What id like to know is, why hasnt some stupid asshole just made smart burning software, so you can drag in 2gig of files and say "BURN THIS BABY" and it would burn the first 700meg, and ask to insert CD2 and burn the next, then cd3, all without having to manually seperate it out. Code wise, no more than a days work to detail the spec and logic/flow charts, and a day or two to code it with 2 days to properly test it. Total price? $1k-$2k at most by the software companies, or $900k for MS which is why it would fail the ROI
So if it's only four days work why haven't you done it yourself?
I feel that math, science, and engineering courses are very important - for mathmeticians, scientists, and engineers (and others who have an interest in said subjects).
But when they enter college many of the mathematicians, scientists, and engineers don't yet know they are going to be mathematicians, scientists, and engineers! Surely one of the functions of higher education is to introduce students to a number of fields, many of which will be soon forgotten, but one of which may lead to a vocation. My first college math class stood my notion of what math was on its head.
Really, which is more important: taking a trig class you'll soon forget, or knowing how to read at a passable level? It's not a difficult call.
Agreed, you certainly won't be able to work competently in math or science without the ability to read and write at a passible level. But why is it a dichotomy? I think a university graduate should be able to read and write and do analytic geometry and basic trig. Tens of thousands of students do graduate each year with these skills, so I don't see why it is unreasonable to ask for it.
Many university graduates will be citizens in democracies and faced with decisions (at least by proxy) about science and technology policy. Wouldn't it be desirable for them to have some basic grounding in science and engineering?
Abstract math as a central subject of high education is not an recent invention. I gather you are a fan of philosophy. Were you aware that the entrance to Plato's Athenian academy is said to have born the inscription "Let no one ignorant of Mathematics enter here"?
What passes for 'education' in the US is gameshow knowledge; memorizing a nearly endless stream of disjointed factoids just long enough to repeat it on a test.
Some mediocre teachers do over-emphasize memorization because it is easy to teach, and even easier to measure. On the other hand I think many bright young minds dismiss memorization too readily because it is tedious and time-consuming. Surely you'd acknowledge that having a grammar and a English-French dictionary is not the same thing as speaking French? Most fields have some base set of facts that must be learned, and memorization is neglected at your peril.
We need to stop kidding ourselves - education in the US isn't worth a damn.
I disagree. I think the US educational system presents opportunities for formal education unparalleled in the history of the world. Yes, it is possible to sleepwalk through Enormous State University and emerge no wiser, and no more educated then when you entered. It is also possible to get a terrific education, but it requires discipline, motivation, and experimentation the part of the student. Unfortunately as many have remarked, "You can lead a student to knowledge, but you can't make them think." Witness the very topic of this thread.
I'm sorry your own experience was so dismal, and I hate to be adversarial, but did you try independent study, study abroad, student reading groups, sitting in on graduate classes, getting involved in research, or did you just fill in the check boxes on your degree requirements?
How anyone can claim it is when we place such heavy emphasis on largely unnecessary, irrelevant and highly specialized classes (like calculus)
Wait a second, this sounds like "assumption of facts not in evidence" to me. Most high schools in the US don't even teach calculus let alone emphasize it. Most universities and colleges that I know (and I know several) will grant a B.A. degree with only a basic algebra or stats class and maybe a "history of mathematics" class. Where is this "heavy emphasis" you are claiming?
Just about any CS degree is about 10 years behind current technology most of the time.
This is only true if you passively accept lectures and homework as the whole of your education. If you show some initiative, go to the seminars, try to read the journals, and get involved in research projects, you can work with technology that won't be state of the art for another ten years.
Just last week I contacted my agent to try to negotiate a better rate with my current employer and one of the reasons was that I am graduating in two weeks with a Math and Comp. Sci. degree. She basically told me that it isn't worth a cent in terms of my rate of pay!
Which would seem to contradict your cliam that the only reason for a CS degree is to improve once's income. Maybe your agent is right and a Math-CS degree is not worth a cent in your pay rate, but it certainly ought to make a difference in the kind of jobs you are qualified for.
but even these terms are deceptive because people say if the micro happens then why shouldn't it lead to the macro?
Seems like a valid question to me. Why wouldn't the continual change of allele frequencies over 10 million years inevitibly lead to species divergence and gross changes in phenotype?
You can get that plus money too. It's called "working".
