Pressure Project 2: Japanese Sign
Posted: October 18, 2025 Filed under: Uncategorized | Tags: Pressure Project Leave a comment »As soon as I heard the requirements for this pressure project, I immediately knew I wanted to use a livestream. Not because I didn’t want to physically observe people in the real world, but because a livestream allowed me to go anywhere in the world. This seemed like a great idea, until I started looking for livestreams. I actually found a playlist of Japanese livestreams on YouTube and as I was going through them I realized one major flaw: there really weren’t many interactive systems to observe.
There was a famous livestream of a major Tokyo intersection, but this seemed too unpersonal and generic. A smaller intersection, while more personal, was still just as generic. Although as a note, barely anyone ever jaywalked in the streams I watched, even when there were clearly no cars at all. I ended up going with a stream of people walking down a street. There were shops, restaurants, many signs, and it all looked very… Japanese. I chose this one because, on top of looking very characteristic for Japan, there was a sign in a great location for observing. Additionally, I watched from 12-1pm (their time) so there was a lot of activity happening. See below for the sign in a screenshot of the livestream.
What made this sign a good candidate to observe though? Firstly, as I said above, it’s framed in a way where you can tell when people interact with the sign. Many of the other signs aren’t on street level, and sense we can’t see the exact location people’s eyes are looking, you can’t tell which sign they are looking at. The sign being on the ground makes it very clear when people look at it. Next, although you can’t really tell in the screenshot, this sign has lights around it that flash in a particular way. This was the “system” what people would interact with. Below is a mockup of exactly what the flashing light pattern looked like:
Now you may be thinking, is this really an interactive system? Perhaps it’s a bit of a stretch but first let’s cover how people interacted with this sign and signs in general. In my opinion, there are four key interactions:
Interaction #1:
A person doesn’t even see the sign. This is the worst case scenario. Either our potential observer was too busy looking at their phone or was in a rush to get through the street who knows, but in the end our sign made no influence on them at all… 🙁
Interaction #2:
A person sees the sign but doesn’t look. This is what I believe to be the most common interaction. I know I said the first type of interaction was worst case scenario, but in a way this one feels worse. People could tell there was a sign there, maybe even glanced at it to see what it was, but the sign’s existence was so ambivalent to them they simply didn’t care to look further.
Interaction #3:
A person sees and processes what is going on with the sign but does not stop walking. This is a great improvement over the first two interactions. People become so interested in the sign that they become curious of what it is. This interaction comes in a range but has an ideal scenario: the head turn. If someone turned their head to read the sign, that means it was so interesting to them, that up until the point where they physically could not see the sign anymore, they were still looking. There is room for improvement here though, as these people still walked by the sign when the time came.
Interaction #4:
A person stops to look at the sign. This is the best case scenario. A person was so interested in the sign, that whatever reason they were walking for become overridden. They must learn more about this sign. This is the only acceptable scenario. I will now redesign the sign to accomplish this goal… 🙂
Simple Changes
Assuming we want to keep things tame with the changes, let’s focus on the lights before adding new components.
Possible Change #1:
Make the light pattern more coherent and interesting. In the mock-up above, you can see the light pattern may vaguely be in a clockwise pattern, but adding more states and making it a clearly clockwise pattern could make people more likely to look, if just to see the pattern.
Possible Change #2:
Add more colors. A pattern of on/off is alright, but a pattern of different colors is definitely more likely to get people’s attention. This also adds an entire new layer to the lights, and that added complexity could keep people’s gaze longer.
Possible Change #3:
Make the pattern flashy. If the pattern has many lights on and then off in quick succession, people may be more likely to look. Especially someone who isn’t really paying attention as a sudden burst of activity may steal their focus.
Intense Changes
The simple changes are largely superficial. While they may get people to look more often and for longer, they’re less likely to get people to stop, which of course is the only goal that matters.
Possible Change #4:
Add many varied colorful and random patterns. The idea here is that there are just so many crazy lights and patterns and flashes occurring that people can’t possibly understand everything happening in the time it takes to walk down the street. People will have to stop in order to get the full pattern, if there even is one.
