4 ms·
Ask HN: Which are the best computer science MOOCs?
I'm finding the current MOOCs sites to became very bloated due to the high number of courses. Now it is difficult to find the good ones. Which are your favorites?
- softwarebeware 5y agoThe school of life. Seriously. Build stuff and play around. And play around some more. :)
- DougMellon 5y agohttps://java-programming.mooc.fi/ https://java-programming.mooc.fi/ (part 1) https://java-programming.mooc.fi/ https://java-programming.mooc.fi/ (part 2) https://fullstackopen.com/en/ https://fullstackopen.com/en/ (fullstack)
- rg111 5y agocomputer science is an extremely broad area. Narrow it down, maybe? 1. Introduction to Computational Thinking at MIT was a great one. 2. Robert Ghrist has the best set of SV Calculus courses on Coursera. 3. Gilbert Strang's LA courses are the best. 4. For Algorithms course, I would recommend Sedgewick and Wayne's course on Coursera. 5. For PL course, go with the WashU course on Coursera. 6. For Machine Learning, you should start with Andrew Ng's Machine Learning. 7. Cryptography I, II by Dan Boneh is a great one. 8. Introduction to Programming with Python at MIT on edX is a good course. And so is CS50 by David J. Malan. 9. The nand2tetris course on Coursera. These are some that I can vouch for. You did not mention your level or what topic you want to learn. The best way is to do a google search for moocs, and try several of them out, and pick one with the teacher whose teaching style you like the most. In MOOCs that are not mandated, the personality of the teacher matters a lot. In my opinion- it is the most important factor. You should also look up courses where all materials are publicly available- tests, assignments, homework problems, etc. If some parts are not, you can always email a prof, and they will almost never turn you down. Follow discussions on Reddit. Engage in discussions, too. You should look up https://teachyourselfcs.com https://teachyourselfcs.com and check out the recorded courses there. Do the ones that you would like.
- strikelaserclaw 5y agostrang, sedgewick and nand2tetris are amazing courses.
- rg111 5y agoYes, the Sedgewick course is filled with wisdom. Even if someone has done an algo course before, or someone do not want to do the exercises, the videos are very worth watching- even like a documentary or movie. One will grow after watching those videos. And Strang is amazing. He does so much with a chalk and blackboard that many with advanced graphics and stuff fail to do. Nand2tetris books is also great.
- L0in 5y agoIt Sedgwick's algo course appropriate as an intro to algorithms and data structures for a CS student with a weak math background?
- rg111 5y agoThey would not get out the most from the course, but I will say, even then it is worth it. They will come out much better from the other side. And the course is not that mathematical.
- rococode 5y agoIf you don't mind missing the "M" part of MOOCs, you can learn a lot from university courses. Most of the top CS schools have slides and homeworks on their course websites. For example, if you want to learn... Artificial Intelligence (Stanford): https://stanford-cs221.github.io/autumn2021/ https://stanford-cs221.github.io/autumn2021/ Programming Languages (UW): https://sites.google.com/cs.washington.edu/cse341spring2021/lecture?authuser=0 https://sites.google.com/cs.washington.edu/cse341spring2021/... Distributed Systems (MIT): http://nil.csail.mit.edu/6.824/2021/schedule.html http://nil.csail.mit.edu/6.824/2021/schedule.html These courses have all been through the test of time and are specifically designed to provide an in-depth education with material written by some of the best and brightest educators in the field. You can simply follow along with the material at your own pace. Occasionally you'll come across something that you can't do (e.g. part of the assignment requires running tests on the school's private cluster), but most of the homeworks you can implement on your own. The only downside is there's nowhere to ask questions if you really get stuck (don't go contacting the course staff...), but with unlimited time you can usually figure it out eventually.
- karmakurtisaani 5y ago> The only downside is there's nowhere to ask questions if you really get stuck Well, there's always stackexchange. Finding open university courses can be a bit tricky. For that, MIT open courseware is amazing. After self-studying a course there, I have to say the depth is so much better than any MOOC I've ever done. No lecture videos though, but for me that's not a problem.
- ibobev 5y agoIt seems that Stanford's "Artificial Intelligence" course requires a login to be able to view the course's materials.
- rococode 5y agoThe videos require a login, but if you go to the modules page there are slides available for nearly everything (all the pdf/html links) https://stanford-cs221.github.io/autumn2021/modules/ https://stanford-cs221.github.io/autumn2021/modules/
- ArtWomb 5y agoYou're right, they have become unusable! https://cs50.harvard.edu/games/2018/ https://cs50.harvard.edu/games/2018/ http://ai.berkeley.edu/project_overview.html http://ai.berkeley.edu/project_overview.html
- karimf 5y agoI’m having good time using this reference: https://teachyourselfcs.com/ https://teachyourselfcs.com/