Week 11
Messing around with pixels on the screen, a method to test autonomous systems, realisations about myself, an interesting signifier and some thoughts on mentoring.
[Experiments] Accessing Colour Values of Pixels In A Live Video Feed
Over the past week, I’ve been experimenting with data that one can take out in real-time from a live video feed, all over an internet browser.
In a very comprehensive series about accessing pixel values by Daniel Shiffman, he mentions how the computer stores an array of the colour values of each pixel on the screen.
Therefore, whatever you see on your computer screen, is always translated into colour values for each pixel in an extremely long array. I decided to make something out of this information that is constantly stored on your computer.
We’ve all seen (and most have probably used) the colour picker tool. Select any colour on the screen and it grabs the colour value for you, allowing you to use that colour in whatever you’re doing. I found it interesting to explore if we could make something that could grab colour values from any object in real life, instead of grabbing colours from a pre-clicked image.
You can try it here and play around with the code here. It was fun to see how simple the concept is but it doesn’t really exist on the internet. I think I should think of more things that I would use myself, to drive the direction of my experimentation.
Next, I plan to track objects of a specific colour and use it to control the computer.
[Learnings] Wizard of Oz Testing
I’ve been working on a very interesting project at TinkerLabs that involves the creation of an AI-assisted chatbot for adolescents. While getting my hands dirty with some secondary research about chatbot design in general, I came across an interesting method to test these systems.
Let’s be honest: it takes a lot of time to get a working prototype of an autonomous system, which means something that aims to operate by itself and the rules provided to it (in the case of computed systems). In order to test these systems before they go into development, John F. Kelly came up with what is now known as the Wizard of Oz experiment as part of his thesis work at the John Hopkins University.
It involves a participant and a ‘wizard’ (the experimenter). The wizard’s role is to simulate an intelligent computer system and the participant interacts with what they believe to be an intelligent computer system.
[Self Realisations] Withdrawing Too Soon To Avoid Conflict and How I Plan To Resolve It
In an interesting and reflective chat with Ankur, I realised that I often to resort to withdrawing myself from high-conflict situations. This majorly happens when my point isn’t put across / acknowledged by the larger room and my “let’s-just-get-it-done” mindset kicks in.
What Ankur helped with, in this entire scenario, was suggesting different methods that I could use to tackle this. Explicitly communicating my thought process and valuing the experience I come with are two methods that I plan to use actively.
Often times, I tend to think that people just can’t seem to think in a particular way. I realise now that the thought process I have in a particular scenario has probably never entered their heads and, so, unless I communicate this tangent to them, they would never know. It’s not that they don’t want to think of a particular thing, it’s just that they’re not aware of this other line of thought.
Upon further reflection, I realised that this is not a new problem but rather something that has frustrated the people who’re close to me. Instead of prompting them to think in another way, I overanalyze and decide that they just don’t want to get it. It’s not that they don’t want to, they just can’t.
[Mentoring] You Can’t Force A Particular Reflection / Overnight Change
It was an interesting week / couple of weeks with regards to mentoring. Initially I couldn’t understand how my juniors were struggling to frame problems (about which I even wrote an article), then I got frustrated about why they unnecessarily overcomplicate the design process without understanding how simple it actually is, and, finally, why they just can’t seem to resolve some things themselves.
As part of my experience co-teaching ThinkLab with my mentor for the first time, I’ve realised how certain things have to come in their own time. She used an interesting proverb to explain this which said that if you’re famished and eating 7 hamburgers satisfies your hunger, you can’t expect the 7th hamburger to come in first and satiate your hunger. You have to go through the 6 hamburgers, in order to reach the 7th one.
Things have to happen over a period of time and as a mentor, one needs to be mindful that their mentees are in their own journey proceeding with their own pace. Certain realisations that you wish they have sooner so that they progress further than you did is a stupid ambition.
It’s their journey. All you can do is ensure that they don’t go astray.
[Observations] A Strategically Placed Signifier to Reduce Confusion
At a move theater in Vile Parle, I entered the lift and saw this:
In order to reduce confusion about where the screens are, they carefully placed a ‘cinema’ tag next to the button for the 4th floor. What a well thought-out solution and what got me even more excited was the deliberate placement.
Instead of putting it anywhere else on the lift / put up posters outside the lift or in the lobby, someone decided to put it right where a person is prompted to make a decision.
Signifiers by themselves are not enough, they need to be in the appropriate place as well.