Critical Analysis #2 - Embodied Interventions

Saw this on my newsfeed and it reminded me of your project :) http://sfglobe.com/?id=13124

@Amal, in order to make the bus waiting time more reliable, we discussed that we can use an app called tiramisu app, that show a more accurate schedule of the bus. Since lots of students live off-campus, I believe the bus schedule will be very helpful for them, it may help them decide hong long they can stay in the library.

Most of the comments I have read above have discussed the idea of also having a mobile app. As one of the originators of the this project, I can definitely see the pros of having an app and think this is a great idea. It will help reduce overcrowding at the kiosks. At the same time, people might ask what the point of the kiosk is, if we already have an app that serves the same purpose. There are two points that I believe still make the kiosk relevant. One is a point brought up earlier by Laura that visitors to the campus might not be willing to download the app to use just for the short time that they are here. Also, some of them might not be tech-savvy enough to use apps or they might not have smartphones. As a result, the people who would use this app more would be current students and faculty, the people who will actually be on campus for the long-term. This leads to our second point, a point brought up by Raisa. We would only include some of the features on the app, the feature that would be most useful to the intended app users. We would have waiting times for dining locations, the list of events on campus, and the waiting times for buses. The interactive map is less likely to be useful for people already on campus for a while and would be hard to navigate on a mobile so we could leave that out.

Hence, the kiosk would still be very useful for new students and visitors. This also means that it would make more sense to have these kiosks in places where more visitors will arrive at as well, like parking lots. This is a point mentioned by Tonya earlier, which is a good observation.

Many of the comments seem to have to do with worries about safety since we plan to implement vibrations; we see your concerns, and would definitely take these into account. We planned to find the combination of sound, light, and vibrations that would have the best effect, and perhaps just sound and light would work great! We do understand that this system may not deter all jaywalkers, but we hope that it will have some impact by drawing both the jaywalker's, and other pedestrians', attention. Thank you for the helpful feedback everyone-- If anyone has any ideas to improve the effectiveness of our system, please feel free to let us know!

@Amanda Marano I agree that this solution doesn't change the number of people physically in the area. However, our goal was to change the manner in which people will wait. Currently, the main problem is that a lot of people waiting to pay at a cash register naturally results in a long line. However, if we shift the waiting time to pick-up,we predict that people might migrate to other areas of the room. Currently, there are tables and chairs on the other side of the room that are often empty. With the large display we implemented above the cart, people can see if their orders are ready from everywhere in the room.

@Amanda Marano @Dan Cheng @Ivan Wang @Kristen Smith
Thank you guys for taking time to analyze our solution!
You guys all had a common concern of congestion of areas outside of the exchange. This is definitely a problem that we failed to address in our solution, and it is also true that it may be hard to change without an infrastructure change to the Tepper building. As of right now I don't have a solid solution, but it will definitely need to be addressed in order for this solution to be successful.

Having read a lot of the feedback that this idea has gotten, I would agree that the lights could be a distracting and occasionally inaccurate way of representing where there are open seats. I think that maps that indicate where there are open seats in the main entrance areas of the UC's second floor would be enough to show users where they can find a place to eat/work rather than having them hunt down a disruptive green light. I agree that the idea of an app would be cool, but, for the most part, unnecessary for this project. I feel that most people would not want to download an app dedicated to finding a seat in the UC when there are maps that are capable of doing so conveniently located there already. However, it would be a nice feature in a larger CMU oriented app that had other services. Eric Wang mentioned that this implementation wouldn't actually help reduce the traffic at the UC and I agree with his observation. This idea was an attempt to ease the difficulty of finding a seat at the UC during peak lunch hours. However, it doesn't address the fundamental problem and realistically there are times when there isn't any vacant places to sit. This is something that I'm currently thinking about and trying to solve.

Thanks for all the input! There are a couple recurring points I see in the comments so far:

> LEDs might be too attention-grabbing, so how can we guide people to seats without distracting people already at work?
> A system of pressure sensors would be expensive and problematic, so what's the best way to tell which seats are occupied?
> How can we accommodate students who just want to go to one of their favorite spots?

These are good points and I don't know that I have answers for them now (though there have been some great suggestions).

I believe the intention was to extend the duration of the beeps so it lead all the way up to the end of the light instead of ending early as it does now. As far as people rushing from further away, do you think decreasing the volume of the beeps so they did not carry as far, as well as adding a timer which was a countdown to the next time it will be safe to cross would help with this? My intuition tells me that the visual cue of the timer would take precedence over the noise at a distance, and people waiting at the crosswalk may stare at the timer instead of trying to jaywalk.

Thanks for the feedback. I agree using an app will definitely prove to be more helpful than having one digital map that everyone will crowd around.
Also, it does seem more practical to have the light turn off when a table is occupied to both be less distracting and also be more energy efficient.
I agree with Anna's concern and I feel we should definitely allow people to see not only if a table is open, but how many chairs are empty at a certain table. We already addressed the problem with the chairs being moved around to other tables and by having weight sensors that connect to a signal that can also be displayed on the screen would be extremely helpful.

@Amanda Marano Your observation regarding the blocking and waiting seems to be the most cited detraction, so it's clear to us that that's something we need to think about and address! There aren't that many places in Tepper, unfortunately, but I think Kristen's idea of moving people into another lobby at the top of the stairs would help alleviate some. I don't have any idea at the moment, but thank you for bringing this to our attention!

@Brian To respond to your question about how we plan to attract not just visitors but also students, we designed that the kiosks so that they not only present a map, but also information that would be of interest of students - dining wait times, campus events, etc. One way we could make the kiosks more appealing to students is including the specific building plan for a kiosk in a building; visitors do not tend to go to specific rooms, but for a student looking for a classroom, the kiosks can be helpful.

