This week was a lot of coding and learning on the relevant material. Apparently, I was a ‘full-stack developer’ according to my friend which I did not expect from working this job as a researcher.

Anyhow, I managed to set up a feature for people to hide their cameras. That’s quite easy but the hard part is to also change the window layout to hide their box instead of blacking it out. Due to the strange coding in the layout, I had a hard time with it before I finally figured out how to reformat the tile layout according to the number of people who hid their cameras. However, it still has an issue where the person who hides the camera can’t see all other people if there are more than three participants but that wouldn’t matter too much in our study so it is not an issue to fix due to time constraints.

I noticed a few bugs such as that the RTT is not transmitting symbols like ‘%& etc. There’s also some missing text due to lag but it’s not important either.

So I have to make like 7 different settings. Two of which we already have. I copy-paste the folders to make a copy but they have a node build that takes up a lot of space and doing that from the network drive is incredibly slow but there didn’t seem to be another option at the moment.

I also tried to do set up Git for them but there were a lot of issues. I resolved one by one but in the end, we didn’t have too much time to actually get the Git implemented. So I did it the old way by just memorizing the good points to return. I may implement it in the future.

So the third setting is to have the captions out of the video. It took a while figuring out where to actually modify the code because our mentor told me it was in the CSS but I knew it wasn’t it because the CSS code doesn’t have it and the window resize function clearly gives different textbox sizes and positions depending on participants so I looked around and finally figured it out. There’s a little bug where if you have 1, 3, or 4 people, then the boxes would overlap or disappear but that is fine because our study will be only 2 people.

I went back to the second setting and tried to add the camera feature but it was quite clunky because the codes were different in that room but I managed to do that as well.

I updated the research agenda according to what we have to better instruct ourselves during a study as well. I also added more to the presentation because I felt like that our main supervisor would be asking for updates so I wanted to easily show what we have left to do and what other plans we have and what we have now all that stuff. Luckily, that was what happened and I impromptu-ly presented quite well (on Thursday).

I wanted to streamline the study process further so I connected all rooms together by having them link to the next one. However, to have all those rooms open at the same time, they needed to have different port numbers which I fixed. The main problem is that they each need one command prompt window which can be laggy for me – who would be recording. So I wanted to make some scripts to start them each up much easier but there was an SSH key issue so I can’t run it straight from my Windows. But I figured that I can just log in and run a script there instead, which worked. I made one script for each setting, and then I made one script to open all these scripts at the same time. If I were to open up all the servers without a script it would be about 28 steps. But with just the script I got it down to just 3 steps process. I can save one more step if it wasn’t for the SSH key issue.

Lastly, looking back on this week, I realized how much more work, effort, and energy I put into just coding alone and it wasn’t really fun. On Wednesday, I signed in early because I wanted to get it done faster but I failed to do that. So I signed in early on Thursday as well. But I didn’t get it to finish in time so technically I’m working overtime to finish some things up and then even more time to write this blog.

Unfortunately, there’s still some work to be done such as live dynamic captions – which is a fancy way of saying “put captions next to the speaker’s face by using face tracking.” That is quite hard with the messy code. It’s like planting corn, but you are doing it in a desert. Then minor adjustments with the last few settings and it would be good to go for a pilot study.

Forgot to add, during the weekly meeting for everyone, they were trying to get us to socialize again but it was quite awkward as always as we didn’t socialize outside of meetings. So they try to set up an official social hour with an interpreter which sounds incredibly awkward but I hope that works out. I guess I’m too tired to be excited about it is why I forgot.

Luckily, there’s no work on Friday, which is why I’m doing this blog now as well. Signing off.