Possible Change #5:
Added speakers and proximity detectors to the sign. A speaker that just makes noise could get people to glance over, but if the audio is proximity based and makes noise depending on people’s distance to the sign, the personal aspect is more likely to get people to actually stop. The sign could say things like “Look at me!” or “You there, stop immediately!” in reaction to someone getting close to the sign and in many scenarios that person will stop.
Possible Change #6:
Makes the lights extremely bright. Now maybe this could have the opposite effect as people can’t look at the something that’s too bright because it hurts, but a light that is extraordinarily bright could cause people to stop in surprise. Although again looking away is not ideal, even if they stop.
Stop Them No Matter What
It’s still possible that the above changes won’t stop people. But what can we do to ensure that people stop no matter what?
Possible Change #7:
Add a fake crowd of people in front of the sign. It should really look like everyone wants to see this sign. How could anyone walking down the street resist the intrigue of what they could be looking at? They may even join the crowd and then strengthen its attraction towards other people…
Possible Change #8:
Add a piece of currency on the ground in front of the sign that is on a string. When people try to grab the money, the sign retracts it back into itself. The act of having to bed down and ground the string will stop someone, and then after they are stopped, they’ll likely look at the sign either in intrigue or confusion.
Possible Change #9:
All other changes have a chance of failure. In this change, a motorized system is added to the sign’s wheels that allow it to move back and forth freely. A motion detector tracks people’s movement and moves the sign to block people’s path so they physically must stop and look at the sign. This is the ultimate solution. I suggest all sign manufacturers invest immediately!
After presenting, the class discussed a few things that are worth noting. It was questioned if people are really “interacting with an automated computer system” by simply looking at a sign, however the changes I made, especially ones related to proximity, easily bring the system as a whole up to that specification. In terms of a less invasion approach, proximity lights were brought up as a possible idea. I kind of had moved this idea to the audio but it could easily work with lights as well. For example, maybe the color changes depending on your distance to the sign or maybe more and more lights turn on the closer you get. Either of these could definitely get a person to stop, especially if they notice that they are the ones controlling the lights.
This was definitely a fun project. I was a little disappointed that I couldn’t find something more interesting in a livestream, but I was satisfied with how I was able to spin something extremely simple into something a bit more complex.
Pressure Project 1: Bouncing Idle Screen
Posted: October 15, 2025 Filed under: Uncategorized | Tags: Isadora, Pressure Project Leave a comment »The idea for this pressure project came to me based on the “achievements” that Alex gave us to work towards. At first I was concerned about how I could possibly keep an audience engaged for over 30 seconds with something non-interactive. But then I thought about something completely noninteractive that does keep people engaged, and that’s the DVD bouncing idle screen. I specifically remember doing this myself whenever I saw it, but I knew other people liked doing it too from seeing it in internet memes or referenced in pop culture. This idea seemed perfect as it could definitely be done within Isadora.
The only issue was that it didn’t feel like it would be enough work for 5 hours. I then decided that because this idle screen plays on a TV, I could simulate someone changing channels. My first thoughts for other channels were full static and the color bars as obviously I can’t animate anything too complex (although maybe a live feed camera could have worked…). This was when I started creating the project in Isadora.
The first thing I made was the TV. I wanted an actual TV to be visible instead of just using the edges of the stage because it looks nicer but also because it just makes it feel more authentic. I also wanted it to be in the classic 4:3 resolution that old CRT TVs had. Another aspect of these older TVs that I wanted to emulate was the curved corners of the screen (technically the entire screen is bulging out but this is a 2D representation). With that plan in mind, I created the first TV with two boxes: the outer TV casing and the screen. I made the outer casing a darkish grey hue and the screen was a sort of dark grey/green thing that I associate with turned-off TVs of this type (the screen also has a thick black border so the entire thing doesn’t go from outer casing to the screen). The first issue came with adding the curved corners of the screen. The best way I could figure out how to do this was to use a shape with odd insets as that was the closest thing to a negative space curve. The issue with this however, was that it couldn’t be layered under border while on top of the screen, as those were both being represented by a single square. See below:
To solve this, I recreated the border casing as 4 individual rectangles so that the layering would allow the corner shape to be on top of the screen and under the border. The also allowed the entire TV itself to have softer edges as the rectangles ended up not perfectly flush. The TV was also made into a user actor where the color of the screen was controllable. The completed turned-off TV is below:
Next was to make the main attraction: the bouncing idle screen. The first thing I did was create a default white square. I used two wave generators for its vertical and horizontal position, with the wave generators in triangle mode as the movements should constant the entire time. To my surprise, this immediately worked in making the square bounce around the screen exactly as I wanted, the only exception is that it was bouncing around the borders of the entire stage. After some scaling and position subtracting (mostly done through trial and error) the square correctly bounced within the TV.
