroll speed being directly related to your own thumb speed and the continuous motion for continuous scrolling is why I believe the clickwheel a is a good interface.
One point against it may be that the scrolling operation is up and down, while the thumb motion is in circles. I haven't put enough thought in to imagine a solution for that...
Re:Fixing fundamental design mistakes?
on
Linus Interviewed
·
· Score: 1
Just a small wish about file extensions: Can we simplify things and have.image,.video,.audio,.doc[ument] or something along those lines? Most of the time I don't really care what compression algorithm this image was compressed with. If I do, I can always view more detailed information about a file (e.g. with the "file" command on Linux, or possibly a right click->properties, or even show the info in one of the file manager views, whatever).
Though, I suppose this may be cumbersome on today's systems because some apps only handle e.g..bmp, while others handle many image formats. To me this is a design flaw, but so fundamentally ingrained into systems with "applications" as to be a lost cause at the moment (I dream of a system where I only need to worry about, say, pdf support once. The system supports pdf. Done. None of this "oh, this program opens pdfs; that one can edit them; and this third one can save its documents as pdfs; but this other app can only save as postscript, so you need to use this converter...").
I'm trailing on into a rant at this point, but filenames (which shouldn't really be necessary in the first place, but assuming we're using a system that demands them) are the domain of the user, and should not be restricted to certain characters, or "must end with.extension" or any of that sillinies. Are you telling me that my 21st century computer can't keep track of what type of file it has? Oh, you are? In the name of backwards compatibility? Probably a necessary evil, but sheesh I hate it so.
As the other ceramic engineer pointed out, glass is considered a ceramic. What makes it good for lenses is the fact that it is amorphous, i.e. has no periodic crystal structure. Thus, light will pass through it without being refracted in strange ways because on a macroscopic level, all the glass is the "same" (or the heterogeneity is small enough and distributed evenly and randomly).
Now, I didn't read the article, though I wouldn't expect it to get into details. I assume by "ceramic" they mean "crystalline ceramic" (versus amorphous). When you make crystalline ceramics, you don't typically make one continuous crystal (like a polished gemstone), but many small (10 to 500 micron is typical) crystals (grains) surrounded by very thin amorphous regions (grain boundaries). Grain boundaries serve to refract the light (I suppose they are much larger and thus less heterogenous than anomalies in the amorphous glass), which is why ceramics (think dinner plate) are opaque. I'm getting to the limits of my [limited] knowledge, but I believe the only real way to make transparent crystalline ceramics is to use a single crystal which contains no grain boundaries. This tends to be costly due to the need to melt the material to coax it into a single crystal and the high melting temperatures of ceramics (see silicon industry).
Given that, I'm not sure why using a single crystal lens would have an advantage over an amorphous glass lens... Perhaps you can achieve a higher index of refraction.
Let's face it, our options suck. Nader? Ya right. That's a vote for Bush, not Nader.
The sooner you realize that the single vote under your direct control has essentially zero effect on the election as a whole (what would the outcome of the last election be if you had voted differently or not at all?), the sooner we can begin to repair the 2-party system that "sucks".
Note that in Florida 2000, the closest thing to an election coming down to a single vote that I've ever heard of, we proved that our voting system does not even have sufficient resolution to determine the outcome in the (next to impossible) event that the election does indeed come down to a single vote.
Can someone please make a car with a stereo that has an audio input? Does such a thing exist even in aftermarket? Assuming I already have a portable music player, I could just plug it in to my car stereo. Instead, I'm stuck using some pathetic mini FM transmitter, or cassette adapter if I have a cassette deck, or buying a whole new HD car stereo.
Slashdot Mirror
I've heard numerous times that for the same power output, a nuclear reactor generates less radioactive material than, say, a coal fired plant. The problem is that the nuclear waste is in a big chunk, and must be stored somewhere. My question is, why not pulverize said nuclear waste and pump it into the atmosphere? At worst, we'd be doing slightly better than coal plants right? And we'd have solved the waste storage problem... right? I'm sure there's something I'm missing (other than the obvious: that's just insidiously stupid).
Encrypted file system I guess... then they'd at least have to subpoena the key from the data owner, even if they can get the hardware through other avenues. Of course, this may be cumbersome on a remotely hosted website.
Let the Dick Cheney jokes begin...
I've long been hearing about alternate voting methods (e.g. instant-runoff) and could not figure out why any rational being would not consider such a system surperior to the current system in the US. Especially given the 2000 presidential election, where many felt that Gore would have won had Nader not ran and "stolen" votes from Gore. If that were the case, then cleary, Gore should have been president and a voting system that could measure that should be instated.
