No such thing as just a "shortage"
on
The H-1B Swindle
·
· Score: 0
Supply and demand: In a free market, there is no such thing as a "shortage". A company willing to pay enough can hire as many programmers of whatever quality they wish. You only get a "shortage" if you try to hold something different to the market-clearing level. In this case, it seems that companies want to get programmers for below-market-clearing wages. What a shock.
You might look into using REALbasic for the conversion. For work that's probably a little less than using the MS conversion wizard, your application would run on Windows, Mac (OS X and 9) and Linux.
The question is beside the point. "As simple as possible, but no simpler" is the correct attitude. Once you get there, there is a boundary of adding features means you need more complexity, but we've got to get there first.
I've used hardly any program or device that manages to get there. The steering wheel is a good example. A good *nix shell is another.
I've never used a word processor that comes even close. Word is closest, but it's still awful (try changing a style definition).
It's economics that drives this: ease of use is a tremendous source of economic efficiency but it won't sell: a better GUI would save how many millions of man hours per year? But M$ won't make money out of it, because it will look *less* flashy.
The world would be better off if you made you program easier to use, but you shouldn't bother, because you won't get paid for it.
In short: you are right, but your Grandfather is even more right.
a joypad. Analog, but with "clicks" for cardinal directions. People want to play games, and up/down/left/right is basic.
Other ideas:
a chorded keyboard. Hold a Palm in your palm. Look where your fingers and thumbs are (probably along the sides). Put buttons there and let me enter text with combinations of buttons. Properly implemented, this would be at least as fast as a regular keyboard.
320x480 screen and a useful but very low-powered 3D accelerator.
a decent-resolution (at least 1MP, but why not 2?) camera. Once you have a memory and processor in this thing, how expensive can it be to add a CCD and a lens? Sound in and out also, obviously.
make the stylus a bluetooth microphone/earphone, if you really must make it a phone as well. I still think separate gadgets makes more sense, so I can access and enter information and talk at the same time.
WiFi. Duh.
GPS. Duh. Tie this into the appointments. Imagine being able to set an appointment along the lines of "The next time I'm at my parents' place, remind me to collect my hat".
A clip on the back. When I clip it to something, it switches to vibrate mode (optional). When it's on me, it vibrates. When it isn't, it beeps.
About the size of a regular wallet. ie bigger than a Palm, but smaller than a iPaq. It should fit easily in a pocket, but be no smaller than that.
The Newton user interface. The best computer user interface ever invented, by an incalculable margin.
Dylan does all that. And you can add multimethods (much more useful than many of the things you mention) and solid functional programming.
There is an excellent windows IDE (www.functionalobjects.com) and an open-source compiler that spits out C (www.gwydiondylan.org) for unix-like OSes.
But the arguments above that you don't want to use an obscure language are correct -- unless you're developing a game. Then you can use any obscure technology you like, because you just do the thing and ship it, and do a few updates, then in six months everyone has forgotten it. Look at how many games are written in UnrealScript. Dylan would be great for that.
But you put it on the desk and it just works like a regular mouse. It's just a mouse with more buttons that you can pick up and use in 3D whenever that's appropriate.
The mouse and keyboard work well together: the keyboard is very versatile (eg for games and typing), and you can switch very quickly between the keyboard and mouse without looking. The pen, OTOH, works poorly with the keyboard because it takes too long to switch: you have to locate the pen and pick it up in the right way before you can use it. You also have to hold it in the air rather than rest it on a surface, so it's more fatiguing.
I propose a development of the mouse I call the wand: it's shaped kind of like a mouse but can also be picked up and used in space, or stood on one end on the desk and used sort of like a joystick. It's sensitive to its orientation and motion in space, and can give tactile feedback as clicks and buzzes and things. It has buttons and levers and whatnot in suitably devious spots for all your fingers so you can work all sorts of games with it.
So: you play Quake with it by sliding it around on a desk, rotating and tilting it. You use it just like a mouse with your word processor. You do 3D modelling with it by just waving it around in the air. And so on.
It can have a little thumb joystick for even wilder input combinations.
It's portable. You can carry it in your pocket to use with a wearable computer (or it can be the computer). It's potentially reasonably cheap. It expands on an existing paradigm and hence is compatible with existing software and interfaces.
It slices! It dices! When will someone make me one?
The biggest development in the next century, hopefully in my lifetime!;-) will be the cure for biological mortality.
The biggest challenge we will face is how much we are willing to change from being human, to being either a melding of human and machine, or being a biological machine whose code/dna we manipulate to our own ends. Want the sense of smell of a dog? Or a prehensile tail? And do you really want your brain wired into a computer? This would also be to be wired into the 'net, along with the thoughts of all the other people who do the same thing. Do you *really* want telepathy? What happens to individuality then..?
