Monday 27 April 2015

Sheepish Endeavours Completed



George is a funny little sheep. Unfortunately, his fifteen minutes of fame are over, and since I have to return the fidgets to the History Department, his ability to respond to his environment is also gone. I removed the phidgets, another invasive surgery, but I was able to stitch him together again successfully. He is just as cute as ever.

George has taught me many things about creating an interactive exhibit, mostly that it is a lot of work and requires almost inexhaustible amounts of tweaking. I am glad I was able to both try my hand at programming and put my sewing skills to work. I would have liked to make George more complicated, but my workload and George's waistline would not have accommodated it.

If I were to do this project again, I am not sure that I would use a distance sensor. It is a rather unpredictable phidget and requires a lot of troubleshooting. The touch sensor was a good call because it is very simple, either on or off, and it was very well hidden in George's head.

I thoroughly enjoyed this project and although my talking sheep is no more, I did leave one piece of technology in George: a speaker. He is now the world's cuddliest sound-system and he will always be able to tell me a sheep joke if the mood strikes me. :)

The Sheep Thrills of Programming 2

So I finally figured out all the programming. Here is a picture of my patcher:


Starting in the top left corner, we have the section of code that turns the sensors on. Once they are activated, the sensors each generate numbers. These numbers are sent out of the bottom of that section ("send distance" and "send touch") and received by the receivers ("receive distance" and "receive touch") on the next two columns of code. The numbers are then scaled down by the “scale” objects. This takes the numbers between 0 and 1000 and scales them down to between 0 and 10. Below the “scale” objects are “if” statements. These send the conditions of action.

For the distance sensor section of code, I have added two “if” statements. The one on the left tells the patcher to activate (known as a “bang” in Max 7) if the number is greater than 0. The one on the right tells it to send a “bang” if the number is smaller than 1. This section of code is designed to prevent the sensor from activating repeatedly when someone is standing near George.

The little box with the pointed, solid line is a “gswitch2.” This object changes the direction of the bang. When it is in the “off” (or 0) position, the bang cannot activate the code below (it is currently in the “on” [or 1] position).

There is another “gswitch2” below the first that basically controls which audio snippet is selected for playback. The “bang” travels through the “switch” and a “trigger” (an object that sends one “bang” to multiple outputs), which activates a “counter” object. This object counts up from 0 to 6 and then restarts. The number it generates will select what audio track is selected. The first audio track is number 1, the second is 2, and so on. So if the counter counts 1 after the first “bang,” it will activate the first track. When it receives another “bang,” it counts 2 and activates the next track, and so on.

Below the “counter,” there is also another “if” statement. This statement sends a “bang” if the number displayed by the counter is lesser than or equal to 1. Below that is a 3-second delay that will “bang” the fifth track. This piece of code was added so that I could play track 1 (“Hi! My name is George”) followed three seconds later by track 5 (“Pat my head if you want to hear a sheep joke!”).

The “trigger” object also sends a “bang” to a “message” that contains the number 1. This “message,” when banged, will turn the “trigger” off so that no new input will be accepted. It also sends a “bang” to a “delay” object that waits for a certain amount of time before sending the “bang” it received. After the delay, the “bang” is sent to a “message” that contains the number 0, which will open the “gswitch2” again and allow it to accept new input. This ensures that two tracks cannot start playing at the same time.

That is basically it. The touch sensor code works the same way. I included some extra messages to the right of the audio tracks that, when pressed, will activate a specific track. I added these for easy access and for demonstration purposes.

Voila! Now you understand what makes George tick. It is a simple patcher once you understand how it works, but it has fairly hilarious outcomes when powering a talking sheep.