These are all useful comments. While setting up a system like this would be very expensive, we were assuming an unlimited budget. It is true that the LED lights could become distracting; we didn't really talk much about that issue. The reason we preferred a gestural interface to a touch screen is that the gestural interface seemed to be a more efficient set up, assuming it works perfectly. Walk up to the area where it reads your gestures, flash your gesture, and it sends you on your way. Again, assuming it doesn't glitch, that seems faster than going through a couple levels of touch screen interaction. The app idea is sound, also, but I feel obligated to remind people that not everyone can afford a smart phone, and generating ideas that only serve the more wealthy portions of the population is unfair.

This is a truly thorough observation of the problem, well done.

Thank you for the feedback. Ivan, you mentioned about having a visual cue - a visual countdown synchronized with the crosswalk sound. This idea was actually proposed in addition to the implementation of the new 'beep-boop's. However this was not added in our final project. I agree that adding a visual cue would give the pedestrian a better time estimate. I also agree with Rikky's point about adding a countdown time for the Forbes/Morewood traffic. Our original intentions were to inform pedestrians with so much (accurate and precise) information (through visual and sound cues) that they would be less inclined to jaywalk. Adding these visual cues would make it easier for someone to gauge their time and pace themselves.

I feel like a system such as the one described above would at minimum be reasonably efficient at cutting down on Jaywalking. People who are convinced they want to Jaywalk still will, irrespective of any such system. But people on the fence could possibly be guilted/shocked out of doing something dangerous by something like this.

That said, a jarring sound and vibrating street could be extremely disorienting for many. Coming home at 2 AM after studying at Hunt for 8 hours straight, tired and hungry, hearing something like that would drive anyone completely up the wall. Furthermore, something like this could actually be dangerous for people with anger/tension/nervousness or heart related issues.

Maybe this system would not be helpful for the Morewood/Forbes crosswalk- since the reason that so many people jaywalk is because they know exactly when to time it so that they can jaywalk. Perhaps at other high frequency intersections in larger cities this system would be more effective.

@Anna Tan

I agree that displaying a wait time based on how many drinks have been ordered and have yet to be made would be very beneficial to customers, La Prima, and that space in general (if people see that there is a long wait, they won't even try to go to Wean 5 to get their coffee and this will reduce the number of people in that space unnecessarily). Additionally, this feature would be fairly simple to implement. Thanks!

Ivan, thanks for the insight about the sound being stressful. We never quite considered how it would make people feel who weren't even crossing the street. Hopefully, though, as people got used to the system, they would internalize it and not be alarmed by it (much like how people get used to the sounds that exist now the longer they stay at CMU).

Our group originally had the idea of installing a giant board above the stoplight that could be seen from down the sidewalk on either side of the street. It, in theory, would show pedestrians the amount of time that they would have to wait for the next light. People may not take the time to interact with a game, especially in the freezing cold when they don't even want to take their hands out of their pockets, but seeing how long they have to wait may help them feel less worried about crossing in a timely manner.

@Amanda As you said, the ideal solution would be to readjust the light timing, but from our minimal knowledge of traffic engineering in large cities, this would disrupt the traffic in much of oakland and require re-timings on almost all the lights around CMU which would be disruptive and prohibitively expensive, so our goal was a non-disruptive embodied solution that did not require significant readjustment on drivers or crossers part, but would be intuitive enough whee you could see decent changes almost immediately.

This is a cool idea, but once you introduce the idea of ordering through the app, the actual line becomes unnecessary. Also, relying on people to start and stop a timer for the wait time to be accurate probably won't end very well. I like the idea to incentivize it, but I think some students will get frustrated with being forced to do this and stop using the app. Here's my proposal: have a touchscreen display near La Prima as well as being able to order from your phone, and automatically have a timer start once an order is placed. A barista can then press a button once the order is filled, removing it from the "orders" list and stopping the timer. This way the wait time will be accurate, because every single wait time is accounted for. I'm sure there are a million ways to tweak and implement this idea, but the framework is there and I really like it.

These comments are all very useful, but I just want to touch on some aspects that might have been misunderstood- the beeping would alarm once you have stepping from the sidewalk and onto the street during a no-walking signal - not from the sidewalk itself while you are on it.

I was wondering myself if I would just keep walking anyways, and maybe I would the first time, but eventually it could stop people from making future attempts at it (they had to cross the street with the annoying sounds going from where they are stepping directly up at them, and would rather wait a little then having to experience that again). I consider this only because of experience with the beeping that sounds in a car if you have not buckled your seatbelt- and the effectiveness of that system.

It is true that the deeper issue is that people do not want to wait, and if they see an opportunity to run across then they will take it, even though it could be potentially dangerous.

@Kim_Lister @Eunice Oh

I definitely see how the crowded feet may impede the functionality of the LED lights. However, I thought it was interesting that you both mentioned that the use of a display might have been helpful. One thing we observed (and that we probably should have mentioned) is that many people are not really paying attention to the menu at all until it is close to the time that they have to order. Instead many people are on their phones or socializing with the people online. Now that you mention it, perhaps we should also look into a way to get people to pay attention to the menus instead so that the conversation between the server and the customer is better so people aren't spending time making decisions as they give their order.

@abhishek

We were thinking of just notifications to users with the app about when their order was ready, but with all these suggestions of an estimated wait-time, I think it would be a great and easy-to-implement feature.

In response to how La Prima workers would increase their productivity:
Well, first, there doesn't have to be one person working the register all the time. Actually, we've seen the person at the register leave to prepare orders in the current system.
Secondly, from our observations, it takes 2 minutes for people to wait to put their order in, and that's just due to the register.
Also, suppose two people wanted the same drink. We were thinking our queue could operate to tell the employees to make the two drinks simultaneously, which would be of no inconvenience to the employee, and virtually no increased time for others waiting.

The approach to observation is really well considered - you've been able to gather some nice metrics about the ordering process. This does rely heavily on the quantitative, but what about the qualitative? What was the experience for people in the queue and how did they respond to having to wait for so long? Is the experience better for people with a friend? How did they bide their time - did they use their smartphone, look around, etc? By focusing onthe quantitative, you have overlooked some of the embodied and experiential components of ordering and waiting. I think this could be equally revealing for your project.