I was watching a poli-sci professor on C-SPAN last night talk about direct election (favored by him) vs. our electoral system. At the end, he stated that one common concern was that having direct election would require a runnoff vote. He dismissed this concern (and I don't understand why the question even came up...), but he spoke of a runnoff vote as if it were the spawn of satan and would "destabilze the two party system". This is the first time I've heard anyone give a real reason in defence of the current voting system, and I was quite shocked to be honest. Apparently many people believe a two party system (as enforced by the vote scoring method) is superior to any other method (which to me would seem to more accurately measure the desires of the public and hence be better).
So, I leave confused and wondering what exactly is so great about a two party system. Can anyone point me to some discussion of this?
Well, technically perhaps but not really in spirit. You could certainly say I'm biased toward a certain view of how markup language *should* be used... But take a look at the slashdot html: Let's see, the Slashdot logo at the top, maybe call it a toplevel heading? Nope, it's table data. The headline above each article, perhaps we should call it a second level heading using? Nope, table data. The paragraph summary of each article? Table data.
It may be technically correct, but lying to my browser about what is contained within those tags annoys me somewhat. In my opinion, and in the opinion of many proponents of [X]HTML+CSS, Slashdot (like most other websites) abuses the markup language, defeating the entire purpose. For some websites, they might as well serve up pdfs.
There are also significant bandwith savings to be had as show by A List Apart which has been posted numerous times before.
I'm by no means an expert, but I was under the impression that a cluster of yesterday's computers would easily be outperformed by a single top of the line computer. So, except for fun and learning, clustering with old computers is just a waste of time.
Absolutely. But see, there's this little problem: education takes time and effort. If I can't decide who to vote for based on sound bites from TV, then hell, I'll vote for Kerry because Bush looks like an ape. Yes, I am the problem with America today. I watch TV and never hear mention of any concepts covered in, say entry level economics or history courses. I hear tax cuts this, free trade that, but have no concept of the long or short term effects of these policies. I don't know what my senator has done; I don't know the name of my current House representative.
Ignorance is bliss, and bliss is god.
touché
Everything else follows quickly and easily from that.
Well, I'm not sure I agree with that, but since telling free software developers what to work on (aside from being wrong) is like herding cats, I guess we're stuck waiting to find out if you're right.
I just don't see the whole interface inconsistency being solved. The more I learn about interface design, the more I realize how different my mindset was before I decided usability is the most important feature of any software program. In free software, I've mostly seen a lone coder or three working on some technically interesting project only to neglect the interface, or use some oddball toolkit, or whatever*. And no one wants to be told how they should code their project. Though, there are steps being made to improve this situation (Gnome usability guidlines and freedesktop.org come to mind) which are, of course, due to developers doing what makes them happy. So perhaps you are right after all.
* Why is it that in shell, there is only one interface (stdin, stdout, commandline args), but in "GUI" there are a multitude of interfaces. Would it not be a fascinating and purely technical challenge to create a gui system that didn't insist upon or enforce a single interface (in the way gnome insists on gtk) but merely was this way by design? From a purely technical standpoint, this would reduce code redundancy (how many GUI toolkits do we have? how many applications have coded a "file->save" function? in which text widgets can I use bold? why does app A have "undo" while app B does not?), and it would be a huge win for usability.
Of course, as some would tell it, if and when significant migration to Linux occurs, so will viruses and other malware migrate to Linux, though this doesn't diminish the other reasons for migration.
Let's assume that you need at least one eye to successfully poke the eye of another. After poking a single eye, the eye-poker is now one-eyed, having given an eye for justice, but can still poke. The eye-poker's eye won't be removed until he is caught. This is the key point, and it allows the one-eyed (or two-eyed) eye-poker to poke as many eyes as she can until someone stops her. So it is possible for a clever eye poker to poke many more eyes than her two eye quota that she would at first appear to be limited to. (Let's ignore for a moment the question of punishing a multi-poker who has but two eyes to lose). Thus, it is certainly possible for everyone to go blind in an eye for an eye world if we allow for an eye-poking rampage.
Or maybe you just shouldn't take everything so literally.
Trademark in the title, copyright in the summary, but a patent on the Rubik's cube. These are all different you know...
Re: your last paragraph. Usability is a hard sell. I remember an Apple commercial ("There's no step three") from a few years ago and it's really the only advertisement I really remember that mentioned usability. (Ok, I guess AOL is "so easy to use, no wonder it's number one"... And infomercials often discuss ease of use, but none of these ever mentions anything really concrete).
