There is a lot to say here but since it is so incredibly depressing, I am just going to write out five thoughts and then be done with it. I won't go into the details of the grading scheme itself; instead, I will try to stick to more general comments.
1. People are very emotional about grading. Even though the course is long over, the discussion boards have suddenly become active again. Earlier in the course, Coursera really could do no wrong; any criticism of the course was generally frowned upon by the majority of students participating in the forums. Now, though, things have changed. Grades are different. Even students who are not likely to be critical of other aspects of the course design are ready to stand up and speak out if they feel they have been wrongly graded. A sad reflection on education in general, where people are more focused on how they are being graded as opposed to the learning process itself.
2. Grading is a contract that must be respected. I am staggered by the idea that anybody at Coursera thought they could just throw out the professor's grading scheme and replace it with a different scheme. What does that tell us about Coursera? Nothing good, in my opinion. Even if they decided that this professor's grading scheme was not sustainable over the long term, they should have asked him to change the scheme for the next offering of the class, rather than retroactively changing the scheme for the class that has already taken place.
3. But … grading really has no place in a MOOC. Seriously, what is the point of all this? Bitterness and acrimony about the grading scheme was a drag on the class week after week, and now it is adding a new dimension of dissatisfaction to the experience… for no good reason whatsoever. Peer feedback is great, because it is sometimes useful - and if it is not useful, you can just ignore it. Grading, however, is different - grading distracts from the emphasis on giving good feedback and it is also very hard to ignore. Yet the grading is also utterly pointless, given that this course is not for credit. The grade literally does not matter.
4. Alternatives to grading. At a minimum, students should have been allowed to opt out of the grading process at the beginning of the class, receiving feedback but no grades and likewise giving feedback but no grades. That would have been a fine option for me and for many others I am sure! There could be a certification of completion based purely on participation as a combination of writing essays, giving feedback, and participating at the discussion board. I would argue that such a participation-based system would actually have been just as effective as any formal grading scheme, given the chaotic nature of peer-based grading to begin with.
5. Poor communication. Throughout the class, Coursera has had an incredibly poor communication strategy overall, and this grading debacle has shown their communication at its worst. I have received two emails from Coursera about the grading debacle, although comments at the discussion board indicate that there are other Coursera emails that I never received (an email about the availability of the end-of-course survey? never got it; an email from the professor about the grading debacle? never got it). The sheer chaos of the discussion boards makes it a very poor vehicle for class-wide communication. The staff postings to the discussion board have been limited to "no comments," so it is not possible to ask questions about what the staff has posted there. Meanwhile, there is still no announcement of any kind on the announcements page of the class; the last announcement is six weeks old and for the entire ten-week course there were a grand total of four announcements. Many of the problems in this class could have been overcome with better communication, but the communication strategy appears to be non-existent, just ad hoc and ad libitum.
I learned a lot from this course - but mostly what I learned are things NEVER to do when creating a massive open online course. And my already bad attitude about grades and grading has definitely not improved, that's for sure. :-)