That said you've a very well rounded documentation of your concept which is really well illustrated and explained.

You might explore some alternative strategies that go beyond the planning for ordering or functional process of queuing. For example, what might help them pass the time while they wait - is there a game they could play that only works in the space? could the application spark conversations going between other people in the queue? etc.

it is nice to see how the discussion is bringing up new applications that tell something about our community. We know that we live and work amongst busy people but we dont have time to see what they all do. What if a network of customized kiosks and mobile apps allowed us to browse work happening at CMU while we wait to get food or take the bus. It helps with the waiting and with the community. This is exactly why group crit is important; it generates new ideas through the collective process.

Thanks for all the good feedback! It is good to point out that many jaywalkers would ignore our implementation and keep walking; those bad habits can be hard to fix. However, the target of our implementation is the group of people who jaywalk without knowing it (ex. if they are texting while walking, or following another jaywalker, or just really really tired), so the goal of our project is to bring those jaywalkers to attention.

However, since many of you have discouraged the use of vibrations, I think our group should consider that option.

I think this is a great idea; I especially love how it would incorporate bus schedules, which can get very frustrating. Because of that, you might want to consider putting some of the kiosks near bus stops to encourage people to check which buses are coming soon. I also agree with Judy and Amanda in thinking that it might be a good idea to have a mobile app, as well. However, because a map on such a small screen can get a bit difficult to navigate, you might include just wait times at dining areas at CMU and bus schedules on the map, which could also potentially solve the problem of overcrowding at the kiosks by making some of the information available on people's phones.

I like how this idea has a good amount of simplicity to it in terms of user interface and its intuitiveness, and will also probably reduce the long queue.

But I also agree, as some people have mentioned before, that the biggest problem this would introduce is a huge backup of orders. With people being able to order all over campus, I think at the end of every class period, there will be a sudden huge influx of orders from people who wish to get a snack on their way to the next class without waiting in line. This places a huge burden on the staff, because they don't have sufficient numbers to handle hundreds of orders at once. Also, if people don't get there right on time to get their order, prepared food will start pooling, and then when they finally do get there, they will have to rummage through the food-reservoir to find the right order. I think an approximate wait time display that constantly updates would be helpful for people wishing to order so that orders do not stack to a ridiculous amount.

This is a great idea to cut down on the line that often takes over the lobby of Wean! The app and touch screens seem very easy for a customer to use, and it does seem to make employee's lives easier too, as they can just read an order from the screen as opposed to taking time to listen to a customer that may be hard to hear in a loud, crowded space. I do have a couple of concerns, though. The first is that I wonder what would happen if something sells out. I like the idea of a "food counter", but this would take away from employee's time to update the app constantly. Also, what if there is just one of an item left and two people order it on the app at the same time? Second, I am concerned about customers that would like to pay with a credit or debit card. Would they swipe their card at the kiosk? What if they order using the app? Besides these minor issues, I think the idea is a good one (maybe it should actually be implemented)!

To me, this idea is incredibly sound in practice. As an events organizer for clubs on campus, the ability to advertise through the kiosk is an added bonus. There are two drawbacks that I invite you to address - you've probably thought about this and I'd like to hear your ideas!
First, this seems incredibly dependent on people reliably reporting their progress through various apps in order to realistically calculate wait time. This in particular strikes me as unlikely - how many people in a line are going to pull out their phones for the purpose of logging where they are? How exactly do you plan to calculate wait times?
Second, this intervention seems to be of limited benefit to those who are already on campus. Most students are probably already aware of where they are going, and if the idea is to keep these out of the way, it seems unlikely that some would go to such kiosks for any reason. This also limits the helpfulness of the events on campus advertising - not many would be pulled in by this because not many would check for it!
I think this project has a lot of potential, and it would definitely help out those who are new to campus, so it solves the problem you all set out to solve. These are just things to think about if you try and expand it. Great job guys!

This definitely seems like it will get the attention of someone jaywalking, but I'm worried it may lead to more safety issues. I could see the vibrations startling someone unaware of the situation, which could be quite dangerous in the middle of the street.

The Exchange do get really crowded during lunch time and it is hard for people to see the menu. I really like the idea of swiping card to get a number and wait to be called, since it gives us a better idea how long we need to wait to get the food, so that we can decide whether we have enough time to eat there or not. However, I agree with Amanda and Eric that space will be a problem, since the people waiting may block the people who are getting food. I think there should be a designated place for waiting. Also, I think it might be better to have the menu on the ticket dispensing machine rather than on the TV, since TV showing news is a good way for people to kill the waiting time.

This definitely addresses a prominent problem many of us have faced before. However, i do agree with some of the previous posts that comment on the control taken away from the user. Sometimes, when I work in public spaces, I know that there are still some open spots. However, since I already have specific tables that I prefer to work at, I will decide to leave if that spot is occupied, even if the area is not completely full. A similar solution that gives the user more control is the reservation system for the work rooms in the basement. Although it would be completely unfair to reserve single seats ahead of time, it may be useful to be able to choose several preferred locations to easily check if those are occupied, and receive alerts when they are open.

I like the idea that the kiosk includes information about restaurants and activities as well as directions. I know I have been asked multiple times for directions from visitors who have maps, but aren't sure where they are or which building they are looking for. As Amanda mentioned, it might be useful to also have this as an app, but I feel visitors (who often come for a short time and may never come back) might not be willing to download an app, and that would be if they knew it existed, which would be another problem entirely. I was curious who your target audience is? I think depending on who this is intended for (students, visitors, etc) you could see if this could lead to a backup at the kiosk or if it would be in use enough to justify the costs associated with its implementation.

As others have said, this is definitely a good idea to reduce the mob that occupies the fifth floor of Wean, but there are some problems that I think are just due to the nature of the problem. As Kim mentioned, there would be the problem of people ordering and not picking up, but if this app can increase the efficiency of La Prima enough than from a business perspective this is still profitable to La Prima, especially if payment happens through the app and not at a cash register. That way the fact that people already paid for their drink would encourage them to always pick up their drinks because they already paid for them.

