Slashdot Mirror
All Science is Computer Science [Y/N]?
angainor sent in this interesting piece: "There is an article in NY Times which claims that in fact all science is computer science. He does some small talk about the fields of modern science where computers have been successfully used. But that's it. Does he really know what he is talking about? Read this piece, but don't be proud just because you are too a computer "scientist"." The writer has a good point about new advances in many fields being due to large amounts of computing power being applied.
I don't think that's being discounted, but simply using mathematics doesn't make one a scientist. A computer programmer uses mathematics in his programming in much the same way that a civil engineer uses mathematics in designing a bridge. Neither of them are really "scientists," but rather engineers who apply scientific knowledge to particular problems.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Most people who get CS degrees are the farthest thing from being actual computer "scientists". Real computer science is basically mathematics - whether it is finite automata or database normalization, it boils down to math.
On the other hand, computer programming, which is really what the vast majority of CS people do, is the farthest thing from science. If it were done with discipline and planning, you might be able to call it engineering, but really when you look at the way software is actually created, it can't even be called that.
So let's not flatter ourselves. The fact that you use computers as a tool in true scientific research, or you program computers to do specific tasks, in no way makes you a computer scientist.
Math is a useful *tool* for science, but science and math are in fact entirely different. Math is *true*
All science, sooner or later, is *false*. That's because theorems can be proven, but hypotheses can only be disproven
And I, being a bioinformatician, am one of the sorts of people the article is talking about! I did my doctorate in a microbiology department (doing bioinformatics really) and did a postdoctoral stint in a computer science department, so I can compare the two fields.
While I think that working in a computer science department gave me an interesting perspective on problem solving, the fact is that computer science really doesn't deal with making actual programs that do things, but with more esoteric things like proving problems to be NP-hard. The sorts of applied knowledge that is useful to other fields isn't really central to the aims of computer science as such. This isn't a slam on computer science -- you can make a similar claim between the basic science of microbiology and the applied knowledge useful for treating infections.
Perhaps for some theoretical non-human scientist that has a consciousness structured so that all facts are always explicit, who can simultaneously sustain conscious awareness of every level of analysis, and then communicate this to their peers, the idea behind this statement might be true. But insofar as human science refers to knowledge possessed by humans, this statement is incorrect.
Software engineering isn't Engineering either.
Uh. All science is not computer science. The article essentially says that a lot of scientific fields rely on massive computations performed by computers. Massive computations != computer science. CS has a lot more in common with logic and math, and formal theories of computation. Fundamentally, it has nothing to do with vanilla number crunching. Following the lead of this article, one might as well say "all science is engineering", because nearly all the sciences rely on well-engineered pieces of equipment to test their theories, gather their data, and a myriad of other things. I think a better way of phrasing it would be that 'computers have become a fundamental part of modern science (and modern life for most of the western world)' -Laxitive
God forbid you actually go into a field you enjoy rather than one in which you hope to become famous.
Besides, last I checked Larry Ellison didn't know much about computer science or genetic engineering.
The real problem with computers is now that they are so dang popular the real advances don't get the press that a new release of Linux 2.4 or a new Athlon does. People aren't interested in trying out new language paradigms because, gall darnit, if C was good enough for K&R it's good enough for them. People aren't interested in trying out new kinds of software because they're comfortable with the old kind.
It's kinda like saying there hasn't been any development in automotive technology when what you really mean is that the cool developments take decades to actually be implemented (if they ever make it) so you never hear about.
Pie in the sky fields like genetic engineering can fill their press releases with things like "some day we may be able to use this technology to cure congenital birth defects". More established industries like computers that already have shipping products have to be slightly more...pragmatic.
Has genetic engineering really made big advances in the past few years? Or is that just the spin that biotech companies have put on it? Or is it our own biases based on our ingrained awe for biology and contempt for mere machines? Even if it has made great leaps in the past, is genetic engineering likely to do so in the next few years as well? How can you even begin to quantify what counts as a "big advance" except through hind sight? I think it was Yogi Berra who said, "It's hard to predict things. Especially the future."
Computers have become an incredible and indispensable tool in the advancement of all the sciences, but that doesn't make "all science computer science". One could just as easily say that "all science is quantum physics" or "all science is math" and it would have the same degree of truth, i.e. some but not enough to be considered a generally true statement.
