The course content is good, some of the ideas may not be directly useful to many CS students at first, but the core concepts are actually very important to all software engineers. Unless you don't want to be a SWE, then this course is probably not useful to you. The core concepts are: machines should be designed to fit humans and consider humans' limitations, instead of the way around. You can read more about this from the book as part of the required readingssssss Design of Everyday Things. There, I just said the core concept, and you probably learned 99% of this course.
Now comes the ugly part, Joyner tried very hard to structure the course, which is not necessarily a bad thing, but the course is still a mess. Although Joyner claims he is a cognitive scientist and bad mouthed the processor model, he did not practice what he preach. For example, he said the lecture videos + assignments require 3~4 hours each week and actually calculated the required time estimated for each subtasks. Of course, it is all bullshit. He underestimates the required time needed for each tasks. He said the required readings takes 1 hour, when he gave you 1 book, 1 article, 2 papers.
It used to be the assignments are easy and can be completed in 1 to 2 pages. Say goodbye to that because it is more like 6 to 8 pages nowadays. Anything less than 6 pages are probably incomplete, but not impossible. If you want to get A, however, 6+ is more realistic. The peer feedback system is bad, like, BAD. Joyner, in his paper, said that OMSCS students overwhelmingly hates peer feedback, yet he still doesn't remove the system after years. He preaches good HCI using conventional behavior, but a lot at the latest Canvas design shows this is not the case at all. But I give him the benefit of doubt because it is the first semester using the new system Canvas.
The exams consist of 30 questions. The first 20 are from lecture videos, easy. The last 10 are from required readings. It is more like 1 question from each reading. Extreme waste of time. The posted examination statistics stated that the correlation coefficient between homework average and exam 1 is about 0.4231, which the highest score obtained was 146 (2 people), which the mean is 123.91 (out of 150). Standard distribution is 11.45.
What's the verdict? Although this post it seems that I hate Joyner, actually no, he is still an amazing instructor, it is just that there are some rough edges readers should beware. I would not recommend this course. (you should just read the book and take another OMSCS course)
Now, if after all the ugly parts I wrote and you are still interested, here's the good parts. Joyner is rated the best instructor for many reasons, he constantly look for feedback from users and improve on his designs, a thing he preaches in HCI (though I prefer the first version to be bit better than what it is currently). Also, the course is easy to get an A. It consists of busy works, which pumps this up to "medium" level. But it is easy to get an A.
30% comes from 10 assignments. 30% comes from 2 projects (of which one is group project), 30% comes from 2 tests and 10 from participation points.
The polls show that most get about 19~20 from assignments, and full 10 from participation points (giving peer feedback, posting on Piazza). So you will only need to aim for 85+ for projects and exams. Let's say you receive 19.5 for all assignments, full 10 participation points, 90 points from projects (a very conservative estimate), getting the mean score of 123/150 for both exams are still enough to land you an A (90+).
95x0.3 + 10 + 90x0.3 + 123/150x100x0.3 = 90.1
Joyner if you are reading this, here are some feedbacks:
- Please remove peer feedback system.
- Please kill the research based required readings.
- The assignments load are okay for 1 course per semester, but might be heavier than expected for those taking two courses at the same time.