While I think this is a good attempt at solving the problem of jaywalking, there are possible unforeseen consequences that you might want to consider. The beeping and flashing light could be potentially distractive for drivers on the road; if they do not know that this system is implemented and someone happens to be jaywalking when they arrive at the intersection, they might be startled and lose focus on driving. Furthermore, they could confuse the beeping with another car beeping or other alarm sounds. For the jaywalkers themselves, they might be inclined to run across to minimize the time of beeping and become even more careless.

Nice idea, based on a real problem; also nice discussion. One concept to think about is that addressing problems like the one described here. or any problem in an adaptive context, may require networks of solutions. In this case a network of kiosks with different functions per location (i need different info when I am standing at the bus stop and different info at the Cohen Union) that are all accessible also through an app, may allow us to have our cake and eat it. In general, like Stravinsky says, if you use a piano make sure you have a good reason. In other words, in the age of mobility, if you use a physical fixed intervention, like a kiosk, there has to be a good reason for it or you need to generate one. At airports, people are rushing by and having many screens available that they can check as they move without having to pull out their mobile, is very useful. When waiting for a bus, finding a way to both kill time and learn something about CMU may be in order (have you seen the poetry in motion posters in the NYC subway, they really work). Overall, good project, good concept and the proof is that it generated good discussion.

I want to commend you guys on the level of thought and observation you put into this project. Not only did you observe the lines at different time intervals but you also tracked a sample of people to observe. The idea of the app itself is great - it definitely would solve the overcrowding line issue and reduce traffic in the area which is important since the line runs through the entrance of wean to the elevators. As I mentioned before though, how would you deal with people ordering on location via the touchscreen and people placing their orders on their phones? Specifically, if I place an order through my phone while I'm walking from my last class in Margaret Morrison because I know it will be ready for pickup by the time I get to Wean, how would the person who already placed there order and is waiting on location to pick up the order feel when they see me come up to the counter, pick up my drink and walk away? I think that there is room for development in the way in which the queue is organized. Also, because the app provides such an increase in convenience, I worry about how popular the app may become. As you mentioned, some students left the line because of the wait but with the app, what if the orders increase in quantity and La Prima becomes busier than ever and the wait to pick up a drink increases?

I like how you guys thought of speeding up the music to play to a person's psyche. It would definitely instill a sense of urgency in them, causing less people to start jaywalking. However, one observation that I have made in my daily usage of that crosswalk is that most of the jaywalking takes place when the light is green for vehicles from Morewood avenue to turn left or right into Forbes Avenue. This could be due to the fact that traffic on Forbes would be moving faster so when that traffic is stopped in the previous instance, people are more inclined to jaywalk. Perhaps a way to improve on this would be some method to show the time left before the traffic light changes from Morewood traffic to Forbes traffic?

I really like the idea of having the daily campus events displayed on it. It's often difficult to keep track of everything that's going on on campus and a lot of the times there might be events that I'd like to go to but have no knowledge of (usually because poster boards are cluttered and difficult to read through)

I agree that it might be beneficial to consider focusing on a mobile app version of the idea since that way it'll be more accessible to everyone and you can use the interface without having to be near a kiosk.

I think that the idea of interactive kiosks across campus is a good and helpful idea for both current students who are used to the school's layout and visitors who may be completely clueless. At first, I was skeptical because the information that I thought would be shown seemed only relevant to newcomers, but there is actually a lot of potential for dynamic content with these kiosks. I really like the example of showcasing campus events that are going on at various kiosks as well as line lengths at different eating locations on campus. It reminds me of similar displays you would find on cruise ships that allow you to easily see which spot on campus has the shortest wait. I also thought that making approximate bus times available would greatly help users decide if is worth the wait or if they need to make another decision about transportation. It is very reminiscent of the signs at NYC subways. As other people mentioned, I feel that all these ideas and more could be combined into a single CMU mobile application that would allow students to easily and conveniently have access to this information. However, I think that have the kiosks in addition to this will still be useful to notify newcomers and passerbyers of what is going on around campus.

This app could be very beneficial to that space in Wean 5, and to students who generally are looking for a place to eat and don't want to/don't have time to wait in a long line. Using crowdsourcing to let people know the wait time for different places to eat could be effective.

I think "locking" a user out if they do not opt into collecting the data for you is a little harsh and impractical. Some users may forget to collect data, or some may feel that giving the app this data is an invasion of privacy. Instead of punishing users for not collecting data, how about rewarding them if they do collect data (coupons to places to eat around campus etc.).

The idea of using a virtual queue is certainly helpful. The thoroughness of your observations helped you to confirm existence of, and resolve, several current problems. However, while changing the TVs from news stations to menus would be helpful, people would probably have a tendency to gather near the TVs and ticket dispensers, even after they have decided on what they want to order. There isn't much space in the restaurant to sit and wait for orders, or to place tangible guides to show that there are separate lines.

To alleviate this potential issue, this design could incorporate some type of system for herding people into different places as they wait so that the space isn't used as inefficiently. Perhaps there could be a separate displays upstairs in the lobby of the building where there is significantly more space than downstairs near the entrance to The Exchange. That way, people could wait in a more comfortable space and still be able to see when their ticket number is being called.

Overall, the original idea is very sound. However, because the root of the problem is something that isn't easily fixed without serious remodeling-- the amount of space in the restaurant-- the design may be more efficient if it somehow found a way around this issue.

My group tried to solve a similar problem for seating in Hunt, so it's interesting to compare and contrast the two designs. I'm going to specifically consider the lights above tables. I think people might oppose installing such equipment because it would significantly change the appearance of the space and have considerable cost, while making only a small difference to function (as others have mentioned, looking for a light and looking for a person sitting there aren't too different). Granted, as far as sharing tables goes, the light would probably work better than the signs used now, but is there a simpler solution that would serve as well?

