The Future of Hardware, Software, and The Physical World – Takeaways from O’Reilly Solid 2014

Last week, I attended the O’Reilly Solid Conference, focusing on solid – the intersection of hardware + software + physical things. While hardware and software have existed for decades and are mature in their own rights, it is often the intersection of disparate things that creates innovation though collisions of innovation. And so over 1,400  curious people gathered together to collide.

I wasn’t sure what to expect of conference.  I am passionate about hardware + software and the new possibilities for interactions and solutions that it enables.  And while wasn’t exactly what I thought it would be, it surprised, delighted, and taught me, leaving me hopeful and inspired in the future of solid.

While I’ll share some focused takeaways and observations in the remainder of this post, perhaps the thing that made the biggest impression on me was the diversity of the attendees and speakers. From professors to mechanical engineers to startup founders to artists to hackers to industrial designers to the just interested, solid is well poised to collide ideas to enable new and unexpected outcomes. It will be interesting to see the collisions it creates in the future, as well as who composes the community in the coming years.

After attending a diverse set of sessions and chatting with many attendees, the following takeaways standout for me:

  1. Don’t make a better solution for an unnecessary process. Focus on the user’s desired outcome and make a solution for that. Hardware + software enables new solutions that were hard or impossible previously.  Think about how things should work, not how they do work. This may mean remaking or even eliminating a process. Disruptive change.
  2. Hardware should know us. And it shouldn’t require a separate identity for each device. It should know our preferences and learn about us, no matter where we are. We need single sign on for the physical world. We should be able to interact with the whole world, not just a lightbulb in our living room.  The entire would should be a continuous interaction.
  3. Making is still too hard, but lots of people and companies are working hard to make it easier.  There is a huge complexity difference between 3D printing and CNC. Additional complexity is free in 3D printing while it’s not in CNC due to extra fixturing, rotation, etc. Also new making machines and materials are enabling new solutions like combo CAD / CAM / CNC / Laser / 3D printing all in one.
  4. Most remaining problems in the world are physical and require a physical solution, be that hardware + software / etc, to solve them. Software can partially solve it, but can’t bring a full solution to bear alone.
  5. We have tons of gadgets that don’t interact with each other. They need to be able to interact. Users want outcomes not brand specific ecosystems and proprietary closed solutions.
  6. Prototypes spark conversations. Build hardware prototypes fast, fail fast, fail often = more time to iterate and make it great. Sometimes the prototype is just a conversation starter – a way to get other people sharing their perspectives, even if they drastically change your initial direction.  Also, firmware can be used to dramatically iterate a physical product.
  7. You are the button. Hardware enables new methods of interaction. Maybe it’s where you are. Maybe it’s where you aren’t. Maybe it’s where you’re pointing. Maybe it’s who you’re with. Maybe it’s tapping one device to another. Maybe it’s flicking content from one device to another. Reimagine. Be playful. Some things will stick, the rest will be fun.
  8. Hardware can be soft. Just because something is physical doesn’t mean it must be rigid and frozen. Fabrics and flexible materials enable soft hardware, much like humans and animals with skin, mussels, and bone.  And often these are simpler to implement.
  9. Some promising tools and services:
    • Temboo: A service for orchestrating the Internet of Things, offering integrations with 100s of APIs and 1000s of API Calls.  Helps to limit the resources needed on limited resource embedded devices.
    • Mellow: A home sous vide machine for cooking food, design more like the Nest in terms of thinking about real use cases you care about.  Examples: It keeps food cold and starts it later in the day so it’s ready when you arrive home.  It asks you how you liked your meal and learn about you to better suite your needs in the future.  I love things the speak, and Mellow is a great example.
    • Node Red: A visual editor for Node.js to orchestrate interactions between the Internet of Things.
    • MQTT: A broker for publishing and subscribing to data using a lightweight protocol, limiting the resources needed on limited resource embedded devices.
    • noam.io: A service for rapid hardware prototyping through easy data storage.
  10. We’ve come so far, but we are only at the beginning… Some of these things may take years or decades to mature, but the attendees of Soid are leading the way.

-Adam

Better Ski Maps – Tinkurlab: SlopeStyle

Tinkurlab is going spatial!

We’re adding maps and spatial data to our visualization techniques.
Keep reading for the first foray into mapping…ski slopes!

——–

You’ve heard it all before.

“This run would totally be a green out west.” “This resort isn’t as hard as my home mountain.” “A blue here doesn’t equal a blue there.”

Really? Are you sure? The resorts themselves state that trail ratings are subjective to that particular mountain. Green – Blue – Black – Double Black…and what are those RED trails about in Europe? Even snow conditions can change the rating of one particular trail.