The coming century is where we confront what it means to be human, when we can *choose* what that is.
Premise: the demand for computer skills is a normal downward-sloping curve.
Conclusion: the price of computer skills will fall. The number of employed computer skills will rise.
Premise: computer skills are a bottleneck in economic growth.
Conclusion: economic growth will rise.
Further conclusion: demand for computer skills will rise faster than it would have had it not been increased in the first place.
Comment: if the price isn't constrained, there is no such thing as over- or under-supply. There is just a different market-clearing price.
Further comment: if all the above is fairly close to right, the only people who might lose will be those with computer skills already in the US, and then perhaps only in the short run.
The features of the language are very important for this. The improvements to OOP that a language like Dylan brings (its concern for efficiency along with the modules and multimethods and a few other things)mean that you can build libraries once for most design patterns, for example, and then just use them.
It's also interesting how much easier Dylan makes plugging into something like CORBA or COM.
Until our languages improve, there are a lot of compromises in trying to widen use of components.
History seems to teach that there are strong institutional factors opposing programming language change. I guess the economics of the investment in existing code and existing skills makes the cost of changing programming language high, at least for most purposes.
Otherwise, we'd all have adopted a programming language like Dylan well before now (take a look at http://www.functionalobjects.com for more on Dylan, or the recent Byte columns http://www.byte.com/column/BYT20000601S0003 and http://www.byte.com/column/BYT20000628S0007 for more on this neat language that gives you efficiency along with improved OOP and functional programming).
The other thing I find interesting is that most of the innovations on the desktop are now being driven by games. This applies in both hardware and software. Unrealscript has some neat ideas, and Tim Sweeney wrote an essay (I can't find the URL) where he discusses a lot of pretty smart ideas about where programming languages should be going, and he's apparently going to do a newer and better language for his next major project.
I've wondered if it isn't possible to attach some sensors to the appropriate nerves and pick up enough information about what we're doing with our tongue, pharynx etc to recognise silently "voiced" words. If we could use a few discreet sensors to get this to work, it would get around a lot of the problems that are mentioned here, and some more (such as background noise). Perhaps this could supplement regular voice recognition, too? Anyone who's worked in the field care to comment?
Slashdot Mirror
Supply and demand: In a free market, there is no such thing as a "shortage". A company willing to pay enough can hire as many programmers of whatever quality they wish. You only get a "shortage" if you try to hold something different to the market-clearing level. In this case, it seems that companies want to get programmers for below-market-clearing wages. What a shock.
You might look into using REALbasic for the conversion. For work that's probably a little less than using the MS conversion wizard, your application would run on Windows, Mac (OS X and 9) and Linux.
The question is beside the point. "As simple as possible, but no simpler" is the correct attitude. Once you get there, there is a boundary of adding features means you need more complexity, but we've got to get there first.
I've used hardly any program or device that manages to get there. The steering wheel is a good example. A good *nix shell is another.
I've never used a word processor that comes even close. Word is closest, but it's still awful (try changing a style definition).
It's economics that drives this: ease of use is a tremendous source of economic efficiency but it won't sell: a better GUI would save how many millions of man hours per year? But M$ won't make money out of it, because it will look *less* flashy.
The world would be better off if you made you program easier to use, but you shouldn't bother, because you won't get paid for it.
In short: you are right, but your Grandfather is even more right.
Most obvious, glaring omission:
a joypad. Analog, but with "clicks" for cardinal directions. People want to play games, and up/down/left/right is basic.
Other ideas:
a chorded keyboard. Hold a Palm in your palm. Look where your fingers and thumbs are (probably along the sides). Put buttons there and let me enter text with combinations of buttons. Properly implemented, this would be at least as fast as a regular keyboard.
320x480 screen and a useful but very low-powered 3D accelerator.
a decent-resolution (at least 1MP, but why not 2?) camera. Once you have a memory and processor in this thing, how expensive can it be to add a CCD and a lens? Sound in and out also, obviously.
make the stylus a bluetooth microphone/earphone, if you really must make it a phone as well. I still think separate gadgets makes more sense, so I can access and enter information and talk at the same time.
WiFi. Duh.
GPS. Duh. Tie this into the appointments. Imagine being able to set an appointment along the lines of "The next time I'm at my parents' place, remind me to collect my hat".
A clip on the back. When I clip it to something, it switches to vibrate mode (optional). When it's on me, it vibrates. When it isn't, it beeps.
About the size of a regular wallet. ie bigger than a Palm, but smaller than a iPaq. It should fit easily in a pocket, but be no smaller than that.
