Schoolin'
10/29/05
Countdown: 41 days until December 9
No, I didn't go back to school. But I have been at a training class all week learning to use the Agilent 93000 SOC tester. It's really a nice piece of equipment. The interface is nothing pretty but it's really fast. The debug tools are probably the most helpful tools I've seen on a tester. The waveform tool really helped me see what the device is outputting and when the tester is trying to compare. If only we had used this tester for our past products it would have saved a lot of debug time. But having used a Teradyne J973 for a bit, I must say that the 93k's shmoo tool isn't nearly as cool as the J973's. The 93k also has a weird way of setting up waveforms. The test program contains all possible waveforms for each pin or groups of pins and then the test vectors just indicate which pre-determined vector to use when. I suppose that might save some execution time over generating the waveforms on the fly, but it can be pretty limiting since you can only define 32 different waveforms for each pin. The I/Os can run at 1.25Gbps using some tricks with the waveforms so maybe the way they do it is useful after all. The training class was good though. It's nice to have a very knowledgeable instructor who not only goes over the course material but gives some practical advice and historical context on the equipment. The sucky part of the training was that it started between 7:30am and 8:30am every day and sometimes lasted until 6pm. Man, that was painful.
Recently, I've realized that education has one goal: seeing into the future. Everything we have been taught since pre-school is either directly or indirectly related to prescience. No one is expecting to become Paul Atreides or Miss Cleo, but those of us with better training are better able to predict how our actions will affect the future. (I take that back, I was hoping to become Paul Atreides but I lost the silly ring.) In early education, we learn to read, write, and do math. All of these basic skills are required building blocks to facilitate the study of arts and science which come later in school. Studying arts and sciences provides knowledge of how things work and knowing how things work allows us to know how our actions today will affect the future.
A simple example is in civil engineering. During the design phase for a highway overpass, a engineer must choose the formula for the concrete that will support the required load and survive the weather conditions of the locale without actually building an overpass. So before construction begins, the engineer makes a prediction: the chosen recipe of concrete will adequately support the predicted traffic load. Without sound knowledge of the properties of available concrete, the engineer could not make this prediction and the overpass would be constructed with some sort of trial-and-error method. By being able to make this (hopefully) accurate prediction, the engineer has turned the costly idea of constructing an overpass into a more practical idea.
What about the arts? Take a writer, for example. A writer needs to know which words to use to create a certain mood for the passage he or she is writing. The author needs to predict how a plot element at one point in the tale will affect the outcome of the story and then predict the reader's reaction to the events in the plot. From a practical standpoint, writers need to predict how well their pieces will be received by the intended audience so they can get paid and be able to feed their cats. Because all writers have cats to feed.
Athletes need to predict their performance in upcoming events. Doctors need to predict how their patients will react to certain treatments so they can effectively heal the sick. Presidents need to predict which mustached dictators have weapons of mass destruction so they can order the troops to premptively strike. Potheads need to predict which snacks will be the most effective in curing the munchies so that they can enjoy the whole smoking experience without making a trip to 7-11. Anyone who does anything is, in one way or another, peering into a crystal ball of knowledge and trying to catch a glimpse of the future. And the crystal ball can only be obtained through education. (Yes, even the pothead needs to be actively enrolled in the school of life to know that Ding Dongs only make you hungrier when you're high.)
(That last bit may or may not be true. I only wrote it because I wanted to say Ding Dongs. There, I said it again.)
The most successful people today (in all forms of "success") are the ones who have had the most effective training in predicting the future. (Or are just the luckiest, I can't tell the difference.) The epitome of any education system is perfect fortune telling. We're nowhere close to perfect. However, we're on the right path. We just need to keep on learning.
Here's a little piece of trivia: shmoo plots were named at Fairchild Semiconductor back in the day. A guy brought his kid to work one day. The kid saw a 2D sweep plot and called it a Shmoo because, well I guess it looked like one. Now the whole industry calls them shmoo plots. The icon for Teradyne's shmoo tool is a picture of a Li'l Abner shmoo. If you don't know what a shmoo plot is, get thee to a test house. Or Wikipedia. But I guess this has nothing to do with seeing into the future.