Tinkurlab's collection of trail maps from the 2013/2014 season...each one different

Tinkurlab’s collection of trail maps from the 2013/2014 season…each one different

Ski trail maps are works of art; designed by a couple people to make you want to visit the resort and display the number of trails- but how much can a color tell you? If it’s the right color, turns out a whole lot.  Enter Tinkurlab:SlopeStyle.

Tinkurlab_SlopeStyle

maps.tinkurlab.com/slopestyle

Earlier this winter, Adam and I relocated from Washington DC to Denver, and had many more ski resorts available to us within a 2 hour drive. Resources like FreshyMap helped us determine which one to hit up with a quick glance.   With so many choices, I was curious about finding a way to objectively compare resorts and tell which parts of the resorts I wanted to ski.  In ski progression, it’s easy to both stay on terrain that you are better than or find yourself in way over your head.  I wanted a way to know what I was getting into before I took that first turn.

Tinkurlab:Slopestyle lets you visually explore the pitch and slope angle of your favorite ski runs.  With this data, you can easily compare these runs with others at the same resort to find new terrain.  Additionally, since the color-coding is the same across the maps, you can now compare resorts with each other.  Pink is pink, no matter where you are!  The site performs the same on a desktop as on a phone, allowing you to plan your runs from   any type of chair you’re sitting on.   Happy skiing!  If interested, please read further for the technical details on how the site was created.

Built by processing USGS elevation data with GIS commands to calculate and thematically render slope pitch in degrees.

What I needed to start was elevation data.  I found what I was looking for thanks to the USGS.  They publish layers in varying degrees of resolution and are available for download using The National Map.  I selected the files for the sections of Colorado and California that I required, and set about processing the data.

I used the GDAL utility as my primary GIS tool.  It’s open source, and allows the user to execute a variety of commands for translating and processing raster data.  Using the command-line operators, I was able to create greyscale hillshade layers and calculate the slope for the terrain I was interested in.  I then selected break points for the slope degrees, and generated a color-reilef map based on the pitch.

Example commands:
gdaldem hillshade -s 111120 DEM.adf hillshade.tif
gdaldem slope -s 111120 DEM.adf slopedegree.tif
gdaldem color-relief slopedegree.tif slope-ramp.txt slopeshade.tif -alpha -co ALPHA=YES
Vail Hillshade TIFF

Vail Hillshade TIFF

Vail Slopeshade TIFF

Vail Slopeshade TIFF

Ski resort data overlaid using OpenStreetMap exports stored in PostGIS databases.

The next step was to get the trail map data.  My initial plan was to find a web service containing the features and add it to my map.  That search did not yield great results, so I switched to plan B and learned how to use OpenStreetMap exports.

I used JOSM to extract the aerialway, downhill and woods features for the ski resorts included in my beta site.   Once I had the .osm data, I used osm2pgsql commands to add the features to a PostGIS extended PostgreSQL database.   I found the LearnOSM tutorial extremely helpful (I’m using a Mac with OSX Mavericks, and was able to follow along with the Windows prompts.)

Visualized using Mapbox TileMill and Mapbox Javascript API.

Once I had the data downloaded and processed, it was time to put it all together and create some maps.  I chose to use Mapbox TileMill to combine my layers and create map tiles.  The Carto CSS-like language enabled styling control, and the built in raster and PostGIS support made adding data incredibly easy.

slopestyle-tilemill

Slopestyle TileMill Interface

The tile packages were uploaded to the Tinkurlab Mapbox account, and I set about creating the interactive web map using their javascript API.   The examples are straightforward, and with a couple lines of code, I was up and running.

SlopeStyle - Mapbox Javascript API

SlopeStyle – Mapbox Javascript API

Packaged in a responsive website.

It was important to house the webmap in code that would render correctly on any device.  You might investigate the resorts on a desktop at home, but on the mountain, you’re going to be on a phone.  The Twitter Bootstrap framework is perfectly suited for this application.  Both Tinkurlab:Maps and SlopeStyle are built using Bootstrap with some custom CSS files.

The application defaults to the I70 corridor resorts in Colorado, but the user has the option to switch to the Lake Tahoe area and back.   I chose to focus on the resorts that we skied the season, but plan on expanding in the future.  Other features in the future might include the ability to select a run to view it’s entire slope profile, as well as rendering the slope colors based on your personal ski preferences.  Keep checking back for updates!

– Val

TinkurCrate – Where’s The Dog?!

Overview

One of my favorite things about the Internet of Things is “giving a voice to things”.  And who’s more deserving of a voice – and a Twitter account – than our roommate’s dog, Tyr.  Tyr, an English Springer Spaniel enjoys running, playing with his toys, and hanging out in his crate.  In fact, he’s got a fancy crate located in a prime location in our living room.  TinkurCrate is the first in a series of projects to learn more about Tyr’s activities and connect him to the IoT.  TinkurCrate uses an Arduino with Ethernet and a proximity sensor to determine when Tyr is in his crate.  The Arduino posts the data to Xively for storage, which sends Web Hook triggers to Xively when Tyr enters and exits his crate – triggering Tweets on TinkurLab’s IO Twitter account – @TinkurLabIO.

 

Parts

 

Wiring

Wiring for the project is pretty simple, just connect the proximity sensor to the Arduino Ethernet shield as follows:

  • Red to 5v
  • Black to Ground
  • Yellow to Analog Pin 5

TinkurCrate Wiring

 

Code

<br />/*<br /><br />Published:  2014<br />Author:     Adam - TINKURLAB<br />Web:        www.TinkurLab.com<br /><br />Copyright:  This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. http://creativecommons.org/licenses/by-nc/4.0/<br /><br />Contributors:<br />-Xively code based on http://arduino.cc/en/Tutorial/XivelyClient<br />-Median library http://playground.arduino.cc/Main/RunningMedian<br /><br /><%%KEEPWHITESPACE%%> */<br /><br />#include &lt;SPI.h&gt;<br />#include &lt;Ethernet.h&gt;<br />#include "RunningMedian.h"<br /><br />#define APIKEY         "xxxxxxxxxxx" // replace your Xively API Key here<br />#define FEEDID         xxxxxxxxxxx // replace your Xively Feed ID here<br />#define USERAGENT      "TinkurCrate" // user agent is the project name<br /><br />// assign a MAC address for the ethernet controller.<br />// Newer Ethernet shields have a MAC address printed on a sticker on the shield<br />// fill in your address here:<br />byte mac[] = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }; // replace your Mac Address here<br /><br />// fill in an available IP address on your network here,<br />// for manual configuration:<br />//IPAddress ip(10,0,1,20);<br />// initialize the library instance:<br />EthernetClient client;<br /><br />char server[] = "api.xively.com";   // name address for xively API<br /><br />unsigned long lastConnectionTime = 0;          // last time you connected to the server, in milliseconds<br />boolean lastConnected = false;                 // state of the connection last time through the main loop<br />const unsigned long postingInterval = 10*1000; //delay between updates to Xively.com<br /><br />int modeSwitch = 1;<br /><br />int incrate = 0;<br /><br />RunningMedian proximityLast10 = RunningMedian(10);<br /><br />int timesincrate = 0;<br /><br />int sensorReading = 0;<br /><br />int sensorReadingMedian = 0;<br /><br />void setup() {<br /><%%KEEPWHITESPACE%%>  // Open serial communications and wait for port to open:<br /><%%KEEPWHITESPACE%%>  Serial.begin(9600);<br /><br /><%%KEEPWHITESPACE%%>  delay(2000);<br /><br /><%%KEEPWHITESPACE%%> // Connect to network amd obtain an IP address using DHCP<br /><%%KEEPWHITESPACE%%>  if (Ethernet.begin(mac) == 0)<br /><%%KEEPWHITESPACE%%>  {<br /><%%KEEPWHITESPACE%%>    Serial.println("DHCP Failed, reset Arduino to try again");<br /><%%KEEPWHITESPACE%%>    Serial.println();<br /><%%KEEPWHITESPACE%%>  }<br /><%%KEEPWHITESPACE%%>  else<br /><%%KEEPWHITESPACE%%>  {<br /><%%KEEPWHITESPACE%%>    Serial.println("Arduino connected to network using DHCP");<br /><%%KEEPWHITESPACE%%>    Serial.println();<br /><%%KEEPWHITESPACE%%>  }<br />}<br /><br />void loop() {<br /><br /><%%KEEPWHITESPACE%%>  if (modeSwitch == 1)<br /><%%KEEPWHITESPACE%%>  {<br /><%%KEEPWHITESPACE%%>    // read the analog sensor:<br /><%%KEEPWHITESPACE%%>    sensorReading = analogRead(A5);<br /><br /><%%KEEPWHITESPACE%%>    proximityLast10.add(sensorReading);<br /><br /><%%KEEPWHITESPACE%%>    sensorReadingMedian = proximityLast10.getMedian();<br /><br /><%%KEEPWHITESPACE%%>    Serial.println();<br /><br /><%%KEEPWHITESPACE%%>    Serial.print("Proximity: ");<br /><%%KEEPWHITESPACE%%>    Serial.println(sensorReading);<br /><br /><%%KEEPWHITESPACE%%>    Serial.print("Median Proximity: ");<br /><%%KEEPWHITESPACE%%>    Serial.print(sensorReadingMedian);<br /><%%KEEPWHITESPACE%%>    Serial.print(" w/ ");<br /><%%KEEPWHITESPACE%%>    Serial.print(proximityLast10.getCount());<br /><%%KEEPWHITESPACE%%>    Serial.println(" samples");<br /><br /><%%KEEPWHITESPACE%%>    delay(1000);<br /><br /><%%KEEPWHITESPACE%%>    if (sensorReadingMedian &gt; 160)<br /><%%KEEPWHITESPACE%%>    {<br /><%%KEEPWHITESPACE%%>      incrate = 1;<br /><%%KEEPWHITESPACE%%>    }<br /><br /><%%KEEPWHITESPACE%%>    if (sensorReadingMedian &lt; 100)<br /><%%KEEPWHITESPACE%%>    {<br /><%%KEEPWHITESPACE%%>      incrate = 1;<br /><%%KEEPWHITESPACE%%>    }<br /><br /><%%KEEPWHITESPACE%%>    Serial.print("Is In Crate: ");<br /><%%KEEPWHITESPACE%%>    Serial.println(incrate);<br /><br /><%%KEEPWHITESPACE%%>    Serial.println();<br /><br /><%%KEEPWHITESPACE%%>  }<br /><br /><%%KEEPWHITESPACE%%>  // convert the data to a String<br /><%%KEEPWHITESPACE%%>  String dataString = "proximity,";<br /><%%KEEPWHITESPACE%%>  dataString += String(sensorReadingMedian);<br /><br /><%%KEEPWHITESPACE%%>  // you can append multiple readings to this String to<br /><%%KEEPWHITESPACE%%>  // send the xively feed multiple values<br /><%%KEEPWHITESPACE%%>  dataString += "\nincrate,";<br /><%%KEEPWHITESPACE%%>  dataString += String(incrate);<br /><br /><%%KEEPWHITESPACE%%>  // if there's incoming data from the net connection.<br /><%%KEEPWHITESPACE%%>  // send it out the serial port.  This is for debugging<br /><%%KEEPWHITESPACE%%>  // purposes only:<br /><%%KEEPWHITESPACE%%>  if (client.available()) {<br /><%%KEEPWHITESPACE%%>    char c = client.read();<br /><%%KEEPWHITESPACE%%>    Serial.print(c);<br /><%%KEEPWHITESPACE%%>  }<br /><br /><%%KEEPWHITESPACE%%>  // if there's no net connection, but there was one last time<br /><%%KEEPWHITESPACE%%>  // through the loop, then stop the client:<br /><%%KEEPWHITESPACE%%>  if (!client.connected() &amp;&amp; lastConnected) {<br /><%%KEEPWHITESPACE%%>    Serial.println();<br /><%%KEEPWHITESPACE%%>    Serial.println("disconnecting.");<br /><br /><%%KEEPWHITESPACE%%>    resetMode();<br /><br /><%%KEEPWHITESPACE%%>    client.stop();<br /><br /><%%KEEPWHITESPACE%%>  }<br /><br /><%%KEEPWHITESPACE%%>  // if you're not connected, and ten seconds have passed since<br /><%%KEEPWHITESPACE%%>  // your last connection, then connect again and send data:<br /><%%KEEPWHITESPACE%%>  if(!client.connected() &amp;&amp; (millis() - lastConnectionTime &gt; postingInterval)) {<br /><br /><%%KEEPWHITESPACE%%>    modeSwitch = 2;<br /><br /><%%KEEPWHITESPACE%%>    sendData(dataString);<br /><%%KEEPWHITESPACE%%>  }<br /><%%KEEPWHITESPACE%%>  // store the state of the connection for next time through<br /><%%KEEPWHITESPACE%%>  // the loop:<br /><%%KEEPWHITESPACE%%>  lastConnected = client.connected();<br /><br />}<br /><br />// this method makes a HTTP connection to the server:<br />void sendData(String thisData) {<br /><br /><%%KEEPWHITESPACE%%>  // if there's a successful connection:<br /><%%KEEPWHITESPACE%%>  if (client.connect(server, 80)) {<br /><%%KEEPWHITESPACE%%>    Serial.println("connecting...");<br /><%%KEEPWHITESPACE%%>    // send the HTTP PUT request:<br /><%%KEEPWHITESPACE%%>    client.print("PUT /v2/feeds/");<br /><%%KEEPWHITESPACE%%>    client.print(FEEDID);<br /><%%KEEPWHITESPACE%%>    client.println(".csv HTTP/1.0");<br /><%%KEEPWHITESPACE%%>    client.println("Host: api.xively.com");<br /><%%KEEPWHITESPACE%%>    client.print("X-ApiKey: ");<br /><%%KEEPWHITESPACE%%>    client.println(APIKEY);<br /><%%KEEPWHITESPACE%%>    client.print("Content-Length: ");<br /><%%KEEPWHITESPACE%%>    client.println(thisData.length());<br /><br /><%%KEEPWHITESPACE%%>    // last pieces of the HTTP PUT request:<br /><%%KEEPWHITESPACE%%>    client.println("Content-Type: text/csv");<br /><%%KEEPWHITESPACE%%>    client.println("Connection: close");<br /><%%KEEPWHITESPACE%%>    client.println();<br /><br /><%%KEEPWHITESPACE%%>    // here's the actual content of the PUT request:<br /><%%KEEPWHITESPACE%%>    client.println(thisData);<br /><%%KEEPWHITESPACE%%>    Serial.println(thisData);<br /><br /><%%KEEPWHITESPACE%%>    client.println();<br /><br /><%%KEEPWHITESPACE%%>  }<br /><%%KEEPWHITESPACE%%>  else {<br /><%%KEEPWHITESPACE%%>    // if you couldn't make a connection:<br /><%%KEEPWHITESPACE%%>    Serial.println("connection failed");<br /><%%KEEPWHITESPACE%%>    Serial.println();<br /><%%KEEPWHITESPACE%%>    Serial.println("disconnecting.");<br /><%%KEEPWHITESPACE%%>    client.stop();<br /><%%KEEPWHITESPACE%%>  }<br /><%%KEEPWHITESPACE%%>   // note the time that the connection was made or attempted:<br /><%%KEEPWHITESPACE%%>  lastConnectionTime = millis();<br />}<br /><br />void resetMode()<br />{<br /><%%KEEPWHITESPACE%%> modeSwitch = 1;<br /><%%KEEPWHITESPACE%%> incrate = 0;<br />}<br />

