There seems to be a theme of duplicate, potentially competing efforts coming from the same universities in the now emerging online course space. First with Coursera vs Udacity coming out of Stanford. Now with MITx vs RELATE coming out of MIT. Wouldn't it make more sense for those within the same university to collaborate in order to produce a consistent brand and user experience? Don't get me wrong, I'm loving the fact that these types of educational systems are beginning to take off. I'll be enrolling in as many as time permits. But individual course content aside, it seems like each is reinventing each other's wheel.
When movie technology first came out, movie makers used it to film plays. I'm glad they didn't take the approach that they should find the best stage performance of each play, film each one, and turn the collection into the standard movie library, and stop there. Fortunately, they soon discovered that movies didn't have to be limited by the constraints on plays.
So many aspects of college-level courses are bound to the institutional needs of colleges. Courses must be taught to several people simultaneously and in sync, must be delivered in semester/quarter/term-long chunks, must be popular enough to support ongoing employment of local faculty & staff, must be taught in a way that allows for objective grading (so, for example, emphasize grammar rules over pronunciation in foreign language, regardless of actual relative importance, because grammar is easier to score "objectively"),and so on.
Right now, these online courses are still basically movies filming plays. These classroom "plays" are the way they are, because that serves the needs of the institutions producing them.
But the most effective way to learn something personally useful probably isn't going to coincide with maximizing an old institution's needs for prestige, profits, politics, and power. I'd like to see these monolithic "degree credentials" broken down to finer-grained "skill credentials" that are available a la carte from competing sources that optimize their approach differently for each specific skill and for learners with different circumstances, with the learning results validated by independent agencies that only care how much you know, not how you learned it, where you learned it, or how long it took.
The last thing I want is a continuation of the "consistent brand and user experience" of today's universities.
Give them some time to diverge and innovate. They're implementing the obvious ideas now, but a diversity of efforts means more room for experimentation when the obvious ideas are mastered. I doubt there's a lack of talent and infrastructure at these schools, so there's no point in limiting the venues for that talent to express itself.
Also, it isn't obvious now, but when they start adding classes in economics, political science, and history, we'll want more diversity, not less. They're likely to go for breadth first and depth later, so each effort will have a single introductory macroeconomics course, a single course about the cold war, and a single course about medieval Europe, with each instructor chosen by a small cadre of people administering that particular set of online courses. Until the software is open-sourced and polished well enough that these programs start popping up all over the place, we'll be limited to the programs that are independently developing their own technology stack. The more, the better.
No, these new learning environments need time to grow before they standardize and this particular one brands itself as research which goes along with MITx very well which intends to grow into a certification system for MIT. Coursera and Udacity must have their differences and we'll all be better off with them competing as long as this trend continues.
The only overlap so far is Coursera and Udacity both teaching computer science 101. That's assuming either one of them actually ever gets a course off the ground. (Coursera is closing in on one month after their announced launch without any signs of life.)
It also seems that this one in particular it particularly designed to test the online system itself. Others are of course also trying to find a suitable format, but this one is explicitly an experiment in the viability of the software they developed.
Anyone interested in learning basically all of physics should pick up Feynman's series on it. It is beautiful and teaches math and critical thinking at the same time.
It's a series of textbooks [1] based on lectures Feynman gave while teaching the introductory physics sequence at Caltech.
To my knowledge, no universities actually use the Feynman Lectures as a textbook for their introductory courses. However, some students who have already taken physics read the Feynman Lectures later and (self-)report a much clearer understanding.
FYI, the audio of the lectures is also available, but can be difficult to find.
Afair, some of its a little scrappy at the start (poor recording setup?) but his delivery is great. I wish there was video available.
I don't like the subject selection for their first course. I think that the benefit of freshman physics isn't necessarily learning classical mechanics (F=MA, FBD's, kinematics, ect.) but teaching students to connect analytical thinking with mathematics and giving them a strong start for a science or engineering curriculum. I am not against learning for the sake of learning but I don't see how this would be useful to anyone who wasn't going to take more science or engineering classes, in which case they would have to re-take a physics course for real credit. I guess that I am less thrilled by their subject selection because the free cs courses already available online are useful and teach easily applied concepts but introductory physics just warms students up to begin working on a full B.S. degree.
Physics education is extremely baroque. There is a set way of teaching which has been done for the past 50 years and nothing is going to change it unfortunately.