@Kim_Lister

What we did not clearly state in the project, but have planned, is to use square readers to read student cards and debit/credit cards. Unfortunately, for those who wish to pay cash, I suppose we could have the app make them go to a cashier and pay them.

Despite this setback, I think that the person running the register would be able to devote more time to making coffee, which is what we thought would also decrease wait times.

Cash of course is a problem overall for the project.

I think if the kiosks were placed in locations that visitors would run into them before they got deep into campus (so near all the parking areas people use), that might help reduce foot-traffic jams, because those locations are not necessarily where students are coming from since many of them walk or cycle to campus (or live on campus). Likewise, CMU has dorms that are specific to first year students, and each of them could have a kiosk in the common space. Perhaps if you were to do a study about what were the most popular things to look at, you could set it up so that 1/3 of the screen permanently displayed those things (like bus schedules) so that not everyone would have to wait in line to touch the screen in order to acquire the information they needed.

This is a great idea. Like some of the other people said, I see some problems with the lighting system, and I don't even think it's necessary. As long as there's a display somewhere with a map of the open tables, and maybe another display that just says how many tables are open (so you will quickly know if all the tables are taken), that will save a lot of time and energy looking for places to sit. Additionally, just like some of the above commenters, I think an app that displays this information would be very useful. Looking at your phone and seeing that there are no open tables before even walking to the UC would save a lot of unnecessary trips. I see a lot of potential with this idea and I hope it's implemented one day.

This is a clever idea for a problem I've always noticed, but never thought to solve! Adding a countdown clock and synchronizing the beep-boop, I believe, should help pedestrians gauge how long they have to cross. However, from my experience walking around cities like San Francisco, this and the accelerating beeping noise may not be enough; in fact, as discussed previously, it may persuade others farther from the road to speed up and cross. Accelerating noise may also cause stress-- personally, I would feel pressured whenever I hear that sound, even when not crossing the street.

Perhaps an alternative, as mentioned in class, is to distract pedestrians. One clever trick I saw on the internet involved a game of pong with a person waiting across the street from you. A simpler solution could be to play calming music as pedestrians are waiting, or display a small animation with the (estimated) wait time. This way, pedestrians can find something to do while standing around.

You clearly illustrated the problems that you were trying to solve here. Creating a "fake" line and explaining how others actually lined up behind you shows how dependent people are on others in this situation, and how changing their dependency on other to dependency on a ticketing system would save everyone a lot of time.

This reminds me of going to a restaurant- if there is a twenty minute wait, you can put your name down and take a walk before coming back when you think that it is time to be seated. It might be useful to have the expected time that it will take to reach their number, so people can make use of the time (for example, print something out, make a phone call outside, etc.) instead of clogging up the area if they are afraid to miss their turn.

I really like your idea! I know I personally tend to take more time than is necessary to find a seat. I do agree with many of the other comments, but I would like to also introduce another issue I foresee. Is it possible that the LEDs may be somewhat distracting to those trying to study on the third floor? I understand that you chose a "pale" green color, but when I think of LEDs, I think of a bright, somewhat harsh light that might be a little distracting to those trying to study or read.

I really liked how detailed the observations are. Your group succeeds in really capturing the minutae of the experience of ordering at La Prima. That said, I feel like your observational data can be somewhat opaquely presented at times. It is a little difficult to understand what is being said on certain occasions.

That said, I like the idea of putting up renders of your UI and system as it would look on location. That helps to really convey what you are planning to do better than any description ever could.

As for the idea itself, I'd like to offer a couple of suggestions and express some concerns. For one, the number one thing that concerns me when I'm running between classes and wanting to get a coffee is 'how long will this coffee take to get?'. A cumulative wait time display would go a long way for your idea. Secondly, while your system makes ordering something at LaPrima much faster, the actually process of making a coffee/beverage or serving food remains the same. How would the LaPrima employees keep up with the increased number of orders?

I like the intent and design of this project, but as others have said, there are some problems with how you could implement this system, first, it would be a massive construction effort to rip up all the floors of hunt to install these LEDs. In addition to being prohibitively expensive, disruptive, and possible problems with having a queue, my main problem is that this would reduce the utility of the library by making it more difficult to move seats once you've found one or to collaborate with other people especially in the ideate spaces. An alternative might be creating an interactive map of hunt using cameras and motion detection algorithms in conjunction with visual markers on furniture to display empty workspaces and concentrations of empty seats. This way would also avoid the problems of enforcement since it would simply be a visual information tool to tell where empty seats and tables are. While the idea is good, no doubt, implementation does need to be thought about especially in a highly trafficked area like hunt library.

I agree that going around campus is one of the hardest things for visitors to do, as nothing is really clearly labeled, and these kiosks will really help with that. During carnival, I felt like the experience was a lot smaller than it actually was, so displaying advertisements about school events on this board until someone interacts with it would be a very good way of notifying students of carnival events.

Personally, I don't think that kiosk backup would be too much of a problem for you guys, I think a set-back could possibly be that no one uses these kiosks. How would you get people to really buy into what these are meant for? It should be easier to draw in those who want directions, but how will you attract students?

This is an interesting idea to solve the problem of packed UC during lunch hours. However, have you thought about the need of a bigger table versus a smaller table for different sizes of groups? From your description, it seems like each light corresponds to a table, and the map would only show the location of the table rather than the size. If there is a group of 8 people trying to find a table, they would probably want something by the swimming pool. Also, if the table is only partially occupied, it would be nice to know how many more empty seats there are; if there is only one person at a 8-seat table, a group of 6 people should be able to know that that table is available.

