Hacker News new | comments | show | ask | jobs | submit login

Big O notation is an expression of a theoretical concept that applies to the algorithm itself. It is independent of any hardware that might be used to implement an algorithm.

Specifically, asymptotic complexity is an expression of the number of abstract operations or units of memory an algorithm will require. The meaning of "operation" or "unit of memory" depends on the algorithm. In the case of comparison sorting algorithms, for example, the operation in question is the number of comparisons.

Even if the CPU time allotted to a running algorithm changes due to system load or overheating, the algorithm still performs the same number of operations. Actual running time under system load can be calculated, but this is unrelated to big-O notation or complexity.


Regarding your explanation, it is excessively wordy and uses awkward phrasing. Your subsequent complaint about the terminology used in responses to your previous questions about P=NP demonstrates a serious lack of prerequisite knowledge. You learn the definition of "polynomial" in algebra; teaching you algebra is far beyond the scope of a discussion on HN.

As someone who is primarily self taught but also finished most of a college CS education, I can understand your frustration. In order to understand one thing you need to learn three other things, and there's no map that leads you from what you know now to what you want to know. Until someone makes that map, or you know enough to make up that map as you read Wikipedia, you'll have to learn things in the order they're taught in school.

So here's what you should do if you want to understand what people are talking about on HN:

1. (Re)learn basic algebra from start to finish. Use Khan Academy. You absolutely need this to understand what a polynomial is.

2. Accept Wikipedia as a source of information. Use a dictionary and other Wikipedia articles to look up terms you don't understand.

3. Related to #2, try to study cognitive biases, critical thinking, and logical fallacies. Studying these concepts will help your brain process information, such as difficult to understand Wikipedia articles. Check out lesswrong.com.

4. Study basic CS. Look for an Introduction to Computer Science course on Coursera or elsewhere.

5. Study algorithms. Take an algorithms course on Coursera or elsewhere.

I studied computer science at the University of Missouri Rolla. My instructor Dr. Pyron said that this stuff wasn't needed to learn Computer Science and that most of it was hoaky and kind of hinky. But he also told our Freshman class that "Ethics don't matter" as well. So I think I was robbed of that opportunity to learn it. BTW I feel that "Ethics do matter" and told him of my opinion, and he said it wouldn't work in the computer business.

I had a TA for College Algebra who couldn't teach it, and the professor was nowhere to be found. They made Catch-22 Jokes that he was like Major Major you could only see him if he wasn't in his office. It made me feel better but did not help me learn. I was in the Delta Tau Delta fraternity and they helped me out, but claimed the TA had messed up some of the problems that could only be solved with Calculus which I didn't learn yet. To make matters worse I got caught up in the party college thing and underage drinking. I quit eventually. But at least I didn't get forced to meet "Alice" like some unfortunate souls. I still have my Intro to Pascal and Pascal with Data Structures books in my basement, I should re-read them and try stuff with Free Pascal.

In 2005 I had a car accident and almost died, was in a coma, and lost some memories as a result. I may just have to relearn everything all over again. Since I am on disability, free online classes and sources are my only hope. Thanks.


Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact