Monday, January 13, 2014

Interesting Articles / Quick Thoughts

I like this article on working hours and flexibility by Jeff Haden.  I think a lot of it is true.

I just read a different article, this one on Big Data (which I won't link); it was discussing how 2014 is the year for enterprise investment in Big Data.  Basically it was an article about nothing.  It's easy to talk about Big Data.  It's easy to collect lots of data.  It's easy to store lots of data.  Actually doing something useful with it is a challenge that eludes most companies... 70% of organizations plan to deploy big data projects in blah blah blah buzzword buzzword blah blah... honestly it doesn't really matter unless you figure out some way to use it.

Is it bad that my favorite parts of when my wife watches "The Bachelor" are the complete and total mental breakdowns?  Does that make me a bad person?

Here's another article, this one about starting your career at a big company.  I'd actually give the complete opposite advice to a software engineer / CS grad starting a career for a few reasons: I think the opportunities are limited in a lot of (not all, but many) big companies for entry-level; one thing I've observed about big companies is that it's easy to "disappear" - I've seen a higher concentration of low-impact people in big companies - because in small companies you can't blend in as easily; I think there are more opportunities to *do* more as a software engineer in a small company vs a large company; and I think there's no less credit in working for a start-up or a small company than there is for a large company.  I'm guessing this one is targeted to more of a business audience, but I don't think much of it applies to the typical Computer Science grad.  Big companies might provide more room for advancement (which is a common complaint in small companies), but in my experience small companies provide more room for learning (which is pretty important in those first few years out of school).

Mobile advertising is where it's at in 2014 (in other breaking news, this newfangled Internet thing is really taking off):

Thursday, January 2, 2014

Maybe we'll all be using MOOCs someday.

Online education seems to be picking up.  Georgia Tech is offering a new MS in Computer Science and it looks like the first time a traditional (and reputable) brick-and-mortar school is offering a degree via massively open online courses (MOOCs).  It's an interesting idea, and I'm watching it closely to see how it unfolds.  Will it be a success?  More importantly, will the students who graduate with a degree from the program be the same caliber as the more traditional students?  Regardless, there's something appealing about earning a degree without having to put on pants.

I've completed 3 degrees completely online and from my experiences I'd say that online education has a few issues and a few strengths.  While online courses can be convenient and cost-effective, when you compare online courses to traditional courses, there are a few things missing:
  • When you're dealing with group work, you don't have to look anybody in the eye and this impacts trust, relationships, and commitment.  Until I started using Google Hangouts to do group meetings virtually face-to-face, I really dreaded group projects in online classes.
  • A lot of the time, you can learn most of what you need by reading the book instead of watching any lectures or reading any class notes.
  • Online exams often turn into an exercise in looking answers up in a book very quickly.
  • It's hard to form friendships with classmates or relationships with professors.
I've also spent some time teaching hybrid online / in-class courses.  It seemed that the students in these classes were only interested in completing the course to move on to the next one and ultimately complete the program.  There was no enthusiasm for learning.  This wasn't true across the board - but it seemed to be prevalent.  I would expect MOOCs may attract individuals with a higher interest in the material - but I'm not really sure.

I'd imagine that MOOCs have similar problems, but especially would lack the personal touch.  In online courses, this is a problem - but it's exacerbated in a MOOC.  Anybody can watch a lecture, just like millions of people can watch a show on television.  It might be educational, but you need to *do* to really master the material in most subjects.  You need to think more deeply.  You need to practice.  Without someone to evaluate the work, guide you through the material, or help with advice, it's just self-study.  Some of the professors I've had in online courses (and offline ones as well) have been good about evaluating my assignments, but I'm doubtful that several of them have done anything beyond checking to see if anything at all was submitted (I never actually submitted Led Zeppelin lyrics, but I think I could've gotten away with it once or twice).  I'd imagine this would be drastically harder to address in a MOOC setting, especially for the type of review and feedback required to help someone through technical courses.

I'm currently teaching Computer Science classes at a traditional university.  I really enjoy teaching, but I've found the last year of teaching in a more traditional setting to be hugely more rewarding (when compared to my past teaching experiences).  I'm establishing real relationships with students and I feel more useful - more like students care about my advice and are willing to come to me with problems.  That personal touch is important.

Establishing relationships is important - but delivering great lectures is also vital.  This is where I think there's a problem today that MOOCs can address.  Look at the professors who have taught the classes you've taken.  Some might have been really helpful, but how many can you say were great lecturers?  Beyond that, how many could you say were possibly amongst the top 10 lecturers in the topic?

The question on my mind is, "Why *shouldn't* you learn from one of the best?"  I expect at some point in the future, an education model might emerge where the top lecturers record material and license it to other universities or provide it as the lecture portion of a MOOC.  A new class of "professor" could be employed in student success, guiding students through assignments, providing feedback, and generally helping students with that one-on-one touch.  When it comes right down to it, there's not really a strong reason that we need to have 1800 different "Intro to Computer Science" lectures given by 1800 different professors in this country, especially if those efforts could be re-focused on working directly with students.

As online education expands, it's likely to impact the way traditional education operates and improve efficiency.  A wave of more knowledgeable graduates could trigger a virtuous cycle of innovation as future graduates improve education which will improve future graduates.  We've been teaching and learning the same way for decades, so I'm really excited to see where the changes lead.