Sheep speak 2

You may recall that I had originally planned to have a very sassy sheep. No longer! George has informed me that insulting people makes him uncomfortable. He wants to be a friendly sheep who makes people laugh. With this in mind, we went to the digital history lab in the UWO History Department and recorded some new phrases for George. This took a little figuring out because the microphones are connected to a Mac, a system I am not very familiar with. Here is what he had to say to my friend Kelly in Ottawa:
George thinks he is a comedian and has quite a repertoire of sheep related jokes. Here are just a few:
The first interaction George will have with someone is when they pass in front of him. He will exclaim, "Hello, my name is George. Pat my head if you want to hear a sheep joke!" At this point, the victim (presumably) will pat George's head and George will reward them with a sheep joke. It is a simple system, but I hope that a randomly talking sheep will intrigue, startle (possibly frighten), and definitely delight passersby. The final presentation of our projects is soon so hopefully George will be on his best behaviour. I’ve warned him not to be baa’d.

Little Sheep of Horrors

The time has come. George is going for his operation today. I have to implant an infrared distance sensor, a touch sensor, and a speaker into his soft, plush body.


George saying goodbye to his mice friends 


All the stuffing removed from George 

I have grown very attached to George and did not want to disfigure him unnecessarily, so I decided the best place to make the first incision was in the seam on his back. Once I had made a large enough hole, I removed the stuffing from his body and head.




De-stuffed George next to the soon to be implanted technology


Touch sensor implanted in George's head 

Now I could access the inside of the fur to sew the implants in. I debated for a time where to attach the touch sensor but decided to sew it into his head so that he would activate when people patted his head. I had to turn his head inside out, of course, to access the area that needed to be sewn.


Sensor implanted in neck 

One sensor down, one to go. The distance sensor, I decided, would be best implanted in George’s neck, where his bowtie was. I removed the bowtie and began the tracheotomy.

The seams of George’s body naturally joined at this juncture, so I just removed a couple of stitches to create a big enough hole and sewed around it, keeping the fur tight around the sensor so that it would not move. These sensors very helpfully included some holes at the extremities that allowed me to secure them very well.

Now that all the sensors were in place, I stuffed George with a small speaker that I happened to own. It was a little big for the hole I had created, but with some persuasion, I was able to squeeze it in. I attached all the necessary cables to the sensors and speaker and made a small hole in a surreptitious area under his tail for the wires to exit.



I proceeded to re-stuff George and sew him back up. Since I had to remove his bowtie so that it would not obscure the distance sensor, I sewed it to his ear. Now that George has his new implants, I just have to teach him how and when to speak. Stay tuned for some more Sheep Thrills of Programing and more Sheep Speak!


Ciber Sheep

Wednesday 1 April 2015

Sheep Speak

George is becoming more vocal every day! Today, he learned to say a variety of lovely things such as "I'm kind of a baaad ass". I also taught him some Shakespearian insults including "away you mouldy rogue, away!" Thanks to Nikki Michienzi for lending me her voice and her witty comments. Now all that remains to do is to help George use his new vocabulary in an appropriate setting. I will once again resort to Max 7 to accomplish this. For the moment, George can speak but he doesn't know when to stop. I am still working through some minor issues. Once that is done, the next step is to implant the sensors and speaker into George.

Here is my favorite thing he says so far:




Wednesday 18 March 2015

The Sheep Thrills of Programming

George still lives! It is still lent and not yet time to cull the lambs. Besides, I need to figure out the code for my project before I can start thinking about implanting things into George.

As proof of his continued safety, here is a picture I took this morning of George and his mouse friends cuddling on my desk.

The bags of wires at his feet are the phidget I am attempting to use. It is an infrared distance sensor and allows the computer to recognize the relative distance of whatever passes in front of the sensor head. Once I have the code figured out, this device will allow George to recognize when something is nearby and to respond accordingly, probably with a sassy remark. Or perhaps with a creative Sheepspearian quote. This should give the unsuspecting passerby a surprise.

I still need to figure out how I will implant the device into George when I am ready. I was thinking of placing the sensor in his bowtie. It is a fairly obvious looking piece of tech that will be hard to hide but hopefully the bow will be able to make it less obvious.