Er, yes I suppose that is true, but you do run the risk that instead of hearing: "You didn't do well on your last assignment, come round at office hours and we'll go over it." you'll hear: "You didn't do well on your last assignment, head over to personel and pick up your severence check."
Going to classes for material that I can better teach myself has kept me from going back to college in the first place
A problem with self-education is that it can be too easy to convince yourself that you've understood something when you really haven't. A real benefit of a college or univeristy education is having a highly trained critic reviewing your work and telling you when you've screwed up.
If the expense were negligable (say 50 cents or so) that would be different. Honestly though, I would rather use the money to feed a few hungry children.
The Freedom of Information Act allows agencies to recover the direct costs of searching, vetting, and duplicating the information. FOIA requests do not compete with hungry children for your tax dollars.
I'm not entirely sure what you mean by "this new ignorance is bliss" trend. I see it more as a division of labor. The problems I'm working with (identifying related DNA or protein sequences using support vector machines) are hard enough to understand in themselves that if I had to worry about finding buffer overruns in my assembly code I'd go nuts. Thank god for languages with run-time bounds checking and garbage collection.
I've also heard it remarked that debugging is twice as hard as programming so if you are as clever as you can be when you write your program you are never going to be able to debug it.
Knowing more is almost always better then knowing less, but I'm still going to try to disagree with you. The problem is that the breadth of computer science has grown so in the last 30 years that there is almost always a 'Guns vs Butter' choice to be made in the course of an education. The time I spent on exercises in assembly language was time I wasn't able to spend on algorithms for computational biology and visa versa.
A few years ago I worked with a gentleman who had graduated from a trade school. He was a generally sharp programmer, but he really put me to shame in database design and SQL. He had completely internalized the relational model, and could write a query in five minutes that was an order of magnitude faster a query I'd taken an hour to come up with. On the other hand he had no idea how floating point numbers were represented in the computer (he assumed everything was BCD). In database design there are so many levels of abstraction between you and the hardware that understanding the CPU architecture is of very limited utility.
I think open source works just dandy for bioinformatics software, and is a hopeless pipe-dream for wet-bench biotech. Bio-tech is like software development in two ways: first, manufacturing and reverse engineering costs are a fraction of the cost of development, second they both require an educated and skilled workforce. Bio-tech is unlike software development because it MUST have a substantial physical plant, and the cost of physical consumables typically exceeds the costs of salaries.
These machines are expensive in part because of the limit numbers produced. If there was more demand for the machines, then the companies producing them would be able to lower margins by taking advantage of mass production techniques
They are also expensive in part because they are complex and contain precision components, so they might not get as cheap as you hope. But granting you that we get -80C freezers for the same cost as domestic fridges, we still have to provide wiring and power for them and the square footage for them to sit on (they are not small and you may need three or four of them per lab). Forget the hi-tech stuff, just consider paying for pipette tips, rubber gloves, and distilled water. Software development doesn't have equivalent consumables.
Do you have an open source model for the army of technicians you will need? I mean I can see a PI working on a project for the excitement, fun and prestige, but what about the research techs doing everything from washing dishes to performing autopsies on the lab mice. Unlike programmers working on open source software they won't be able to tele-commute and a part-time commitment would be problematic.
After all, it is common knowledge how to make donuts, but lots of companies do it and make good money doing it
Yes, but none of them had to invest a couple hundred million to figure out how to make dougnuts. The issue is not recouping manufacturing costs, the issue is recovering R&D costs. The money can come from compulsory taxation (as with NIH/NSF), voluntary taxation (as with the Polio vaccine), or it can come from speculative investment. I still haven't seen anyone explain how open source is going to raise the money.
the government can simply grant limited production rights to the company that puts up the cash to get something through trials
Limited production rights? isnt't that basically a patent?
Many math programs place the calculus sequence as in introduction to mathematics to force a mathematical way of thinking, but a discrete math sequence could serve the same purpose
I don't think it is just tradition. Calculus has the advantage of being relevent to many technical fields while discrete math has (so far) a more limited scope. After all, these are first year undergraduates we're talking about. Many of them will be end up switching to other majors. Having gotten calculus out of the way in the first year keeps a lot of options open for the second year.
Perhaps it is a chicken and egg problem, but I would think that most first year students would be baffled by the relevence of discrete math until they've written a few programs with non-trivial data structures. On the other hand, most of them will have a nodding acquaintance with area, velocity, and aceleration.