That may be true (although I'd like to see you prove it), but would it be meaningful or useful? As another reply has pointed out, you're expressing a standard reductionist position, but it's one that isn't even held by most good physicists. For example, here's a quote from quantum physics professor Howard Georgi of Harvard (taken from here):
Another good intro to these kinds of issues is Murray Gell-Mann's book, "The Quark and the Jaguar". Gell-Mann's credentials as a quantum physicist are beyond reproach, but he is by no means a reductionist, and has a keen appreciation for the unique properties of complex systems - the jaguar in the title of his book is a metaphor for this.
Since many other physicists and philosophers more qualified than I have written on this topic, I'll restrict my reponse to a freewheeling, extended analogy: quantum physics can be compared to a CPU's instruction set, or "machine code". On top of that, we layer assemblers, and then compilers and interpreters for various languages. Using compilers and interpreters, we build various systems and applications. Since ultimately, all of these things are done using machine code, is it meaningful to say that all applications are "just machine code"? There's a sense in which this is true, but let's examine it further.
With the CPU analogy, we can do something we can't do in our single physical universe: we can take an application and compile it on a different type of CPU - a CPU with a different instruction set. Compiled for this CPU, the application still behaves identically. So by claiming that an application is machine code, we're clearly missing an important point, since the same application functionality can be achieved with completely different machine code. [Of course, both CPUs follow a more fundamental set of information theory laws, but that's not important to the argument.] The point is that complex systems exhibit "emergent properties", characteristics which arise from interactions between components of the system in question, and which can't be meaningfully analyzed, or even easily inferred, from the perspective of more basic, underlying systems.
To cut this short (well, shorter than it would be otherwise), I'm going to make a few leaps. Imagine for a moment that we could build a toy universe in the laboratory, with different physical laws than our own. Even though its physical laws are different, it's not impossible - in fact it's quite likely - that complex systems in that universe could share some of the properties of complex systems in our universe. To take an extreme example (as I said, I'm leaping), imagine an intelligence in this other universe, and assume we could communicate with it somehow. We would probably find that we share some basic characteristics with this alien intelligence. For example, it is a common characteristic of living systems that they have a strong bias toward survival, simply because those that don't, die out. This survival instinct is something that's not a direct or obvious consequence of quantum mechanics - it's actually rooted in simple logic (perhaps all science is logic?!)
Even if you could somehow come up with a QM model for the survival instinct, it would miss the point, since it's quite conceivable that a survival instinct could arise in a universe not based on QM - it really has nothing to do with QM. The survival instinct is just one example of an emergent property of complex systems - in this case a living system - that has little or nothing to do with the physical construction of the system.
AP, UPI (NY)
In a stunning announcement today, the American Automobile Association stated "All science is actually automotive engineering." A recent study has shown that over 98% of scientific developments required the use of a car. "Without these amazing machines, I'd have to walk 45 miles to work, up hill both ways!" one scientist was quoted as saying.
In a related event, the Carpenters Union announced "All science is actually construction." Nearly 100% of lab experiments take place in buildings that were built by builders. Without us, there'd be no labs, no checking of theories - in short, no scientific advancement." Theoretical mathematicians scoffed at the announcement, but other scientists confirmed that most labs are not in caves or other natural structures.
Meanwhile, representatives of the National Restaurant Association are preparing press releases to explain that all science is actually eating. 100% of scientists contacted by this reporter confirmed that they would be unable to do science if it weren't for food.
Most programmers use (formally or informally) the scientific process to debug software defects. They witness unexpected behaviour, create a reproducible test, postulate a cause, compare the new test results with the expected/control results. Are programmers now scientists when debugging, but not when writing new code?
cpeterso
The naive ignorance and general gullibility of the public never ceases to amaze me. Claiming that everyone who knows how to drive requires is also a mechanic does not compute. Very broadly speaking, there are 3 branches of mathematics which feed into computers
- statistical = accounting = infosys
- discrete = binary/automata = computer science
- continuous = scientific fields = computational science
Basically computers have matured to the stage where nowadays CSEE are nothing more than software engineering techniques, but the level varies according to the stage of hardware->firmware->software->wetware. (as Intel? CEO once said, hardware is nothing but frozen software). Computers are useful because they act as mental accelerators allowing you to do stuff overnight or in between coffee breaks or QUAKE sessions. But by itself, the theory is rooted in various branches of maths split into the business of computing (variations of the accounting equation), art of computing (Knuth/algorithms/etc) and the science of computing (complex systems/quantum effects/etc). For some strange reason fun and money seem to have an inverse relationship along this continuum.
For the average layperson who barely recognises how to access the internet (gee-whiz, moving text) the distinctions are superfluous but it doesn't help when the media confuses mathematics with their applications.