Source code also available on GitHub at https://github.com/TinkurLab/TinkurCrate/.

 

Xively Setup

Xively is a service for powering the Internet of Things, providing API access for data storage, data retrieval, triggers, and data charting.  After registering for a free Xively account, perform the following steps:

1) Add Device.

2) Add Channels for “incrate” and “proximity”.

Cursor_and_TinkurDogCrate_-_Xively_1

3) Setup Triggers for “incrate” status change.  One Trigger calls the Zappier “Entering Crate” Zap, and the other Trigger calls the Zappier “Exiting Crate” Zap.

Cursor_and_TinkurDogCrate_-_Xively

 

Zapier Setup

Zapier is a service for orchestrating and automating the Internet of Things and popular online services.  After registering for a free Zapier account, perform the following steps:

1) Create a new Zap, using a Web Hook – Catch Hook event, triggering a Twitter – Crate Tweet event.

Cursor_and_Editor_-_Zapier

2) Authorize your Twitter account if needed, and add a Filter to only trigger when the Xively trigger is “1” (when the dog enters the crate).

Cursor_and_Editor_-_Zapier_2

3) Setup a Message to post to Twitter and save the Zap.

Cursor_and_Editor_-_Zapier_1

You can view Tyr’s live status at https://xively.com/feeds/1765156749 and his Tweets at @TinkurLabIO.

What’s Next

While TinkurCrate is a good beginning, I hope to add a few more sensors to Tyr’s world.  Possibilities include:

  • Pressure sensors to determine when Tyr is on the sofa
  • Continuity sensor to determine when Tyr’s water bowl is empty
  • Processing visualization of Tyr’s in crate / out of crate activity by time of day and day of week

Have any other ideas or suggestions.  Tweet @TinkurLab.

-Adam

Tagged

TinkurBooth Photobooth

I remember the first time I visited my friend Jim at college many years ago.  Upon walking into his small two person dorm room, the first thing I noticed was all the Polaroid photos hanging around his door frame – of smiling friends and a few funny faces.  Jim had a tradition of taking a Polaroid picture when someone new visited him – a way to remember all the great people that one encounters at college.  I’ve always had a love of candid photography – photos the capture the essence of a moment or a person.  Photobooths and Polaroids are great tools for capturing candid pictures – they’re easy, quick, and produce a tangible result.

TinkurBooth Prototype

A basic prototype of TinkurBooth.

While I still love the Four Frames Photobooth, it takes a bit of time to transport, set up, and take down.  Although Four Frames Photobooth has had a busy life – attending multiple weddings, a few business parties, and even a happy hour – I wanted something the would be, well, easier, quicker, and a ‘lil bit more playful.  Meet TinkurBooth – a platform for taking quick and candid photos with endless possibilities for innovation.

TinkurBooth was my first project using a Rasberry Pi.  Many of TinkurLab’s creations have already used microcontrollers such as the Arduino.  However, I’ve been looking for an excuse to try a Raspberry Pi.  Not only are Pis cheap ($30-40) the Pi Camera module is also cheap ($30) and you can use a cheap ($10) wi-fi adapter for connectivity – < $100 for a photobooth is a pretty good deal.  And honestly, there’s just something geekily awesome about taking pictures using shell commands and switches!

