Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: Is computer science really a science?

news.ycombinator.com

Post reply on HN
Print view View original

1–10 of 22 posts

Ask HN: Is computer science really a science?

#1

Post by stevejobs »

How did it come to be that the field of computing became synonymous with science to form the term computer science? As I understand it science is about figuring out and interpreting the natural world as it is and has evolved over time (biology/physics/chemistry). Laws and theories have been formed to understand and clarify how objects are created and interact with each other.

Computer science OTOH is more akin to math and engineering than science. While some parts of computer science deal with the natural world like the work on artificial intelligence (which is more about psychology/philosophy than science) most of it deals with programming and electronics than the study of nature.

That is not to say that computing does not have value. It certainly does. I just think it is questionable if it deserves the science moniker.

Re: Ask HN: Is computer science really a science?

#2

Post by frossie »

The problem is that only a few decades ago, computer science was a fairly pure branch of mathematics - now you can argue whether mathematics is a science or not in general, but applied mathematics is nearly indistinguishable from physics in many areas, and computer science certainly flirted with applied mathematics.

What gets taught as computer science has evolved rapidly in the last couple of decades and it is my impression that quite a bit of it comes under software engineering now.

This is why I get annoyed when you see jobs in the software sector or software management that require a Comp Sci degree. It's actually a subtly ageist requirement - twenty years ago, computer science was not what you did if you were interested in computers - you did physics so they would let you play with their VAX cluster.

Re: Ask HN: Is computer science really a science?

#4

Post by bendmorris »

Yes. But the word "science" is very broad.

Sciences are generally classified by "hardness" or (perceived) quantitative rigor. In that respect, from "hardest" to "softest," the sciences are usually ranked roughly like this: Math (yes, math is as much a science as anything else), Physics, Chemistry, Biology, Psychology, and the Social Sciences.

Computer Science is, as most things are nowadays, interdisciplinary. As such, it really depends about what aspect of Computer Science you're referring to. Some areas of computer science are more rigorous, and have a lot in common with math and physics. Others, such as "software engineering" etc. fall on the softer end of the spectrum.

With that said, I don't think that all programmers are necessarily "computer scientists." They're practitioners of an art. Just like not all doctors are medical scientists - some are, but most are just practitioners.

Re: Ask HN: Is computer science really a science?

#5

Post by idoh »

It depends on how you define science. Personally I define science as a process, whereby one has theories and then tries to systematically falsify them. By this measure CS isn't really much of a science, it is more like Math or Engineering.

But it is not exactly a black or white issue. It is more of a science than, say, Political Science, in terms of rigor and logic.

Re: Ask HN: Is computer science really a science?

#6

Post by ablerman »

hottoh wrote: post #3

Computer science is the study of algorithms, the study of algorithms fits the scientific method. Programming and electronics is not computer science. QED.

Science is the study of the fundamentals. Algorithms is clearly a science as is compiler design and even electronics ( someone has to create new types of components, transistors, memristors, etc... )

Building things with those fundamentals is not science, it's engineering.

Re: Ask HN: Is computer science really a science?

#7

Post by Zev »

How do you think the parts of the computer work and are developed? Especially processors. There's a lot of chemical and material sciences behind the scenes.

I suppose this is also why "Computer Engineering" as a discipline is somewhat growing in popularity.

Re: Ask HN: Is computer science really a science?

#9

Post by hristov »

Computer science is definitely a science. Mathematics is a science and computer science is part of mathematics. Thus computer science is a study of computation or the automatic solution of problems or processing of data, etc.

The problem is that in most universities the computer science label is used for things that aren't computer science. Most computer science departments are mostly concerned not with studying or extending the science part of computer science, but with applying computer science to solve real world problems. This really should be called computer engineering or software engineering, just as applying certain aspects of physics to create electrical devices is called electrical engineering.

Re: Ask HN: Is computer science really a science?

#10

Post by thingie »

That's just a name. It is of course not a natural science, but a formal one, as mathematics (it could even be a branch of maths). There are even other kind of sciences, social ones, like psychology (which many people wouldn't call a science, though for different reasons, after all it does study natural things, doesn't it?). And of course, there should be a distinction from engineering (electrical e., software e.), though boundaries aren't clear and never will be. If you had a better name for those different kind of sciences, call them in that manner. Also remember, that by saying that something is not a science, it might seem that you are implying that the branch is somehow inferior to the "actual science". (This is not the case, but it is going to be perceived as that.)
Post reply on HN