The Newton user interface. The best computer user interface ever invented, by an incalculable margin.
Dylan does all that. And you can add multimethods (much more useful than many of the things you mention) and solid functional programming.
There is an excellent windows IDE (www.functionalobjects.com) and an open-source compiler that spits out C (www.gwydiondylan.org) for unix-like OSes.
But the arguments above that you don't want to use an obscure language are correct -- unless you're developing a game. Then you can use any obscure technology you like, because you just do the thing and ship it, and do a few updates, then in six months everyone has forgotten it. Look at how many games are written in UnrealScript. Dylan would be great for that.
But you put it on the desk and it just works like a regular mouse. It's just a mouse with more buttons that you can pick up and use in 3D whenever that's appropriate.
The mouse and keyboard work well together: the keyboard is very versatile (eg for games and typing), and you can switch very quickly between the keyboard and mouse without looking. The pen, OTOH, works poorly with the keyboard because it takes too long to switch: you have to locate the pen and pick it up in the right way before you can use it. You also have to hold it in the air rather than rest it on a surface, so it's more fatiguing.
I propose a development of the mouse I call the wand: it's shaped kind of like a mouse but can also be picked up and used in space, or stood on one end on the desk and used sort of like a joystick. It's sensitive to its orientation and motion in space, and can give tactile feedback as clicks and buzzes and things. It has buttons and levers and whatnot in suitably devious spots for all your fingers so you can work all sorts of games with it.
So: you play Quake with it by sliding it around on a desk, rotating and tilting it. You use it just like a mouse with your word processor. You do 3D modelling with it by just waving it around in the air. And so on.
It can have a little thumb joystick for even wilder input combinations.
It's portable. You can carry it in your pocket to use with a wearable computer (or it can be the computer). It's potentially reasonably cheap. It expands on an existing paradigm and hence is compatible with existing software and interfaces.
It slices! It dices! When will someone make me one?
Why aren't we making spinning space stations like on 2001? Surely this would solve this whole problem.
http://www.ddj.com/articles/2001/0165/0165f/0165f. htm
The biggest development in the next century, hopefully in my lifetime! ;-) will be the cure for biological mortality.
The biggest challenge we will face is how much we are willing to change from being human, to being either a melding of human and machine, or being a biological machine whose code/dna we manipulate to our own ends. Want the sense of smell of a dog? Or a prehensile tail? And do you really want your brain wired into a computer? This would also be to be wired into the 'net, along with the thoughts of all the other people who do the same thing. Do you *really* want telepathy? What happens to individuality then..?
The coming century is where we confront what it means to be human, when we can *choose* what that is.
Premise: the demand for computer skills is a normal downward-sloping curve. Conclusion: the price of computer skills will fall. The number of employed computer skills will rise. Premise: computer skills are a bottleneck in economic growth. Conclusion: economic growth will rise. Further conclusion: demand for computer skills will rise faster than it would have had it not been increased in the first place. Comment: if the price isn't constrained, there is no such thing as over- or under-supply. There is just a different market-clearing price. Further comment: if all the above is fairly close to right, the only people who might lose will be those with computer skills already in the US, and then perhaps only in the short run.
The features of the language are very important for this. The improvements to OOP that a language like Dylan brings (its concern for efficiency along with the modules and multimethods and a few other things)mean that you can build libraries once for most design patterns, for example, and then just use them. It's also interesting how much easier Dylan makes plugging into something like CORBA or COM. Until our languages improve, there are a lot of compromises in trying to widen use of components.
History seems to teach that there are strong institutional factors opposing programming language change. I guess the economics of the investment in existing code and existing skills makes the cost of changing programming language high, at least for most purposes. Otherwise, we'd all have adopted a programming language like Dylan well before now (take a look at http://www.functionalobjects.com for more on Dylan, or the recent Byte columns http://www.byte.com/column/BYT20000601S0003 and http://www.byte.com/column/BYT20000628S0007 for more on this neat language that gives you efficiency along with improved OOP and functional programming). The other thing I find interesting is that most of the innovations on the desktop are now being driven by games. This applies in both hardware and software. Unrealscript has some neat ideas, and Tim Sweeney wrote an essay (I can't find the URL) where he discusses a lot of pretty smart ideas about where programming languages should be going, and he's apparently going to do a newer and better language for his next major project.
I've wondered if it isn't possible to attach some sensors to the appropriate nerves and pick up enough information about what we're doing with our tongue, pharynx etc to recognise silently "voiced" words. If we could use a few discreet sensors to get this to work, it would get around a lot of the problems that are mentioned here, and some more (such as background noise). Perhaps this could supplement regular voice recognition, too? Anyone who's worked in the field care to comment?