Anyway, when was the last time you shopped for a new stove / oven (yes, I know stove shopping isn't a weekly thing) and based your decision on whether or not the four stove control knobs were layed out in a square,thus having a direct mapping to the actual stove, or layed out in a single row, thus requiring some sort of symbol or instruction to map a knob to a stove burner (my grandmother started a fire because she turned the wrong burner on once). My guess is never, but this is a usability issue and, in my experience, a safety issue as well.
people have adapted to this new method of navigation and are quite comfortable with it.
People? Not my people. (or, not my parents.) They have been using the Web for about three or four years now, though on a highly irregular basis, and they find it extremely confusing at best. Most of the time they have trouble remembering how to get "here" or "there" (they don't quite get bookmarks, and bookmarks don't always work on highly dynamic websites anyway). I'd say much of this is due to poorly designed web sites with no clear cues hinting at the overall layout of the site, or worse, many conflicting cues (links across the top in several rows, side bar at the left, links in the body, etc.) combined with much glitz and glitter.
Jef is seeking to create an interface that works the way his brain works.
I must respectfully disagree here. I have not read the article, however I suspect you have not read Raskin's book The Humane Interface which makes it clear that he uses the limits of the [majority of] human minds to direct his designs. For example, the human short term memory makes it impossible to concentrate on two things at once. A human can do many things at once, but all but one of those things must be automatic (i.e performed without thinking). I can't really prove this, and some people surely can switch between concentrations more quickly than others, but imagine trying to solve a calculus problem while at the same time looking for an error in your checkbook...
An interface with a mouse and menus typically requires fairly precise mousing maneuvers to activate a menu item. It is impossible to [completely] automate these maneuvers, due to variations in the mouse cursor starting position, and the position of the window containing the menu (or button, etc.). So, when typing a document, any use of a menu item requires a context switch of the mind, interrupting the typing (which may not be a huge deal depending on what you are doing exactly.. it may have required a context switch anyway). This serves to create a cumbersome interaction experience (another unproven conjecture, but one I doubt most would disagree with).
Notice how many "experts", or those who need to use an interface for extensive periods, take the time to learn "keyboard shortcuts", which have the advantage that they may be automated once learned (unlike many mousing maneuvers)*. This is all fine and good, BUT, when beginning to learn an interface, one must make the choice to either use the easy to discover menus or the efficient keyboard shortcuts (which are not efficient until they become automatic). I'm not sure it is possible, but I believe it is a *very* worthy endevor to create an interface that combines the ease of learnability or occasional use of a graphical menu system with the ability of keyboard commands to be automated.
I'm sorry this turned out to be quite long... and perhaps not a strict response to the parent post. Reading the posts here I get the feeling that there is rampant misunderstanding of Raskin's goals, possibly due to Raskin's personality or whatever.
* I am aware of Bruce Toggnazini's tests that show that those using keyboard shortcuts believe they are faster than when using the mouse only to discover the stopwatch tells a different story. This only applies to certain cases, and these tests were performed on programs designed for mouse use with keyboard shortcuts added afterwards. You know as well as I do that "rm *.jpg" is faster than selecting all jpgs with the mouse and dragging to the trash (for an example of an interface designed for keyboard use vs. one designed for mouse use).
leaving the mice running free is not an option
How 'bout lettin' em run, just not free. Hook the little guys up to a generator or something :-)
Or how about a few cats...
...but you've already got to scan through the headers to find out who it's really "from" anyway. Seems like From: and Reply-To: is a lost cause. And from a usability standpoint it just seems like a confusing disaster. "Wait, who is this from?" "I hit reply, but it wants to send it somewhere else?? why? oh yeah, Reply-To:."
And the MUA needs to display some sort of sender information. I'd rather have all messages from Joe show as joe@foo.com rather than joe@whatever.server.i.sent.this.from. which may require the display of an additional from address (the Reply-To:, which isn't a from address at all).
I suppose more intelligent minds than me have worked hard at this whole email mess... but there has to be some way to verify or trust the sender without imposing rules on the From: header.
One more ding against the nipple/joystick: You move the stick down to scroll down, and proceed to hold it still while the scrolling continues. With the clickwheel, your thumb is in motion with the scrolling motion, and held still to hold the scrolling still.
I haven't used an iPod for more than a few seconds, but I think "non-intuitive" is a rather pointless statement. Sure, you might not get it immediately, but once someone shows you, it's a piece of cake; you probably won't forget it. Remember all the jokes about people trying to use the computer mouse as a foot pedal, or putting it on the screen? The mouse is simple enough, but only after you know. Nothing is really "intuitive". The real goal is to be easy to learn and use, and difficult to forget once learned, which I think the clickwheel achieves.