Baaey for now.
Nicole and George

Sunday 1 March 2015

Sheepish Endeavors: My Interactive Exhibit

Museums are often thought of as haunted, and this is a large draw for many people. The haunted doll is often an object of curiosity and concern for museum workers and visitors alike. I want to validate those concerns by making my own "haunted doll" by placing a bunch of sensors inside a doll to get it to react to different stimuli, such as touch, sound, and being picked up. I am not going to pretend there is any historical value to this project; it is purely an experiment to allow me to test the limits of my technological abilities.

So here is my plan: My first objective is to secure a doll of some kind. As we are nearing St. Patrick’s Day, the stores are obviously full of adorable pink and yellow stuffed animals and chocolate eggs. I found this cutie at the Dollarama for a mere three dollars; he should prove to be a worthy test subject. I probably shouldn't have named him but I did; his name is George.
George the Sacrificial Lamb
If my experiments with George are successful, I may crochet myself a new doll for the final exhibition because I fear that George will be horribly disfigured by my experiments.

Now that I have secured my subject, I will have to cut him open and gut him to stuff some sensors into him (I will sew him back together later). I am going to start with an infrared distance sensor to try to get him to notice when somebody walks by, but I hope to incorporate a volume sensor, an accelerometer, and some touch recognition. I will attempt to use Max7 to accomplish the programming side of the project. If all goes well, I will create a frankensheep that will talk when you interact with it. *Now accepting suggestions for sheep back talk.*

Tuesday 24 February 2015

Brainstorming Interactive Exhibits

It has been a while since my last post but since I am now embarking on a new project for my Interactive Exhibit Design Class, it is time to start blogging again. In this class, we have to create an interactive exhibit for which I have a number of ideas. I hope to use the model I built last semester and put it to good use. Here are some pictures of the final result.




Building on my last post and the SketchUp model of Pinhey's Point I made last semester, I would like to make a video game level, such as Left 4 Dead. I mentioned in my last post that I hoped to achieve this during the Christmas break. Unfortunately, that did not happen as I was rendered computerless over the holidays and the process seems to be more trouble than I originally thought. Hopefully, my model is compatible with Valve Hammer Editor (the program used to create custom level maps), but this will require some testing. My goal when making the model originally was to have a digital version of the house that people could explore when the museum is closed or when accessibility is an issue. Making the house into a Left 4 Dead map seems to be the easiest was to make the model fully explorable in first-person view. Shooting zombies is just a bonus and might be an interesting Halloween activity.


My next idea for an interactive exhibit is taking historypin and geocaching and combining them. Geocaching is an activity where participants find caches using a GPS or mobile device to help them navigate. There are geocaches all over the world and it is easy to start hiding and seeking by downloading a geocaching app onto your phone. Historypin is a website that allows users to pin historic photographs to their modern locations. Images are pinned to their Streetview equivalent and a fade filter is used to toggle between modern Streetview and the historic image. They also have an app that allows people to explore the area themselves (although it didn't work for me when I downloaded it) and hold the historic picture up to the actual building.

By combining these two activities, I hope to develop a fun game that visitors to museums, especially historic houses, can do while they visit the museum. The activity would start with a hunt for the exact location the historic picture was taken. It is fairly easy to upload geocaches to the geocache website for others to find. Once the location is found, the cache would include a reproduction of the historic picture or a QR code that could lead them to a website with the historic picture. The cache/website would also contain some interpretive material about the history of the house and metadata about the image.

Unfortunately, this might be difficult to demonstrate at the end of the term since it is an activity that requires space. I would probably have to confine myself to Lawson Hall and may need to use pictures that are less than historical to prove the concept.

Finally, my last interactive exhibit design idea is to incorporate QR codes into the exhibit that the Museology class is developing. These QR codes would lead to supplementary information such as a YouTube video demonstration of how a pill cutter (which we intend to exhibit) was used (see here). It is a simple idea but it will hopefully enhance our exhibit and make it more engaging for viewers.