Calculus, I don't think you actually need it to be a programmer
Anything in machine learning, computer vision, natural language processing, data mining, even network theory, will require a strong and advanced Statistics and Probability class
You are going to have a pretty rough time in your advanced statistics and probability class if you don't know calculus!
GM and selective breeding are two TOTALLY different processes
They are certainly different in the details of the process, but you'd be suprised at how much they have in common at the lowest level.
Both amplify a novel allele. In the case of breeding you have to wait for a mutation that produces the trait you want. In many cases the mutation occured a long time ago but just never spread widely in the population. However it was still a mutation, and a therefore a gene not found in the species until that point. Furthermore, genes do jump from species to species in nature (google "horizontal gene transfer").
I don't doubt that GM poses hazzards that deserve consideration. I don't understand though why people assume that the random processes of nature are benign.
Tomatoes have never needed fish genes before, so why would they suddenly need them now?
Humans never needed penecillin before 1920, so why should they suddenly need it now?
Breeding doesn't modify genes per se, but to make breeding fruitful (pun intended) you must have a source of novel genes. Some of the mutations may have taken place a long time ago, but the desireable alleles certainly came from somewhere.
There seems to be an undercurrent that novel genes introduced by a random process are "good" while novel genes introduced by design are "evil". While I can understand the mistrust of design, I don't understand why people assume a random process is begign.
I think you overestimate the value of the command line interface for the "long run". Command lines are efficient for tasks that are done frequently. However, computer users at all levels are faced with chores that come up infrequently but require a lot of technical knowledge to manage in their bare form. Good GUIs and menu systems are valuable because they cue the user to possible actions, and prompt the user with information they need to make decisions. When a casual user sits in front of a command prompt there are no cues, and they have to depend entirely on their training and memory. If it's done right, "eye candy" provides reminders of what it is possible to do and makes suggestions as to what should be done.
Eric Raymond's rant about UI was triggered by a GUI that was not good. It required him to understand the contents of the samba.conf file just as thoroughly as a sys admin editing it with emacs. He was pissed off because he had to spend hours learning arcane minutiae that won't leverage to any other area of his technical life, and will be forgotten if not obsolete by the next time he finds himself installing a network printer. An administrator who is setting up printers every week probably won't touch a GUI no matter how good, but a good GUI (or menu system) would have given Mr. Raymond several hours of his life back.
I'll chime in to support the parent. Despite the tremendous strides it has made in a very short time, the Linux desktop is still not suitable for general consumption. Did you read Eric Raymond's rant on the Linux user experience? Gnome and KDE are worthy efforts which I use daily, but as Mr. Raymond points out there is more to creating a productive end user interface then poping up dialog boxes that allow you to type in the same obscure strings that you would otherwise have typed into/etc/foo.conf using emacs.
I'd abandoned Apple back in the OS 8 days because I'd though the OS was languishing, but I just bought an iBook P4, and with OS X Apple really has made a Unix desktop that really is usable by the legions of Aunt Tillies.
You obviously know very little about computer science. How pray tell, does integral calculus have anything at all to do with the design of a programming language?
It is getting to be one of my pet peeves that folks take their little corner of computer science and presume it to be the whole thing. I think you are probably right: integral calculus is probably not a useful tool in the design of computer languages. However, this ignores the dozens of other subfields of computer science where the integral calculus is essential. Anything involving stochastic modeling is going to pull in probability theory, and you won't get far in probablility theory without the basics of integral/differential calculus. On top of this most programmers and computer scientists are using computers to solve some problem external to computer science. To understand many of those problems you will need the basics of integral/differential calculus.
I am still trying to figure out why Integral Calculus is forced down everyone's throat. Computer Scientists are better off studying proof theory, axiomatic set theory, lambda-calculi, etc...
There are subfields within CS that make use of Integral Calculus... but most subfields of CS do not use it and instead use things like proof theory, set theory, etc
Most undergraduates do not have an accurate idea which field, let alone which subfield, they are going to end up in. The integral/differential calculus is a gateway topic. If you don't have it you've foreclosed study in many areas: physical sciences, numerical methods, probability and statistics, much of engineering, economics etc. I agree that the lambda calculus belongs in every CSci undergrad education, but lacking proof theory or axiomatic set theory closes very few doors, interesting topics though they may be.
In practice, the average CSci B.S. is going to end up writing database front ends for a business somewhere, so you could argue that all of these topics are academic and irrelevant. I just think that the first two years of college are too soon to drop out from so many areas of human knowledge.