“And honestly, there’s just something geekily awesome about taking pictures using shell commands and switches!”

I wanted TinkurBooth to be more then a “once and done” photobooth.  I wanted it to be a platform for experimentation and play – for trying new ideas.  Coming from my day job as an agile leader and coach, I am often reminded about the countless opportunities for experimentation – to try new things, test a hypothesis, and validate the outcome.  The TinkurBooth experience is built around four steps, which allow for many possibilities: Trigger + Interaction + Capture + Sharing.

 

Trigger + Interaction + Capture + Sharing

“Aren’t you going a bit too far?  It’s just a photobooth!” is what you’re probably thinking right now.  Let me explain.  Thinking of each of these steps as separate but related, helps me think about the possibilities of changing one or more of the steps to create a different experience that engages different people in different ways.  For example, one day the TinkurBooth could use a motion trigger while another day it could use a sound trigger that listens for a clap.  As another example, TinkurBooth could capture a single black and white photo one day while another day it could take four photos and merge them into an animated GIF.  In fact, my goal for the next year is to have a new version of TinkurBooth every month or two to see how each variation changes the user experience.

 

Trigger

This step is about the system knowing the user wants to interact with it (or convincing the user to interact with it).  Examples of triggers could include:

  • Button: Pressing a button

  • Motion: Passing within a certain area (width and depth)

  • Sound: A loud sound or a certain type of sound (ex. a clap or whistle)

  • Light: A dramatic change in light, indicating something has changed

  • Distance and Movement: Coming within a certain distance

  • Time: A certain amount of elapsed time

  • Proximity: Being within a certain distance, although not necessarily within visual proximity (ex. think about using geo location)

  • Something You Have: Needing a smartphone to interact with it

  • Pattern: Press a button in a certain pattern

And let’s not forget about the possibility of mashups such as Distance + Time.  For example, an early prototype of TinkurBooth required the user to be standing within a 2 foot area before it would take a picture.  Too close and it would yell at you to stand back.  Too far away and it would start to worry and ask you to come back.  Really far away and it would ask you to come play.  And if you stood in just the right spot for two seconds, with would take your picture.

 

Interaction

While the trigger could directly move to the capture step, there’s a great opportunity for interaction at this point to engage and delight the user.  Examples of interaction could include:

 

  • Engage: In the Distance + Time example above, the photobooth is playful, using an LCD screen with text based on distance and time to engage the user.

  • Inform: The LCD screen could also be used to tell the user how to interact with the photobooth.  In the above example, before starting taking pictures, the photobooth asked the user to make a funny face and then showed the user a random word such as “crazy” just before taking the picture.

  • Give and Take: Interaction could also consist of the user providing something, such as their Twitter username, in return for the photobooth taking a picture.  The photobooth could then use the Twitter username to tag the photo.

 

Capture

The photography step of the equation.  Examples of capture could include:

  • Timing: The time between pictures or a random time

  • Number of Pictures: The number of pictures taken or a random number

  • Shape of Pictures: Vertical, horizontal, square

  • Filters: Color, B&W, think Instagram

  • Lighting: Flash, No Flash, Ringflash

  • Post Processing: Keep pictures separate, merge photos into a 4 x 1 strip, merge photos into an animated GIF

 

Sharing

One of the greatest challenges with photography is doing something with all those pictures.  Sharing is the part of the equation that allows people to share and remember moments.  Examples of sharing could include:

  • Twitter: Posting to Twitter

  • Printing: Printing a photostrip

  • Emailing: Emailing to the user

 

Prototype

However, before experimenting with variations, I had to create a working prototype.  There are lots of great tutorials and troubleshooting posts all over the Internet, so I won’t provide step by step instructions here.  However, if you have specific questions, feel free to contact me.

One of the first animated GIF examples posted to Tumblr http://tinkurbooth.tumblr.com/ by TinkurBooth.

How It Works

The following is an overview of the TinkurBooth Platform workflow:

  1. Run sudo python boothsnap.py script

  2. Script monitors for motion

  3. When motion is detected, four pictures are taken using the raspistill command and saved locally; code based on https://gist.github.com/benhosmer/5653641

  4. The four photos are merged together using ImageMagick and saved locally as an animated GIF

  5. The annimated GIF is uploaded to a Gmail account; code based on http://mitchtech.net/connect-raspberry-pi-to-gmail-facebook-twitter-more/

  6. The If Then Then That (IFTTT) service monitors the Gmail account and runs macros to post the animated GIF to Tumblr (and thanks for Tumblr for being so awesome as to actually support annimated GIFs!)