No, I didn't go back to school. But I have been at a training class all week learning to use the Agilent 93000 SOC tester. It's really a nice piece of equipment. The interface is nothing pretty but it's really fast. The debug tools are probably the most helpful tools I've seen on a tester. The waveform tool really helped me see what the device is outputting and when the tester is trying to compare. If only we had used this tester for our past products it would have saved a lot of debug time. But having used a Teradyne J973 for a bit, I must say that the 93k's shmoo tool isn't nearly as cool as the J973's. The 93k also has a weird way of setting up waveforms. The test program contains all possible waveforms for each pin or groups of pins and then the test vectors just indicate which pre-determined vector to use when. I suppose that might save some execution time over generating the waveforms on the fly, but it can be pretty limiting since you can only define 32 different waveforms for each pin. The I/Os can run at 1.25Gbps using some tricks with the waveforms so maybe the way they do it is useful after all. The training class was good though. It's nice to have a very knowledgeable instructor who not only goes over the course material but gives some practical advice and historical context on the equipment. The sucky part of the training was that it started between 7:30am and 8:30am every day and sometimes lasted until 6pm. Man, that was painful.
Recently, I've realized that education has one goal: seeing into the future. Everything we have been taught since pre-school is either directly or indirectly related to prescience. No one is expecting to become Paul Atreides or Miss Cleo, but those of us with better training are better able to predict how our actions will affect the future. (I take that back, I was hoping to become Paul Atreides but I lost the silly ring.) In early education, we learn to read, write, and do math. All of these basic skills are required building blocks to facilitate the study of arts and science which come later in school. Studying arts and sciences provides knowledge of how things work and knowing how things work allows us to know how our actions today will affect the future.
A simple example is in civil engineering. During the design phase for a highway overpass, a engineer must choose the formula for the concrete that will support the required load and survive the weather conditions of the locale without actually building an overpass. So before construction begins, the engineer makes a prediction: the chosen recipe of concrete will adequately support the predicted traffic load. Without sound knowledge of the properties of available concrete, the engineer could not make this prediction and the overpass would be constructed with some sort of trial-and-error method. By being able to make this (hopefully) accurate prediction, the engineer has turned the costly idea of constructing an overpass into a more practical idea.
What about the arts? Take a writer, for example. A writer needs to know which words to use to create a certain mood for the passage he or she is writing. The author needs to predict how a plot element at one point in the tale will affect the outcome of the story and then predict the reader's reaction to the events in the plot. From a practical standpoint, writers need to predict how well their pieces will be received by the intended audience so they can get paid and be able to feed their cats. Because all writers have cats to feed.
Athletes need to predict their performance in upcoming events. Doctors need to predict how their patients will react to certain treatments so they can effectively heal the sick. Presidents need to predict which mustached dictators have weapons of mass destruction so they can order the troops to premptively strike. Potheads need to predict which snacks will be the most effective in curing the munchies so that they can enjoy the whole smoking experience without making a trip to 7-11. Anyone who does anything is, in one way or another, peering into a crystal ball of knowledge and trying to catch a glimpse of the future. And the crystal ball can only be obtained through education. (Yes, even the pothead needs to be actively enrolled in the school of life to know that Ding Dongs only make you hungrier when you're high.)
(That last bit may or may not be true. I only wrote it because I wanted to say Ding Dongs. There, I said it again.)
The most successful people today (in all forms of "success") are the ones who have had the most effective training in predicting the future. (Or are just the luckiest, I can't tell the difference.) The epitome of any education system is perfect fortune telling. We're nowhere close to perfect. However, we're on the right path. We just need to keep on learning.
Here's a little piece of trivia: shmoo plots were named at Fairchild Semiconductor back in the day. A guy brought his kid to work one day. The kid saw a 2D sweep plot and called it a Shmoo because, well I guess it looked like one. Now the whole industry calls them shmoo plots. The icon for Teradyne's shmoo tool is a picture of a Li'l Abner shmoo. If you don't know what a shmoo plot is, get thee to a test house. Or Wikipedia. But I guess this has nothing to do with seeing into the future.