Those sound like perfectly reasonable problems. I think the time constraint might be a bit rigid giving that the problems are coming at them with no warning in a stressful environment. Are you standing next to them with a stopwatch counting out the time in 10 second intervals?
How big is your sample size? My experience as a developer tells me that there is a wide distribution of programming skill in the world, so if you have a decent sample size and your test says that everyone is an idiot, then either something is wrong with your test, or something is wrong with your recruiting.
I've suffered the same disappointment, but this is one area where I think the employer is justified in imposing very strong filters. I believe there is an imposing body of FDA regulations governing the development of software for medical devices and it makes sense to try to find programmers who have already worked in that framework. Similarly for avionics systems.
"Why do I have to take a course in assembly? We never need to program in it."
The exponentially expanding breadth of human knowledge drives the division of labor. Even computer programming is now so broad a field that sub-specialists can legitimately ask questions like "Why do I need to understand assembly language?" Certainly such a question would seem foolish coming from someone specializing in procssor engineering or compiler design, but is it equally foolish coming from someone specializing in bioinformatics or artificial intelligence?
Why stop at assembly language? Should all programmers be able to write microcode? How deeply should they understand logic gate arrays? It would be wonderful if we could all understand everything, but life is fleeting and there are guns versus butter choices to be made in our educations.
I knew my way around regedit and the Windows registry; he didn't.
But why is knowing your way around the windows registry more fundamental to life as a free citizen then understanding any of the other myriad technologies we depend on every day? How deep is your understanding of AC power systems? Ceramics? Metallurgy? Polymerase chain reaction DNA amplification? Diesel engines? Domestic plumbing? Carpentry? Each of us stands on top of vast mountains of diverse technology which we have no technical expertise in.
I can understand complaints about the general level of technical education. I don't understand why so many computer technologists insist that the particular technology they are expert in is the technology every good citizen must learn.
But I don't want people to become better end users! I want them to become better at whatever it is they actually do for a living. It is my job as a programmer/sysadmin to try to advance their goals using my years of specialized study in computer technology. It is called the division of labor, and given the depth and sophistication of our technologies it is the only way to go.
I had to have a CAT scan today. CAT is an incredible blending of multiple advanced technologies: math, physics, electronics, computers and medicine. It's run by a technician who knows the gory details and gotcha's of its operation, and the results are interpreted by a Radiologist who usually doesn't even touch the machine any more. I don't want my Radiologist spending his or her professional time studying processor architectures or HTML. I want them studying their radiology journals and sharpening their diagnostic skills.
I may be wrong, but I had always been told that once you have been infected by a virus, you can't be infected again.
Things are never that simple in biology. Immunity depends (in part) on keeping around at least a few members of the class of white blood cells that recognize the disease organism. If you rarely encounter the target bug that class of white blood cells will dwindle away, and given enough time and bad luck may die out entirely leaving you vulnerable to another attack. On top of that, the virus may change subtley over time so that the white blood cells that recognized last years virus doesn't recognize this year's virus.
Slashdot Mirror
Agreed, you certainly won't be able to work competently in math or science without the ability to read and write at a passible level. But why is it a dichotomy? I think a university graduate should be able to read and write and do analytic geometry and basic trig. Tens of thousands of students do graduate each year with these skills, so I don't see why it is unreasonable to ask for it.
Many university graduates will be citizens in democracies and faced with decisions (at least by proxy) about science and technology policy. Wouldn't it be desirable for them to have some basic grounding in science and engineering?
Abstract math as a central subject of high education is not an recent invention. I gather you are a fan of philosophy. Were you aware that the entrance to Plato's Athenian academy is said to have born the inscription "Let no one ignorant of Mathematics enter here"?
I disagree. I think the US educational system presents opportunities for formal education unparalleled in the history of the world. Yes, it is possible to sleepwalk through Enormous State University and emerge no wiser, and no more educated then when you entered. It is also possible to get a terrific education, but it requires discipline, motivation, and experimentation the part of the student. Unfortunately as many have remarked, "You can lead a student to knowledge, but you can't make them think." Witness the very topic of this thread.
I'm sorry your own experience was so dismal, and I hate to be adversarial, but did you try independent study, study abroad, student reading groups, sitting in on graduate classes, getting involved in research, or did you just fill in the check boxes on your degree requirements?