LL
Well, my subject about says it all. Just because most scientists can and do use cars and pencils, we don't refer to them as racing drivers or pencil-operators.
"All scientists are programmers" would have been a truer headline, as would "All programmers are not computer scientists".
"If I were 21 years old," he said at a company conference in New Orleans, "I probably wouldn't go into computing. The computing industry is about to become boring. I'd go into genetic engineering."
:-)
This rings true to me. Much of Comp Sci (my chosen profession, though I suck at it) seems to have a lack of discovery and / or innovation these days, with the exception of nanocomputing. Much of the rest of it is innovation, not invention / discovery. How many Turings do we have in Comp Sci now?
Computing / computer science is a skill rather than an industry. While I'm a programmer, and have worked for a computer company (Acorn), I've also worked for a medical company and a couple of communications companies. I'm sure I could get a programming job at a medical company doing genetic research if I set my mind to it.
Secondly, how many people working in genetics are making fundamental discoveries, and how many are just grunts doing their job? For that matter, how many people's jobs in *any* field allow them to do blue skies research of the type that may lead to fundamental discovery?
I've long ago realized I had to separate my intellectual interests from my job. While I've been lucky to have extremely interesting work assignments, it's at home that I become the "mad scientist".
You're overestimating the potential impact of quantum computers. Unless something's changed since I stopped paying attention, QCs are Turing-equivalent -- no more, possibly less. At best, they're nondeterminstic (or do a good impression), but that's hardly a breakthough for theory. Everything we know about algorithms, formal languages, and computation still applies. On the other hand, they're hard to build, fragile, and almost useless as general purpose computers (I haven't yet heard developed proposal for I/O, which is vital to just about everything most computers do).
All science is computer science is true in as far as it goes. The more accurate assertion would be to say that computer science is a new way of looking at information theory (which pre-dates even Babbage).
For example, high energy physics is the process of deducing, without knowing the underlying properties of the Universe, what behavior we will see when we "crank up the heat" of the universe. If we knew the underlying properties, however, math could tell us the rest. This math, it seems, is more complex than the pre-Dirac world had thought. It does, in fact, seem to involve some rudimetary logic. Hence, the study of the universe is the study of an information system with logic, math and vast "memory", which is not unlike Turing's paper tape.
Computer science is math, and math is the Universe. As computer science expands and more generically encompases all of mathematics, the lines get grayer. If it is fair to say that all science is math (and I think it is), it is getting increasingly more accurate to say that all science is computer science.
A computer is a tool. Its use, like its construction, is a technique. In the early days of computing it made sense to pull together multidisciplinary teams from mathematics, physics, philosophy and engineering together to make the things work in the first place. That's been done.
There are still interesting things to be done in Physics and Engineering which may, in the fullness of time, lead to better hardware, and there are still interesting things being done in Mathematics, Linguistics and Philosophy which will, in the fullness of time, lead to better software.
But there is fundamentally no such thing as Computer Science.
I'm old enough to remember when discussions on Slashdot were well informed.
-- Multics
I bet most the scientists drive cars (or use public transportation) between their place of residence and their place of work. In fact, many important discoveries would not have been possible if they had no way to transport themselves to the lab. Does this make them car designers or mechanics?
I didn't think so.
--
SecretAsianMan (54.5% Slashdot pure)
Washington, DC: It's like Hollywood for ugly people.
I'm no math wiz, but isn't CS basically all based on number theory, computability, and mathematics that were around before anybody actually assembled a physical computer (computers themselves were just thought experiments until they became a physical reality, right?).
And in turn, we just found that math is basically "full of holes".
It's 10 PM. Do you know if you're un-American?
Next you'll be telling us that, since art can be produced, manipulated, reproduced and analysed with the aid of computers, that all art is computer art.
Since speech can be produced, manipulated, reproduced and analysed with the aid of computers, that all speech is computer speech.
Since music can be produced, manipulated, reproduced and analysed with the aid of computers, that all music is computer music.
That is just ridiculous. Computers are simply our way of patching our brains to make up for the difficulty most of us have with performing sustained, repetitive calculations. Cept we haven't managed to 'open the source' to our brains and compile in the changes yet, which is why we're fucking around with these dynamically linked modules we call computers.
I gots ta ding a ding dang my dang a long ling long
Exactly correct. Computer Science is the science of computers not the science of fungi. Now had the author said that all sciences were math, he may have had a point (not too sure about this myself, but I throw it out there).
