The authors start by quoting an anti-MOOCer: "If I had my wish, I would wave a wand and make MOOCs disappear." There are many fantastic MOOCs that are on par w

The authors start by quoting an anti-MOOCer: "If I had my wish, I would wave a wand and make MOOCs disappear." There are many fantastic MOOCs that are on par with or exceed the best college classes I took at a top ranked school. Yet article after article from certain sectors proclaims MOOCs a failure and asserts or suggests students are getting ripped off or cheated of a proper education, doomed to failure, should be erased from history, etc. Since these people clearly do not know what they are talking about, yet clearly choose to intentionally misrepresent things to promote various agendas, they are unreliable sources and their opinions about everything else are unreliable as well. However, in this case, this article is written by Compeau and Pevzner. I have taken their Bioinformatics sequence, which I found to be among the better MOOCs around. It is very challenging and the interactive textbook which compels the student to write code in any language they fancy and solve problems with that code works quite well. Therefore these authors do know what they are talking about in this case. Whether this approach can or should be extended to history or art classes is not clear. Textbooks are expensive and can be difficult to access for students outside the western world. A lack of texts has been an issue in many MOOCs and has limited the depth of some classes. This has been addressed in many well done science classes, such as Melbourne's remarkable Epigenetic Control of Gene Expression by selecting open access academic journal papers and assigning them to students to read, and requiring the articles be understood to pass the assessments. Some other classes have arranged with publishers to make copyright restricted materials available for a limited time period to students enrolled in the class. These approaches, are quite good as well compared to a textbook, perhaps even better as they can cover cutting edge results for which textbooks have not even been written. Yet another approach is to recommend textbooks as a supplement to the class material. I have found this useful for some classes, especially when I am able to get previous editions used for close to the cost of shipping. Another approach is when the professor self publishes a print on-demand text for the class, and makes it optional, with the text also available for free. In these cases these texts have often shot to the top of the sales rankings in their category. It's very true that interactive texts that give immediate feedback are a valuable pedagogical tool which when implemented well approaches the value of working with a tutor. It's also valuable to have well designed forums where students can discuss problems they are having with other students who have solved the same problems. Compeau and Pevzner's texts have comments per page and don't ban discussion of techniques used to solve problems, or tricky bits that were not obvious. These discussions are very helpful for non-trivial problems that take hours to solve when one gets stuck. Students need to refrain from posting actual code of course and this has been the case. These context tied mini discussions are comparable to having a very good study group or TA session available in a university class. General and sectional discussion forums not tied to specific pages, are also needed as well. On Coursera in particular this is well designed and there are often fantastic contributions to new research posted by knowledgable students, as well as the discussions by students that one would experience at a brick and mortar school, but with a transcript and references built in, which makes them in some ways even more valuable than face to face conversations.