 

Parts

 

Build

  1. Follow the awesome Adafruit Raspberry Pi Tutorials

    1. Prepare SD Card and Install Raspbian OS ://learn.adafruit.com/adafruit-raspberry-pi-lesson-1-preparing-and-sd-card-for-your-raspberry-pi

    2. Configure Pi http://learn.adafruit.com/adafruits-raspberry-pi-lesson-2-first-time-configuration

    3. Configure Network http://learn.adafruit.com/adafruits-raspberry-pi-lesson-3-network-setup

    4. Configure VNC (so you can program from yoru desktop or laptop) http://learn.adafruit.com/adafruit-raspberry-pi-lesson-7-remote-control-with-vnc

    5. Setup GPIO (General Purpose Input Output) Libraries http://learn.adafruit.com/adafruits-raspberry-pi-lesson-4-gpio-setup

    6. Trying Sensing Some Movement http://learn.adafruit.com/adafruits-raspberry-pi-lesson-12-sensing-movement

  2. Setup and configure the Raspberry Pi Camera http://raspberrypi.werquin.com/post/43645313109/initial-setup-of-the-raspberry-pi-out-of-the-box
  3. Download the TinkurBooth source code from GitHub at https://github.com/TinkurLab/TinkurBooth

 

Helpful Tutorials

 

What’s Next

Stay tuned for future posts about TinkurBooth.  Forth the 1st “TinkurBooth” of the month, I’m going to be creating a version that is activated by distance and uses an LCD screen to interact with the user asking them to play a game to act out a event that will be turned into an animated GIF.

And if you’re wondering about that Polaroid from Jim’s dorm room, here’s it is – Jim, Adam, and Val.

Jim Adam and Val Polaroid

-Adam

Tagged , ,

Raspberry Pi Camera Command List

Now that I got my Raspberry Pi’s Camera Module working, I’m sure I’ll need to keep a list of the software commands handy.  The following is the list of commands supported by v1.1 for the RaspiStill Camera App:

Use from a command line.  Usage: raspistill [options]

Image Parameter Commands

-?, –help : This help information
-w, –width : Set image width <size>
-h, –height : Set image height <size>
-q, –quality : Set jpeg quality <0 to 100>
-r, –raw : Add raw bayer data to jpeg metadata
-o, –output : Output filename <filename> (to write to stdout, use ‘-o -‘). If not specified, no file is saved
-v, –verbose : Output verbose information during run
-t, –timeout : Time (in ms) before takes picture and shuts down (if not specified, set to 5s)
-th, –thumb : Set thumbnail parameters (x:y:quality)
-d, –demo : Run a demo mode (cycle through range of camera options, no capture)
-e, –encoding : Encoding to use for output file (jpg, bmp, gif, png)
-x, –exif : EXIF tag to apply to captures (format as ‘key=value’)
-tl, –timelapse : Timelapse mode. Takes a picture every <t>ms

Preview Parameter Commands

-p, –preview : Preview window settings <‘x,y,w,h’>
-f, –fullscreen : Fullscreen preview mode
-op, –opacity : Preview window opacity (0-255)
-n, –nopreview : Do not display a preview window

Image Parameter Commands

-sh, –sharpness : Set image sharpness (-100 to 100)
-co, –contrast : Set image contrast (-100 to 100)
-br, –brightness : Set image brightness (0 to 100)
-sa, –saturation : Set image saturation (-100 to 100)
-ISO, –ISO : Set capture ISO
-vs, –vstab : Turn on video stablisation
-ev, –ev : Set EV compensation
-ex, –exposure : Set exposure mode (see Notes)
-awb, –awb : Set AWB mode (see Notes)
-ifx, –imxfx : Set image effect (see Notes)
-cfx, –colfx : Set colour effect (U:V)
-mm, –metering : Set metering mode (see Notes)
-rot, –rotation : Set image rotation (0-359)
-hf, –hflip : Set horizontal flip
-vf, –vflip : Set vertical flip

Exposure mode options :
off,auto,night,nightpreview,backlight,spotlight,sports,snow,beach,verylong,fixedfps,antishake,fireworks

AWB mode options :
off,auto,sun,cloud,shade,tungsten,fluorescent,incandescent,flash,horizon

Image Effect mode options :
none,negative,solarise,sketch,denoise,emboss,oilpaint,hatch,gpen,pastel,watercolour,film,blur,saturation,colourswap,washedout,posterise,colourpoint,colourbalance,cartoon

Metering Mode options :
average,spot,backlit,matrix

Tagged , , ,

Sticky Notes – Awesome for Brainstorming and Collaboration

There’s a stereotype out there – that agile teams use a lot of sticky notes.  Well guess what – it’s true!  If you walk around the workspace of almost any agile team, you are likely to see countless colorful sticky notes on walls and windows everywhere.  But why?  In my experience, sticky notes are one of the most powerful and versatile tools for collaboration I’ve ever used – including software tools.  But I’ve also learned it can take a bit of convincing and learning by doing to help others understand the “Power of Post-Its“.  Through a series of blog posts, I hope to turn you into a believer too!