Someone you trust is one of us.
Utilizing computational power is NOT computer science. Designing processors is computer science (and computer engineering). Designing computationally efficient algorithms is computer science. Analyzing algorithm complexity is computer science. It's hilarious that so many people equate computer science to "using computers." It's like equating "hacking" and "cracking."
Computer science (and computer engineering) lays the foundations for other fields to effectively use computers. Where would physicists and biologists be if significant time and effort had never been invested in developing programming languages, communication protocols and designing processors?
Anyway, that NYT article is just plain silly. The Larry Ellison quote tops it off. I'll agree that much of the computing industry is boring, but computer science is an academic field and it'll be a l--o--n--g time before CS begins to get boring. Go ask good 'ol George Johnson what he thinks of Artificial Intelligence...
Jason
"This rings true to me. Much of Comp Sci (my chosen profession, though I suck at it) seems to have a lack of discovery and / or innovation these days, with the exception of nanocomputing. Much of the rest of it is innovation, not invention / discovery. How many Turings do we have in Comp Sci now? "
I disagree. Computer science is about to get much more interesting in the way you mention, when quantum computing starts getting taken seriously. The entire field of algorithms needs to be rewritten for quantum computers. The fields of cryptography, compiler design, languages, and even theory of computation need to be rewritten. NP-hard doesn't necessarily mean what it used to (it doesn't make a problem intractable with quantum computers). The whole heirarchy of decidability has to be looked at a litle differently.
-- "Those who cast the votes decide nothing. Those who count the votes decide everything." -Joseph Stalin
Most (if not all) sciences now use computers as tools, but that's no different from using calculators as tools, or calculus as a tool, or statistical analysis as a tool. That does not mean that all sciences are mathematics or engineering. Physicists now need to be able to write code and use computers in fairly sophisticated ways, but they do NOT need to be computer scientists. Computer scientists do NOT just write code; they're generally developing more theoretical stuff, such as the theory of computation, or artificial intelligence, or advanced operating system design. It would be like calling someone who uses physics on a daily basis (gee, pretty much everyone, though I had in mind someone like a radiologist) a physicist.
The difference is between using tools and theories (which does not make someone a scientist in that discipline, in this case computer science), and DEVELOPING those tools and theories, which is the job of scientists in various disciplines.
-- "Those who cast the votes decide nothing. Those who count the votes decide everything." -Joseph Stalin
Read the EFF's Fair Use FAQ
Computer Science is a mathematical field, not a scientific field. Not sure who coined the name, but it is definitely a bad one. Computer science has more to do with proof-theory, than any scientific occupation.
I do systems work for a math deptarment at a large university. One of our Professors is very involved in using computation and simulation to study various bio-medical phenomenon. Simulation is an important part of her work. By taking experimantal data, she can create simulation models that closely approximate the "real" world. Experimental data is needed to create these models, and then to validate results, but simulation is useful for running multiple senerios to see where to look for experimental research. I don't think that simulations will ever totally replace experimentation, but it will increasingly focus it, and reduce the amount of experimentation necesary to gather specific data.
I don't need a million points of light, just two points of multi-mode fiber and a 10 Gig-E router.
The role of computation in science is magnified beyond its usefulness because futzing with computers is fun, easy, and something to do when you're out of other ideas.
One example is global climate modeling. The predictions of these computer models are cited all the time, but no one really knows if they're putting out valid results or garbage. (Since these models can't predict the weather 10 days out, one must wonder about their century-term results.) That's not real science.
Another example, I'll bet, is the computational archaeology mentioned in the article. It is easy to imagine these guys assigning variables to a lot of inexactly quantifiable phenomenon, writing equations for things that are not precisely equatable, and plugging in estimates for unknowns. Garbage in, garbage out. That's futzing, not science.
(For that matter, has the Santa Fe Institute ever produced any useful science? As far as I can tell, they're a sensational press release factory.)
The Logical flaw is supposing that all computation is computer science.
There is the science of the problem you are trying to solve, and then there is the science of the tools you use to solve the problem. The two are not the same.
Solving the Human genome is different than programming the computer to analyse the data.
but there is an overlap. In the same way that it helps to have business and accounting experience to be a systems analyst in a business. Although alot of system analysts do not have this either.
"It is a greater offense to steal men's labor, than their clothes"
That's why, when considering a career in the sciences, you really need to weigh interest a lot more heavily than the job market. Right now I'm seeing a lot of people around me trying to jump into fields that are "hot" at the monent. But, like you said, what's hot and what's not will change before they're done with their schooling.