One of the problems mentioned in previous comments was that of possible overcrowding of people in the waiting area. In my opinion, this shouldn't be that big of a problem as the fact that you can also order from the app instead of being there physically means that many people would order well before they actually reached the cafe. It is what I would do as I enjoy the idea of having my order ready to be picked up by the time I walk to the cafe. This would cut down on the amount of people physically waiting at the cafe. However, at the same time, it poses another problem - Having a pileup of orders that have not been collected yet. Just leaving the orders lying at a counter unattended might result in the 'wrong' person taking them, something that happens at the underground too. Other than this point though, i believe the groups solution is highly likely to solve the problem that they intended: the congestion in front of the cafe and the time taken to pick up food.

I think this is a good solution to the problem, but I also think that it creates some other interesting questions. You address the fact that an overhanging light might be distracting, but I think that alone is a critical issue that requires more attention. In addition, such a setup creates expectations - a green light means open. This prevents people from getting up to get water or perform any other action that would cause them to leave the table as the weight sensors would indicate that the seat is free. Perhaps that's actually a good thing though, in a situation where there are few tables for many people. Regardless, this is one of the more creative ways of addressing the problem, and it definitely has the potential to reduce a lot of headache!

I think this is a great idea for getting people to their seats more efficiently and it sounds like it could provide a little bit of fun for those that are coming to Hunt, especially when they'r getting ready to sit down for a long study session. However, I do agree with Judy that the LED lights may potentially cause a huge distraction. It might especially be a problem on the third floor, where people are trying hard to focus, if they see a green or red light pulsing every once in a while near them. I think a better idea might be one that kdsmith just posted, which is to have an interface at the bottom of the library that just shows what seats are available at the time. This solves both the problems of distractions from the LED lights and the problem of not being given "your" seat. There could be an option possibly placed in the interface that allows you to select the open seat that you want; afterwards, the seat would no longer be shown as available, which would potentially prevent others that came after you from taking the seat that you wanted before you got to it.

Very interesting idea. Might be a little overkill on the technology usage as maybe a touchscreen interface showing open seats on floor plans would be more efficient?

This is a good attempt to use embodied ideas to influence the situation, but I feel like the problem of jaywalking in this situation may be a bit harder to fix. Personally, if I was going to jaywalk before the noise was implemented, I would still do it afterwards. This is because I, as a CMU student have crossed that street enough to know the light patterns. I know that after the walk signal ends, cars from Morewood get to go first. If no one is turning right (happens quite often), then I can cross because everyone else is at a red light. Thus, when I'm looking to cross, my focus is solely on the edge of Morewood. The added sounds may affect how I jaywalk, but in the end my decision to jaywalk would be based on other factors.

Cool idea! How is the kiosk going to determine the wait time in real time though? Could it possibly be connected to some interface on the vendor's end, and they can update the wait time as they believe appropriate?

Is it necessarily safe to try to "startle" jaywalkers into behaving? To me I would feel this is further motivation to remove myself from the sidewalk as quickly as possible; if anything I would be more likely to jaywalk so I wouldn't have to deal with the startling stimulation on the sidewalk.
I think a better approach than trying to strong arm people into not jaywalking is to place more trust in them. Give them more information about the time remaining, and give them a longer chance to cross. If pedestrians have proper communication and accommodation I think they are more likely to behave rationally.

Thanks for the feedback! I do agree that looking down might be difficult or inconvenient in the midst of a large crowd. However, the intent for footprints were mostly to guide the first customers to the lines; as we observed, once a line or two forms, humans will naturally follow from social cues. That said, an overhead display is a fantastic idea. Perhaps we could improve the idea by moving the LED colors up, with additional text to guide customers or let them know how full each of the lines is.

This definitely sounds like a good idea, but as other people have said and you somewhat addressed in the writeup it may be a little impractical to execute it the way you have defined it now. The app definitely sounds useful, or making the lighting system simpler where something like light on is open and light off is taken.

I can definitely say that I have experienced this problem firsthand many times, and that this solution has potential.

But just as some others have said, I think there are a few problems that need to be flushed out. Specifically, I think enforcement will be a huge problem, because if there is only one seat finding machine, then at some point a line will probably end up forming because too many people are coming at the same time. At that point, I feel that it is extremely likely that people will stop using the system and just try to get to a seat before the machine can assign it to the people in front of them in line, and then people using the machine may have their seats taken away from them as they are being led to their spots. What are you planning in terms of scale and enforcement? Are you planning on using as the main way of seating in the library? Or just for people coming in who are wondering if there could possible be a seat?

It will definitely be difficult to constantly stare at the floor to determine whether or not there are people up in front. I agree with Kim with having an overhead display because staring at the floor could distract most of the people on the line and possibly cause more traffic. Maybe you could have a combination of both the overhead display and floor lights to make sure people aren't staring at one area for too long. But I do like your idea of splitting up the line so people don't get confused.

@Zach: I think our idea was actually to have the pressure sensors in the chairs themselves. I agree that even that could be error-prone, though, and a system where the students themselves actually indicate when they're present or about to leave might be better. Prompting every few minutes seems like it would be distracting, though. It might also cause a problem with people abusing the system to "reserve" seats when they're not actually using them.

I like the idea of using a virtual queue, but I agree with Amanda about managing space; perhaps you can section off an area for waiting so they don't take up eating space? I'd also like to point out the potential problem of a customer missing when their number is called, or if a customer decides to leave after taking a ticket. This could be alleviated by a short countdown, but maybe look into adding a queue of passed numbers as well off to the side (so skipped customers can return without having to go back to the end of the line).

Overall I think this would do an excellent job of solving the problem you observed, and I wish they used this system already!

I like the ideas for a "food counter" and estimated wait time that others have suggested. Also, your design kind of implies this, but customers should probably have to pay when they place their order, right? Otherwise you could have problems with people ordering and then not picking up their items. I think the crowding Amanda mentioned could be an issue, but hopefully if this system were implemented the area could be rearranged somewhat to add more seating (which is an issue now anyway) so that fewer people would just stand around.

A few details of your design that seem especially well thought out: the progress display, since there will likely be many people waiting who would appreciate knowing about how much longer they have; the three displays so there can be three short lines rather than a single long one; and the ability to order from anywhere on campus, because I can see people ordering before they actually get to Wean so that their coffee will be ready when they arrive.