For another view contradicting yours, consider this: Woman rejected by Oxford college urges study in US
Wait a second, this sounds like "assumption of facts not in evidence" to me. Most high schools in the US don't even teach calculus let alone emphasize it. Most universities and colleges that I know (and I know several) will grant a B.A. degree with only a basic algebra or stats class and maybe a "history of mathematics" class. Where is this "heavy emphasis" you are claiming?
This is only true if you passively accept lectures and homework as the whole of your education. If you show some initiative, go to the seminars, try to read the journals, and get involved in research projects, you can work with technology that won't be state of the art for another ten years.
Which would seem to contradict your cliam that the only reason for a CS degree is to improve once's income. Maybe your agent is right and a Math-CS degree is not worth a cent in your pay rate, but it certainly ought to make a difference in the kind of jobs you are qualified for.
Seems like a valid question to me. Why wouldn't the continual change of allele frequencies over 10 million years inevitibly lead to species divergence and gross changes in phenotype?
I'm not entirely sure what you mean by "this new ignorance is bliss" trend. I see it more as a division of labor. The problems I'm working with (identifying related DNA or protein sequences using support vector machines) are hard enough to understand in themselves that if I had to worry about finding buffer overruns in my assembly code I'd go nuts. Thank god for languages with run-time bounds checking and garbage collection.
I've also heard it remarked that debugging is twice as hard as programming so if you are as clever as you can be when you write your program you are never going to be able to debug it.
Knowing more is almost always better then knowing less, but I'm still going to try to disagree with you. The problem is that the breadth of computer science has grown so in the last 30 years that there is almost always a 'Guns vs Butter' choice to be made in the course of an education. The time I spent on exercises in assembly language was time I wasn't able to spend on algorithms for computational biology and visa versa.
A few years ago I worked with a gentleman who had graduated from a trade school. He was a generally sharp programmer, but he really put me to shame in database design and SQL. He had completely internalized the relational model, and could write a query in five minutes that was an order of magnitude faster a query I'd taken an hour to come up with. On the other hand he had no idea how floating point numbers were represented in the computer (he assumed everything was BCD). In database design there are so many levels of abstraction between you and the hardware that
understanding the CPU architecture is of very limited utility.
They are also expensive in part because they are complex and contain precision components, so they might not get as cheap as you hope. But granting you that we get -80C freezers for the same cost as domestic fridges, we still have to provide wiring and power for them and the square footage for them to sit on (they are not small and you may need three or four of them per lab). Forget the hi-tech stuff, just consider paying for pipette tips, rubber gloves, and distilled water. Software development doesn't have equivalent consumables.
Do you have an open source model for the army of technicians you will need? I mean I can see a PI working on a project for the excitement, fun and prestige, but what about the research techs doing everything from washing dishes to performing autopsies on the lab mice. Unlike programmers working on open source software they won't be able to tele-commute and a part-time commitment would be problematic.
Yes, but none of them had to invest a couple hundred million to figure out how to make dougnuts. The issue is not recouping manufacturing costs, the issue is recovering R&D costs. The money can come from compulsory taxation (as with NIH/NSF), voluntary taxation (as with the Polio vaccine), or it can come from speculative investment. I still haven't seen anyone explain how open source is going to raise the money.
Limited production rights? isnt't that basically a patent?
I don't think it is just tradition. Calculus has the advantage of being relevent to many technical fields while discrete math has (so far) a more limited scope. After all, these are first year undergraduates we're talking about. Many of them will be end up switching to other majors. Having gotten calculus out of the way in the first year keeps a lot of options open for the second year.
Perhaps it is a chicken and egg problem, but I would think that most first year students would be baffled by the relevence of discrete math until they've written a few programs with non-trivial data structures. On the other hand, most of them will have a nodding acquaintance with area, velocity, and aceleration.
Both amplify a novel allele. In the case of breeding you have to wait for a mutation that produces the trait you want. In many cases the mutation occured a long time ago but just never spread widely in the population. However it was still a mutation, and a therefore a gene not found in the species until that point. Furthermore, genes do jump from species to species in nature (google "horizontal gene transfer").
I don't doubt that GM poses hazzards that deserve consideration. I don't understand though why people assume that the random processes of nature are benign.
Humans never needed penecillin before 1920, so why should they suddenly need it now?