That's why I chose an area of study that might never be hot [probably because it's so complicated that it makes normal biologists' heads swim (signal transduction and biochemistry)], but happens to interest me greatly. True, I'll probably never make $100,000+ a year. But at least I'll always be happy doing what I do. And after living a student's lifestyle ($8,000 per year), even $30,000 per year will seem like the high life.
---
4-star general in a one-man army.
Now that genome projects are hot stuff, people have started to take a good look at the biotech industry. And if they think what's happening now is exciting, wait until they see what's in store.
I'm willing to bet that the protein folding problem will be solved in the next 50 years. Soon after, we should start to see protein design hitting its stride. What does this all mean?
Take a bunch of E. coli bacteria. Use the genomic info you already have to insert a new gene for ProteinX that you've designed. The bacteria then make ProteinX in a huge vat, churning out billions of copies of the protein you need within a few days.
Think nanomachines are hot stuff? ProteinX is only a few nanometers in diameter, has no conventional moving parts (just changes in conformation) and can be regulated just be adding different chemicals into the mix.
OR
Think spider silk is strong? ProteinX could be modified silk fibrin, designed for more elasticity and higher tensile strength.
The sciences will always come out with incredible discoveries. Companies that use these discoveries will always have stock that's worthwhile to own. Maybe in the short-run things might dip, but it'll always make a come back.
---
4-star general in a one-man army.
Wait, they were all "pencil and paper" scientists as well. Damn, were they well-educated or what?
And Archimedes was a "stick and dirt" scientist, right?
Give me a break. The computer is a tool. A very powerful tool, in fact indispensible now, but a tool nontheless. I'm an Electrical Engineering researcher, and I spend a lot of time writing computer programs for my research in a variety of languages. Please don't call me a Computer Scientist, though, or I might just throw up. I use oscilloscopes a lot too. Does that make me an "oscilloscope scientist"?
Computer science is a well developed discipline in which very smart people devise new ways to solve problems. People in other fields, like me, use what computer scientists come up with. We are not computer scientists in our own right.
Last month a leader in the software industry, Larry Ellison, the chief executive of Oracle, predicted that the focus of the intellectual excitement will shift again.
"If I were 21 years old," he said at a company conference in New Orleans, "I probably wouldn't go into computing. The computing industry is about to become boring. I'd go into genetic engineering."
This rings true to me. Much of Comp Sci (my chosen profession, though I suck at it) seems to have a lack of discovery and / or innovation these days, with the exception of nanocomputing. Much of the rest of it is innovation, not invention / discovery. How many Turings do we have in Comp Sci now?
Now genetics, this stuff is freaking AMAZING. My girlfriend is going into it, and I'm regularly amazed by the discoveries that are being made in the field. It may well be that computer science is no longer the frontier of human knowledge; I don't know.
The article is, of course, dead wrong. Mr. Johnson needs to have his head examined if he thinks that just because computers are used as tools in many professions, that thereby all professionals are computer scientists. He wrote an article for the NYTimes online, probably using a word processor - thus by his definition he can claim to be a computer scientist.
The thing he's dimly perceiving, but failing to adequately put into words, is how computers have become ubiquitous in the professional and academic world, and how a working knowledge of how to USE computers is fast becoming utterly essential. However, he fails to see the vast difference between being a competent end user, and being a discoverer, an inventor, a creator-of-new-things in the computer world.
So all in all, the article is only interesting in that the author accidentally brings up something else that's worth thinking about: computers and their involvement in genetics research. Now what I want to see is more development in the field of biological computing... the day when genetics and microbiology combine with comp sci and nanotechnology / nanorobotics, will be a portentuous day.
-Kasreyn
Kasreyn: Cheerfully playing the part of Devil's Advocate to hairtrigger
I remember a different time, not so long ago, when all science was Slide Rule Science. And my grandaddy told me of a day when all science was Abacus Science.
-----------
-----------
If you ever drop your keys into a river of molten lava, forget 'em, because man, they're gone. -- Jack
that must mean when I was 8 and I got one of those home science kits for my birthday that I was actually doing computer science. I gotta update my resume for that extra 5 years of CS experience I have now. Big raise here I come!
The global economy is a great thing until you feel it locally.
The Aztecs had an advanced number system, and developed many of the algabraic formulas we use today. They were also amazing astronomers who made many wonderous discoveries about the heavens. All without the help of a computer.
Slashdot: Open Source, Closed Minds.