I am still not really convinced this technology would help reduce traffic, as green light will pretty much be the same as seen an open table. Though I do think the digital map is a good idea.

Very cool idea! But using pressure to determine whether a seat is occupied seems like it may be problematic. Installing pressure tiles all over the library would be time consuming and expensive, and there are many possible sources of error for these sensitive instruments. If someone were to set a particularly heavy backpack down, for example, and leave it, the tile would be unable to tell that the seat is actually unoccupied. I think having a small screen accompanying the seat that asks the student at set intervals - say once per half hour - if they would like to continue holding the seat.

I agree with Francisco, showing where all open seats are would be more helpful to me than just making me blindly find a computer system. It gives the person more control.

Creative title. Maybe instead of just slowing down the beeps, you could have a voice that counts down once the orange hand starts.

I agree with others that floor lights might be difficult to see around people's legs and feet. It might also cause people to stare at the floor and miss when the workers behind the counter are trying to catch their attention. Maybe some kind of overhead display would be better, since people are looking up at the menus and such anyway.

Great idea. It reminds me of parking lights in newer parking lots. This idea is really convenient, as I find myself searching for seats for what feels like forever.

I agree with Amanda; I have a tendency to pick the same few seats on the quiet study floor and if it doesn't direct me to my seat, I'll end up taking my own. It may be more beneficial if the screen at the front of Hunt just shows open seats, then makes it unavailable to other people when you go upstairs and sit in it. However, I really like the idea of using gestures to make inputting preferences faster.

To clarify, I think the team was thinking of a pressure pad with vibrations similar to those in a haunted house. Just to clarify.

I think this is a good idea. The only thing I would say is that what happens when people are ordering on location through the touch screens and people ordering on their phones from a remote location for pick-up? Which of the orders get priority?

I think in this case, people would not stop jaywalking, although your intuition about sound indicating the necessary pace to safely cross the street seems fair. The longer someone is exposed to something, the less impact it has on them, and since most of the people crossing these streets are going to/from school several times a day, 5+ days a week, the sound would eventually be ignored to some extent. Perhaps there's a way you could work around that?

Personally, if I'm in a rush, I would keep walking. Think of how the crosswalk already beeps when the time to cross is appropriate. Having vibrations in the ground can be a potential hazard (cause imbalance while walking) and then it would be even a greater threat if they cannot pick themselves up and get out of the road in time. Also, if the sound is too loud, yes it would get out attention, but at the same time for people who were making an honest mistake/not intending to jaywalk, a large noise might frighten and confuse them. However, I did appreciate the steps that you outlined you would take during the experimental phase of this project.

I would just keep walking. I think the idea is really good, but the implementation doesn't have a big enough dis-incentive to potential jaywalkers. I'd be concerned that vibrations wouldn't necessarily be recognized because when you're jaywalking, you're usually moving quickly, so your feet might not stay on the ground long enough for you to perceive the vibrations.

I liked how you implemented gestures as a way of interacting with the system - the quiet area would be undisturbed and it is quicker than typing in answers into the system. However, I wonder if there was a better way to guide people to their seats instead of an LED trail on the floor - just because in areas where tables are close together, having LED lights moving across the floor could be a potential distraction.

I agree with Christina about the app. It would also be cool if the app had a reservation function in it (so you could reserve a table for a set amount of time). That might be difficult to enforce, or to keep fair, but it might also help.

I also think it might be confusing for people for one line to turn into two lines--how will they know which line to break off into for what? If it's crowded they might not be able to tell. Is there something you could add to make that more apparent?

I really appreciated this project because I have experienced this problem myself. I really enjoyed your solution and I liked how you even thought about what would happen if people didn't mind sharing a table. The only thing I would comment on is your choice of light color. Because the tables are not so far apart from each other, having a green, red, or yellow light above each table can mean that it may serve as a distraction to other tables (since having a table glow green next to you might draw your attention). Also, one the table is occupied, is it necessary for the light to glow red? Why not have the indicator light just go off completely?

I agree with Amanda that this may have been better served as a mobile app. While the kiosk idea is great, you may experience overcrowding (like you mentioned in the write-up) due to many people wanting to use the kiosk or by the time it takes to interact with it. If people are searching one by one, it would take a while for the kiosk to serve everyone's needs in which case just asking someone for directions might have been faster. Of course, there are limitations even to a mobile app. Overall, I enjoyed the renderings and the features you wanted to implement in the kiosks

Excellent idea. I think maybe changing the results to not include the vibrations but rather a very obvious use of lights or sounds might be good. Vibrations seem really difficult to implement on concrete since it's so rigid.

I like the idea of splitting up the lines to avoid one giant one. A possible area of concern might be that, given that the exchange is always really full and chaotic during lunch, there might be so many people that following the directions on the floor will become difficult simply because there's so many feet!

I think this is a great idea. I know personally when I toured the campus I had no idea where I was going most of the time, and maps on the web served little help. This would greatly help tourists and freshmans get acquainted with the campus. To respond to the earlier comment, I actually think physical large maps located at various points is better than an app. It is much more convenient and intuitive more the general masses.

I agree that the system of seat-finding in the library could benefit from some overhaul via technology.

It may make more sense/be more feasible to (instead of implementing LED lighting all over Hunt library) have an app that has a map of Hunt and shows the student where an open seat is or where ALL open seats are, and then the student can choose their favorite out of the available options.

I really like the idea and the detailed design picture, UC do get very crowded and it is hard to find a seat. However,I agree that cost will be a big problem, having a weight sensor under each chair is at great cost. It may get damaged or lost. Also, if there are a lot of people trying to find seat, it is still really hard to see the light. Maybe instead of lights, an app that shows where the seats are open is better?

It is an awesome idea. One suggestion is that maybe the app can show how much food are left(say the number of croissants left), because lots of people buy food there and food sold out very quickly. So there should be a system that keep track of the how much food is left. And maybe the app can cancel the order if the student does not pick up the food at certain time? Overall, it is great idea and I really like the pictures too.

