I mentioned on Friday that I have three final pieces in order to finish off my degree (user study, writing my dissertation, and submitting to a journal). While I don’t have the three completely hammered out, I have started working. And am hoping this week I might be able to come up with an initial outline for my dissertation (and get that moving ahead).
However, the main thing I’ve focused on, has been the user study that was discussed. And I started by checking into the feasibility of what was proposed. If I had tracked time from the beginning of my PhD, I bet I’d find that I have spent upwards of 30% of my PhD determining if something was feasible or not. Every single step of my PhD has started with a hypothesis followed immediately by “can we do this using x?” or “is it even possible to do y?”. Which has meant that I start off by spending a lot of time not directly working on the hypothesis, but figuring out if our decision on how to proceed makes any sense. And sometimes, it hasn’t, and we’ve had to re-think and change direction. The worst time this happened, was during the beginning, when we had to really change direction and re-think my entire project (and ended up “wasting” almost 2 years*).
And so, with our planned user study, I knew the first thing I needed to figure out was whether or not this was even going to work. And so I started by researching the idea from home. What could I find out about the tools we had chosen? What had other people done? Was there any promising news that any of the potential initial red flags were going to be solved soon? After doing that, I met with another student who is using the system and we sat down and looked through what you could actually do. On one hand, it’s really cool, and it’d be very neat to re-do some of my stuff here and be able to show the “robustness” of my solution. On the other hand, there are some serious limitations in their design – limitations that I don’t get and are so confusing as to why they are implemented they way they are.
So now that I’ve done some initial feasibility exploration, it’s time to sit down with my supervisor and figure out how to proceed. Is there a way we can modify what we want to do such that the limitations are no longer a problem? If not, is there another reasonable user study that we could do that would be similar in design? Or, would the best solution be to just not do a user study, since this “failed”? I don’t know what the answer is going to be. And I imagine we won’t be able to solve it in an initial meeting. On the plus side, at least this time, I actually have other stuff to work on, while we sort this out.
* It wasn’t really a waste. It helped frame and set up the motivation for what I’ve done for the rest of my time. But it definitely took a long time to figure this all out.