Also, the scrolling speed of the clickwheel is directly proportional to the speed with which you move your thumb. The scrollwheel on a mouse also achieves this, but I find it more difficult with the rubber nipple style scroll stick. Finally, to continue scrolling, you just keep moving your thumb in the same circular direction. The rubber nipple also requires no change to continue scrolling, but a mouse's scrollwheel does (you must scroll a bit, then move your finger to the top of the wheel and scroll some more). This combination of the scroll speed being directly related to your own thumb speed and the continuous motion for continuous scrolling is why I believe the clickwheel a is a good interface.
One point against it may be that the scrolling operation is up and down, while the thumb motion is in circles. I haven't put enough thought in to imagine a solution for that...
Just a small wish about file extensions: Can we simplify things and have .image, .video, .audio, .doc[ument] or something along those lines? Most of the time I don't really care what compression algorithm this image was compressed with. If I do, I can always view more detailed information about a file (e.g. with the "file" command on Linux, or possibly a right click->properties, or even show the info in one of the file manager views, whatever).
Though, I suppose this may be cumbersome on today's systems because some apps only handle e.g. .bmp, while others handle many image formats. To me this is a design flaw, but so fundamentally ingrained into systems with "applications" as to be a lost cause at the moment (I dream of a system where I only need to worry about, say, pdf support once. The system supports pdf. Done. None of this "oh, this program opens pdfs; that one can edit them; and this third one can save its documents as pdfs; but this other app can only save as postscript, so you need to use this converter...").
I'm trailing on into a rant at this point, but filenames (which shouldn't really be necessary in the first place, but assuming we're using a system that demands them) are the domain of the user, and should not be restricted to certain characters, or "must end with .extension" or any of that sillinies. Are you telling me that my 21st century computer can't keep track of what type of file it has? Oh, you are? In the name of backwards compatibility? Probably a necessary evil, but sheesh I hate it so.
With the wireless keyboard and Internet hot keys, you can go to your favorite places on the Web from anywhere in your room.
Whoa! Internet hotkeys! Sign me up!
Are there any better pictures? This thing looks fairly rickety and cumbersome to set up or move around. It's not exactly elegant looking.
AVP: No matter who wins, we lose.
CSI Miami has the worst dialog ever. David Caruso's character needs a long damn vacation.
Two detectives investigation a juror's possible murder by peanuts, find a peanut in the jury room
Detective 1: (picking up peanut) Take me out to ball game?
Detective 2: Buy me some peanuts.
Me: (shoots TV). Ah, blessed silence.
As the other ceramic engineer pointed out, glass is considered a ceramic. What makes it good for lenses is the fact that it is amorphous, i.e. has no periodic crystal structure. Thus, light will pass through it without being refracted in strange ways because on a macroscopic level, all the glass is the "same" (or the heterogeneity is small enough and distributed evenly and randomly).
Now, I didn't read the article, though I wouldn't expect it to get into details. I assume by "ceramic" they mean "crystalline ceramic" (versus amorphous). When you make crystalline ceramics, you don't typically make one continuous crystal (like a polished gemstone), but many small (10 to 500 micron is typical) crystals (grains) surrounded by very thin amorphous regions (grain boundaries). Grain boundaries serve to refract the light (I suppose they are much larger and thus less heterogenous than anomalies in the amorphous glass), which is why ceramics (think dinner plate) are opaque. I'm getting to the limits of my [limited] knowledge, but I believe the only real way to make transparent crystalline ceramics is to use a single crystal which contains no grain boundaries. This tends to be costly due to the need to melt the material to coax it into a single crystal and the high melting temperatures of ceramics (see silicon industry).
Given that, I'm not sure why using a single crystal lens would have an advantage over an amorphous glass lens... Perhaps you can achieve a higher index of refraction.
Let's face it, our options suck. Nader? Ya right. That's a vote for Bush, not Nader.
The sooner you realize that the single vote under your direct control has essentially zero effect on the election as a whole (what would the outcome of the last election be if you had voted differently or not at all?), the sooner we can begin to repair the 2-party system that "sucks".
Note that in Florida 2000, the closest thing to an election coming down to a single vote that I've ever heard of, we proved that our voting system does not even have sufficient resolution to determine the outcome in the (next to impossible) event that the election does indeed come down to a single vote.
Can someone please make a car with a stereo that has an audio input? Does such a thing exist even in aftermarket? Assuming I already have a portable music player, I could just plug it in to my car stereo. Instead, I'm stuck using some pathetic mini FM transmitter, or cassette adapter if I have a cassette deck, or buying a whole new HD car stereo.
*sigh*