While I understand the implication of group tables for groups and smaller tables for single people, I foresee a huge potential problem. As someone who works on the third floor of Hunt often (almost daily), I know from experience that not all seats are created equal, and I definitely have seats that I favor instead of others. And if this machine directs me to a seat that isn't "mine" and that one is open, I'll take my usual seat. How will that mess with the system if the user doesn't choose the chosen seat for them. Another problem is that not all study cubicles on the third floor have outlets near them (although most do). What would happen if the machine directs me to a seat without a nearby outlet and I need to use one? Would the machine know and ask me about my preferences?

I think this is a really good way to cut down on waiting time in line, however the people working there can only make a drink so fast, and if a lot of people still want drinks at the same time (peak time), then there is going to be a backup and its still going to take quite a few minutes for the drink to be made (considering it takes 3-4 minutes to make a custom drink). In that sense, even though these people aren't waiting on a "line" anymore, they still have to wait in that area, right? So in a sense, aren't they still blocking traffic flow by waiting in a crowd by the kiosk? If anything this could make it worse for commuting in and out of the building because now the big crowd is closer to the doors and harder to weave in and out of. Is there a way to have an app or some sort of noise or alert that your drink is ready be available so people don't have to crowd the kiosk and can wait outside or a little bit farther nearby maybe?

As I said in the other project with the same topic, I think a lot of people jaywalk because the "beep boops" don't really last long enough for people to cross fully anyway, and because there is such a large gap in times between light changes (good observations), people realize that they can get away with crossing a little later and not having to wait for at least 2 extra minutes to cross. When I had to cross the street more often, I found that, when walking at a normal pace and only crossing once the beeps started, I don't make it to the other side before the beeps stop. I think in this case fewer people would jaywalk if the beeps lasted a bit longer and the wait between crosses was a little bit shorter.

How would you have the road vibrate without compromising the structural integrity of the pavement? It sounds like you want to start an earthquake...not so safe either. I also think that a lot of people jaywalk because the wait times between cross signals is much longer comparatively than the walk signal is, and people don't really want to wait for that long, because it might make them late. Maybe instead of punishing people who try to walk not during the crosswalk signal, maybe make an option so people would rather wait than cross, or would make them less annoyed with the prospect of waiting.

I think a solution for your crowding of kiosk problem would be to have the map available as an app (I'm not sure if you mentioned that in your writeup). Also a question I had was: would you be able to use the map while being logged in to it through your andrew-id? I know that there are certain events on campus that require an RSVP, and I think it would be helpful if, as an app, there could be one convenient location to RSVP. Overall I think this is a great idea, as it is very hard to keep track of all of the events happening on campus!

I like this idea. and how you thought out solutions to many possible problems- specifically the "share a table" problem. Many times the "share a table" signs that people place on their tables can be somewhat awkward because you don't know if it was just left there or if the person sitting there actually doesn't mind sharing a table. Having the option to actively press a button that turns the "share a table" signal on makes sharing it easier.

Also, the maps would probably be more helpful than the lights themselves. It might be hard to see the lights from a distance, especially when most tables are taken and you are looking carefully for that one green light.

I agree with your potential problems with ordering ahead, but what about the people who choose to use the app and order like they normally would now? There would probably be a large group of people standing around the kiosk like they would be now, and the problem of blocking the entrance to Wean would still be in effect, because with shorter wait times people would continue to place orders until the wait time becomes noticeably longer on the app, and how long would that take, depending on how many people show up and now quickly? I'm imaging a scenario not unlike disney world lines that show the wait time, but when that time is short more people add to the line, and the wait time only updates every 5-10 minutes.

This is a great idea! However, I think it may still be difficult to determine if a table is occupied if it is behind a pillar or wall. (It definitely helps with figuring out if tables far away are occupied though.) Maybe you could consider adding an app to your system that CMU students can download with has a map on it that they can check as they are looking for a seat, rather than just having a large map that could be hard to read, especially if many people are gathered around it?

While I do agree now that this resolves the issue of waiting in a (physical) line, there still is the problem then of where all of these people are going to wait while their food is being prepared, especially if they order a sandwich, which takes longer to prepare than boxing up hot food. In that case, I've found that people waiting in groups tend to take up more space by loitering in different areas. How would this affect traffic into the academic building and the new people coming in to place orders where now people waiting might be blocking them? In my experience, I think people are more willing to wait closer to what their waiting for, but this might block up the table areas in the Exchange. Maybe if there was a designated waiting area with differently colored floor tiles to keep people out of the way of Tepper traffic?

I feel like while this will speed the line up, there is a certain rate limiting step, which is the speed at which a sandwich made-to-order can be made. In this case, the line can still become really long and then doesn't really solve the initial problem of the line spilling into the building proper and blocking traffic. Although, with this method, people wanting hot food will not have to wait nearly as long, which is an incentive to get hot food instead of a sandwich. I think in general, because the hot food line will be so much faster, that this will be a great improvement, though.

I definitely see how this could this could make lines go faster, but I do wonder if there is a way you could adjust this so that customers do not end up staring at the floor to figure out where they are supposed to be going. I feel that this could be an issue if, say, one line is very full and the other is not. That very full line could end up spilling over into the other line and covering up the LEDs. I do like how the system reacts to the environment though by adjusting the light color depending on the traffic!

This is a clever idea! The gestural and lighting interfaces are both simple and intuitive to follow. I can see a few potential problems with the LEDs, however. What if while one student is halfway to an empty seat, another student takes it? Or if two paths intersect, how will each group figure out which path to continue on? While these can potentially be subverted with advanced pathfinding algorithms, there is a technological (and monetary) tradeoff with the ease of use.

What if someone would like to order something, but has a short break only? I think it would be even better if the displays and the app can show the approximate wait time for people to decide if they should place an order or not. Otherwise, I would love to see this implemented!