Now that I have something bouncing, it’s time to make that thing change colors every time it hits as edge. I did this by using an inside range actor connected to the wave generators. Every time the wave generators left the range of 0.5 – 99.5 it sent out a signal. This perfectly corresponds to when the shape bounces off a wall. I then connecting this signal to three random actors and connected those to a color maker RGBA actor’s red, green, and blue values to generator a random color for the shape. Now every time the square bounces off a wall, it also changes color.
The final thing I needed to do was replace the default square shape with something else. I didn’t want to recreate the original exactly, so I replaced the “DVD” text with “ACCAD” and just put a small empty oval underneath it similar to the original. I turned this into a user actor to simplify the views and after a few more operations it looked great. See below:
I was extremely happy with how this turned out, but I still needed a bit more to complete the project. The next thing I created was the static screen. At first I wanted it to be as accurate as possible by having a large number of shapes act as pixels, but this quickly showed to be not possible. At one point I had over a hundred little “pixels” that would get a random black and white color and position in the screen but the lag this caused was too great to continue. Not to mention the fact that it looked horrible! I then briefly thought about using several images of static screen and cycling between them, but I couldn’t remember how to display and swap images and this seemed like the easy approach any way. I ended up using a couple dozen large “pixels” to simulate a sort of static. By coincidence, they ended up in a similar pattern to how the color bars looked and so I was satisfied enough. The squares simply get a random black and white color in a pretty fast frequency. See below:
The last screen I made was the color bars. This was very simple as it was just static colors, although getting the exact correct positions was a little annoying sometimes.
Finally, I decided to simulate the TV turning off as it felt like a pretty iconic old TV action and a satisfying conclusion. For this animation, I used two wave generators set to sawtooth and to only play once. One wave generator squishes a big white square vertically into a line, and then the other squishes it horizontally until it’s gone. The end result was surprisingly satisfying! See below for the color screen into turning off:
Now that I had all the scenes complete, I needed to link them together. For the idle screen, I decided to start a counter for the number of times it bounces off the top or bottom wall. After 20 bounces it switches to static. For both static and the color bars I simply had a pulse generator activating a counter to switch after an amount of pulses. There was probably a better way to do this, but I was running out of time and there was one more thing I wanted to do.
The very last thing I added was channel text to the corners of the static and color bar scenes. This would further signify that this was a TV the viewer was looking at. Ideally, this would be solid and then slowly fade away, but given the time crunch it was just a very slow constant fade. Because these scenes only play briefly, it isn’t too noticeable.
The complete (although slightly shortened) result can be seen below:
The feedback I received on this project was amazing! I seemed like everyone made at least some noise while it was playing. One person said they were getting physically engaged in the idle bounces. Some people didn’t realize it was a TV until it changed channels which actually surprised me as it seemed obvious given the context of the idle bouncing. I hadn’t thought about how someone who wasn’t completely familiar with it wouldn’t know what was happening or what the border was supposed to represent. I was extremely happy when someone pointed out the curved corners of the screen as I thought nobody would even notice or care about it. There were also feelings of nostalgia and anticipation among the viewers as well.
This pressure project was a ton of fun! Isadora is just a blast to create things with and pushing its capabilities is something I loved exploring. If I had more time, I definitely could have done a lot more with this project, but I’m looking forward to creating more interactive experiences in future projects!