There seems to be an undercurrent that novel genes introduced by a random process are "good" while novel genes introduced by design are "evil". While I can understand the mistrust of design, I don't understand why people assume a random process is begign.
I think you overestimate the value of the command line interface for the "long run". Command lines are efficient for tasks that are done frequently. However, computer users at all levels are faced with chores that come up infrequently but require a lot of technical knowledge to manage in their bare form. Good GUIs and menu systems are valuable because they cue the user to possible actions, and prompt the user with information they need to make decisions. When a casual user sits in front of a command prompt there are no cues, and they have to depend entirely on their training and memory. If it's done right, "eye candy" provides reminders of what it is possible to do and makes suggestions as to what should be done.
Eric Raymond's rant about UI was triggered by a GUI that was not good. It required him to understand the contents of the samba.conf file just as thoroughly as a sys admin editing it with emacs. He was pissed off because he had to spend hours learning arcane minutiae that won't leverage to any other area of his technical life, and will be forgotten if not obsolete by the next time he finds himself installing a network printer. An administrator who is setting up printers every week probably won't touch a GUI no matter how good, but a good GUI (or menu system) would have given Mr. Raymond several hours of his life back.
I'll chime in to support the parent. Despite the tremendous strides it has made in a very short time, the Linux desktop is still not suitable for general consumption. Did you read Eric Raymond's rant on the Linux user experience? Gnome and KDE are worthy efforts which I use daily, but as Mr. Raymond points out there is more to creating a productive end user interface then poping up dialog boxes that allow you to type in the same obscure strings that you would otherwise have typed into /etc/foo.conf using emacs.
I'd abandoned Apple back in the OS 8 days because I'd though the OS was languishing, but I just bought an iBook P4, and with OS X Apple really has made a Unix desktop that really is usable by the legions of Aunt Tillies.
In practice, the average CSci B.S. is going to end up writing database front ends for a business somewhere, so you could argue that all of these topics are academic and irrelevant. I just think that the first two years of college are too soon to drop out from so many areas of human knowledge.
Those sound like perfectly reasonable problems. I think the time constraint might be a bit rigid giving that the problems are coming at them with no warning in a stressful environment. Are you standing next to them with a stopwatch counting out the time in 10 second intervals?
How big is your sample size? My experience as a developer tells me that there is a wide distribution of programming skill in the world, so if you have a decent sample size and your test says that everyone is an idiot, then either something is wrong with your test, or something is wrong with your recruiting.
I've suffered the same disappointment, but this is one area where I think the employer is justified in imposing very strong filters. I believe there is an imposing body of FDA regulations governing the development of software for medical devices and it makes sense to try to find programmers who have already worked in that framework. Similarly for avionics systems.
Why stop at assembly language? Should all programmers be able to write microcode? How deeply should they understand logic gate arrays? It would be wonderful if we could all understand everything, but life is fleeting and there are guns versus butter choices to be made in our educations.
But why is knowing your way around the windows registry more fundamental to life as a free citizen then understanding any of the other myriad technologies we depend on every day? How deep is your understanding of AC power systems? Ceramics? Metallurgy? Polymerase chain reaction DNA amplification? Diesel engines? Domestic plumbing? Carpentry? Each of us stands on top of vast mountains of diverse technology which we have no technical expertise in.
I can understand complaints about the general level of technical education. I don't understand why so many computer technologists insist that the particular technology they are expert in is the technology every good citizen must learn.
But I don't want people to become better end users! I want them to become better at whatever it is they actually do for a living. It is my job as a programmer/sysadmin to try to advance their goals using my years of specialized study in computer technology. It is called the division of labor, and given the depth and sophistication of our technologies it is the only way to go.
I had to have a CAT scan today. CAT is an incredible blending of multiple advanced technologies: math, physics, electronics, computers and medicine. It's run by a technician who knows the gory details and gotcha's of its operation, and the results are interpreted by a Radiologist who usually doesn't even touch the machine any more. I don't want my Radiologist spending his or her professional time studying processor architectures or HTML. I want them studying their radiology journals and sharpening their diagnostic skills.
Things are never that simple in biology. Immunity depends (in part) on keeping around at least a few members of the class of white blood cells that recognize the disease organism. If you rarely encounter the target bug that class of white blood cells will dwindle away, and given enough time and bad luck may die out entirely leaving you vulnerable to another attack. On top of that, the virus may change subtley over time so that the white blood cells that recognized last years virus doesn't recognize this year's virus.