Everyone learns from the same exact textbooks. My professors were using the same exact edition of Kleppner and Kolenkow (Into Classical Mechanics) textbook I used. It's even a point of pride for them. The canonical set of books are:
K&R -> Intro Mechanics
Taylor -> Advanced Mechanics
Griffiths -> E&M and Quantum
Sakurai -> Quantum
Boas -> Mathematical methods
(I'm prolly missing a few)
Every physics student in the US is familiar with these books.
Part of the reason for the intransigence is that it weeds out the weaklings. They make Physics very difficult and impenetrable in part so that 1/2 the class drops out and goes to do Bio or CS(which are taught sooo much better) and you're left with the most hard working masochistic students of whom the top 10% will become grad students and slave away for minimum pay for 7 years.
I mean.. they're not actively planning this in an evil-planning-room, but that's the end results and the physics establishment is happy and no one is pushing for new methods of teaching.
MIT still hasn't figured out a great way to teach physics, I'd look at this as another attempt.
Currently MIT offers 4 classes that count for the 8.01 (mechanics) requirement: 8.012, 8.01, 8.01L, or 8.011
Before they even place you in a class, they have you take an analytical exam to determine which class is recommended. This happens during orientation before classes begin, and it's the first of many things MIT does to help you pass the 8.01 requirement.
8.011 is the easiest, but they don't let you take it unless you've already failed one of the other subjects.
8.01 and 8.01L cover the same content, but 8.01L offers an extra month to go over that content. If you try and pass 8.01, but end up with a D (non-passing), they'll let you switch to 8.01L for the extra month.
8.01 is also a TEAL class, which is specially structured (and costs a lot of professor time) to encourage students to work together, and offers time in class to individually work on examples (as opposed to just listening to a professor).
The amount of effort MIT puts in to getting kids to pass the 8.01 requirement on their first try is exceptional. But with all their attempts, many still need to take 8.011.
I don't have numbers, but I'd guess that MIT spends more money teaching students mechanics than any other single graduation requirement. That probably made it a good guinea pig for this course, being that it's focused on education research.
I think the problem is that many of the stronger incoming students at MIT (at least from the US) will have already done AP Physics and thus not be taking mechanics at all.
I think it's a great subject for their experiment though: they want to design a system to teach deep problem-solving skills, and introductory physics (of the variety aimed at physicists and mathematicians, not engineers and liberal arts students) is generally the first place students are really challenged in that regard. The other place would be a formal mathematics course, but that would make an even gnarlier subject for an online course and would have even narrower appeal.
Those analytical skills stand very well on thier own without the need for further study to justify them.
Calculus and mechanics are part of a well rounded education in the same way as Shakespeare's principle tragedies, or an understanding of the causes and consequences of WWI are.
The fact that these building blocks of knowledge are going to be available in the near future for free to anyone with an internet connection is a modern wonder of the world.
This is a project from a Physics Education Research group, so the topic is pretty much their bread and butter. I think they probably hope that their methods will apply throughout the sciences and beyond eventually, but you can't blame them for building on their own field first. (PER is also the best established science ed. discipline, as far as I'm aware.)
I wonder if MIT will offer free web development courses. I know Harvard's David Malan offers free online courses (academicearth), but it's mainly PHP. It'd be nice to have a legit Ruby on Rails course.
Unfortunately, not sure if this would happen through MITx. The standard MIT curriculum doesn't have a Ruby on Rails course. It has project classes in which you could choose to develop with Rails, but courses don't really teach languages, save the few intro ones. 6.001 (no longer running) taught Scheme, I believe 6.01 teaches Python, 6.170 "reviewed" Java in 2 courses.
I think courses avoid specific technologies to focus on the concepts behind them.
The current implementation of 6.170 is "build a webapp in python". I seem to recall that they were using Flask, but that might have changed as this is pretty new.
6.00 is a class on programming in python. There are IAP (January) classes on C++, python, and a few more.
Other than that, though, it's all in the context of "do a project with language x" and not "learn language x."
I think once they have something mature enough to productize, they will do so. Basically every online consumer service ever started free and stayed that way until enough people recognized its value. There's definitely enough value here to do so -- people will start to discriminate between professors and course designers -- but as a platform there is no way it is mature enough for anyone to give themselves that much of a competitive disadvantage. We are still in the "Mad grab for market share" phase of online learning.
While I understand why this was downvoted, you must get how it's frustrating to see tuition rising every year while great potential revenue streams are being ignored. I'm still wondering what they're doing with dividends from Bose :/
