Technology in the Real World
I've been interested in computer for a long time, since I realized that you could get them to do *anything* you can think to program them to do. In fact, anything people can't get computers to do yet (perfect translation, etc.) is due to lack of understanding the process ourselves. Of course we have to understand how to do something before we can write specific instructions on how to do it.
More than that, however, early computers were limited by interfaces. For example, the first large computers could perform math operations, but they had to be specially coded and fed into the machine using switches or punched cards. The results would be output and deciphered in a similar way.
One can understand the reaction of a non-scientist who would say "Well that's great, it can do 100 math problems per second, but.... what does that mean to me?" Uses of early computers were limited. With the addition of a monitor and keyboard, suddenly they became much more compatible with humans. We could put data in, and get data out. Later, the mouse was invented and better graphics became possible. While this increased the usability of computers, it didn't really change the fact that they weren't really CONNECTED to anything. For example, if you wanted to add together 100 numbers from somewhere else, you would have to key them in. If the source was another computer system, you might be able to load a file or something, but it was all very manual.
I remember the show "Quantum Leap", in which the main character had a computer the size of a credit card which could project 3D holograms and access all sorts of knowledge, as well as control certain things. In the movies, it seems anything computerized can control anything electronic. While that would be nice, the truth is closer to the opposite situation. So for me there are three kinds of interaction to think about:
a. Interaction between the computer and the user.
b. Interaction between the computer and other computers and computerized devices.
c. Interactions between the computer and the natural environment.
A is about input and display devices.
B is about protocols, networks, and sensors.
C is about sensors.
The more these three things grow, the more useful computers will be in the real world. Processing power is already at a point where it is more than sufficient for many tasks. The ability of computers to connect wit the physical world, though, isn't.
Consider a door lock. You want to use your credit card and have it unlock the door (like many hotel doors and company offices with card-keys). The amount of processing power is truly trivial, but you need two things:
1. A sensor to detect your card.
2. A physical controller for the lock that your computer can control.
Since neither of these are standard equipment, your computer can't help you. Turning the lights in your house on and off has the same issue. Even computerized devices like your DVD player lack a standard interface (both hardware and software) for talking to your computer, so the control is not physically possible (even though it theoretically should be).
Likewise there have been programs to process photos for a long time, but they had to be scanned into the computer first. More recently they could be taken with a digital camera, and then downloaded - eliminating a step. Finally, web-cams became popular on laptop computers, and now we have a usable sensor for real-time interaction.
While the progress in interaction with other electronic devices has been disappointing, things have been moving forward in other areas - particularly environmental sensors. Lately, Web-cams have been increasing in quality and ubiquity, orientation sensors have started to be included in many laptops, and more mobile phones. Sony has started including smart-card sensors in their laptops, etc.
Of course hardware is only half of the solution. As the hardware becomes more common, more people will write code to use it, which will further drive it's adoption. As these things becomes more common, the processing power in modern PCs (and mobile phones, etc.) will be unleashed, making them truly useful to more people.
More than that, however, early computers were limited by interfaces. For example, the first large computers could perform math operations, but they had to be specially coded and fed into the machine using switches or punched cards. The results would be output and deciphered in a similar way.
One can understand the reaction of a non-scientist who would say "Well that's great, it can do 100 math problems per second, but.... what does that mean to me?" Uses of early computers were limited. With the addition of a monitor and keyboard, suddenly they became much more compatible with humans. We could put data in, and get data out. Later, the mouse was invented and better graphics became possible. While this increased the usability of computers, it didn't really change the fact that they weren't really CONNECTED to anything. For example, if you wanted to add together 100 numbers from somewhere else, you would have to key them in. If the source was another computer system, you might be able to load a file or something, but it was all very manual.
I remember the show "Quantum Leap", in which the main character had a computer the size of a credit card which could project 3D holograms and access all sorts of knowledge, as well as control certain things. In the movies, it seems anything computerized can control anything electronic. While that would be nice, the truth is closer to the opposite situation. So for me there are three kinds of interaction to think about:
a. Interaction between the computer and the user.
b. Interaction between the computer and other computers and computerized devices.
c. Interactions between the computer and the natural environment.
A is about input and display devices.
B is about protocols, networks, and sensors.
C is about sensors.
The more these three things grow, the more useful computers will be in the real world. Processing power is already at a point where it is more than sufficient for many tasks. The ability of computers to connect wit the physical world, though, isn't.
Consider a door lock. You want to use your credit card and have it unlock the door (like many hotel doors and company offices with card-keys). The amount of processing power is truly trivial, but you need two things:
1. A sensor to detect your card.
2. A physical controller for the lock that your computer can control.
Since neither of these are standard equipment, your computer can't help you. Turning the lights in your house on and off has the same issue. Even computerized devices like your DVD player lack a standard interface (both hardware and software) for talking to your computer, so the control is not physically possible (even though it theoretically should be).
Likewise there have been programs to process photos for a long time, but they had to be scanned into the computer first. More recently they could be taken with a digital camera, and then downloaded - eliminating a step. Finally, web-cams became popular on laptop computers, and now we have a usable sensor for real-time interaction.
While the progress in interaction with other electronic devices has been disappointing, things have been moving forward in other areas - particularly environmental sensors. Lately, Web-cams have been increasing in quality and ubiquity, orientation sensors have started to be included in many laptops, and more mobile phones. Sony has started including smart-card sensors in their laptops, etc.
Of course hardware is only half of the solution. As the hardware becomes more common, more people will write code to use it, which will further drive it's adoption. As these things becomes more common, the processing power in modern PCs (and mobile phones, etc.) will be unleashed, making them truly useful to more people.
Comments
Post a Comment