Pair programming versus solo programming for SAAS Part 1

pairprogramming

For the SAAS course part 1 we are being asked to do pair programming for all assignments. 
A portion of the mark is attributed to submitting a video of ourselves performing pair 
programming. Therefore I am submitting this document in lieu of a pair-programming video. 

I did do pair programming on HWK 0, with Jay A Student. Unfortunately we were unable to 
record the session successfully - getting Google hangouts to work was difficult and it 
seemed to be necessary to make the hangout public i.e. "on the air" in some way in order 
to record it. At the time this did not seem possible. The behavior of hangouts seemed 
pretty flaky. However we did do the work together. There were some pluses and minuses 
from this:

- I got the chance to meet someone else on the course. 

I do a lot of MOOCs, and I tend to avoid the discussion forums because things often 
descend into trolling and arguments. I am in full time employment, and since August 
last year I have completed 26 MOOCs. Generally I am following up to ten MOOCs at any 
one time. At the moment I am just doing six, as I had international travel scheduled 
this month. This means I have to manage my time quite carefully. Therefore, due to the 
often confrontational nature of MOOC discussion boards, particularly on Coursera, my 
policy is to avoid them wherever possible. This may seem paranoid but here is a 
practical example: on a recent course, we were using a data set with a license in a 
peer marking exercise. I wrote a comment (based on experience working with open 
source at a number of companies, although I didn't qualify it like that, it might be 
seen as inflammatory) that, even though I had not deducted any marks for it, it was 
important to quote the license in any derivative works. I think this is just basic 
good practice. However this produced a surprising amount of down voting and complaint. 

So meeting someone else on the course, in a non-confrontational situation, was good. 
I appreciated the chance to do this.

However it did take extra time, both to "get to the point" of doing the assignment, 
and it took longer to do the assignment itself. As I outline above, I am a bit short 
on this. Although I do have software development experience, I am new to Ruby, so I 
find I am constantly having to refer to Google in order to complete assignments. I 
also find this course tends to ask challenging, corner case questions, rather than 
building up our knowledge with basic practice questions. The course is demanding on 
the student, which is a good thing, but in a pair programming context, I would 
normally expect to go into it prepared with a base, functioning level of knowledge - 
it's meant to be pair programming, not pair teaching. Here I feel unprepared for the 
assignment. Also, in the assignment I did paired, I found it's too easy to only do 
half the assignment, i.e. leave the other half to the other person. This doesn't even 
have to be deliberate - we used Cloud9 but we had quite a bit of trouble getting update 
to work so we ended up having to use manual syncing which also slowed us down, but also 
impacted our ability to discuss the solution.

So in summmary I think doing pair programming is a good idea, this is a good opportunity, 
but I would prefer not to have to do it for every homework exercise, or for it to take a 
smaller proportion of the marking scheme, so if I omit I do not feel bad about it. Due to 
my international travel, and work commitments, it's been hard for me to do the last two 
assignments via pair programming before the deadline, but I will attempt a later assignment 
via pair programming to see if I can overcome some of the difficulties I experienced the 
first time around.