Sticky Notes

As an agile practitioner and coach, I am often facilitating collaborative sessions – to brainstorm, to prioritize, to turn a vision into a plan, to collect feedback, to estimate, to sync on progress.  And my number one tool of choice are sticky notes.  The seemingly simple pieces of paper are a powerful tool to get the whole team to participate and rapidly share their knowledge and perspectives while collaborating towards a shared goal.

While I’ll discuss how to use sticky notes for many different types of collaboration in future posts, let’s start by looking at the many benefits of sticky notes as a tool:

Quick

Capturing a thought on a sticky note is quick.  Just grab a pen or Sharpie marker – my favorite writing implement – and jot down your thought in a few words.  Great!  Now write down another.  And another.  Awesome – you’re rapidly brainstorming!

Compact and Concise

With limited time and attention it’s often important to distill the most important facts down to a few meaningful words or sentences – or a picture.  Sticky notes have a finite amount of space, requiring each note to contain a simple and focused idea.  Can’t fit it all one sticky note?  Good – use a second and third, making sure each item is focused.   I bet you didn’t think less space was a good thing!

Divide and Conquer

Sticky notes allow a whole team of people to brainstorm independently at the same time.  Just ask each person to write down as many ideas as they can within a 5 to 10 minute time box.  Don’t worry about evaluating, discussing, or prioritizing sticky notes yet.  Brainstorming is all about unfiltered idea anyway.  And by asking everyone to work in parallel without discussion, you’ll get lots of ideas written down.

The Sticky Factor

Sticky notes stick to just about anything – walls, windows, desks – especially the Super Sticky Post-It Notes which I highly recommend.  When I’m facilitating collaborative sessions, I often use painter’s or artist’s tape to create an appropriate set of buckets on the wall and then ask team members to add their sticky notes to each area.  For example, while facilitating an team retrospective to determine how to improve the team and their process, I’ll make buckets such as “do more of”, “do less of”, “what if…”.

Visible

One of my favorite reasons for using sticky notes is that they’re colorful and visible. By putting your sticky notes in a public space such as an office wall, you and your team are constantly reminded of the sticky notes and encourage others to check out what’s happening.  It’s a great way to radiate information and invite other to participate in collaboration.

Touchable

You can touch a sticky note.  It’s real.  When you’re discussing it, you can pick it up so others know which thing you’re talking about.  When you’ve made progress, you can move it from an in progress bucket to a done bucket.  When you’re done done with it you can tear it up and throw it away – my favorite!

Simple or Complex

Just like software tools, sticky notes can support just about any process and metadata you can invent. Want to track new features vs. defect – use different colored sticky notes. Want to track how many days something has been in progress – add a dot to the lower left corner of the sticky note each day. Want to track who’s working on what – give each person a unique sticker to add to sticky notes.

Repositionable and Groupable

Combined with a wall or window and some painter’s tape to create some buckets, sticky notes make a great system to track just about anything.  And best of all – they’re built to be moved around over and over and over.  Try reconfiguring your software tools to match your process that quickly!

While I also use many other agile tools, sticky notes are one of my favorite.  Try it out yourself and see if you become a believer!  Or check out some other team’s sticky notes.

-Adam

Tagged , , , , , , ,

What’s the best tool in our hackerspace workshop? A Dremel. Hands down.

Tales of Tools and a Panning Timelapse GoPro Timer.

So what is the best tool in TinkurLab’s workshop?  A Dremel!  Hands down!  Well, 1 minute quick drying epoxy is a close second, but let’s save that for another day.

It can cut.  I can grind.  It can drill.  It can route.  It can sand.  It can polish.  It’s pretty small.  You can get new bits cheap.

So how did we get to this epiphany?  I recently completed a small project to update my timelapse panning timer – affectionately named TinkurLapse.  TinkurLapse rotates my GoPro camera as it takes a photo every few seconds, creating a panning timelapse video.  My original timer was basically a $2 Ikea kitchen timer with a ¼”x20 screw glued to it to screw into the GoPro’s tripod mount.  BTW a standard camera tripod mount takes a – you guessed it – ¼”x20 screw.  Remember that for your next trivia night.  Anyway, in the spirit of most Tinkurlab projects, the goal was to get a minimally viable product out the door quickly to start really learning about it and to make improvements.  “Learning by doing” in other words.

TinkurLapse Panning Time-Lapse Camera Timer

And learn I did.  While the first version of TinkurLapse worked, it had a few issues.  First and foremost, having a tiny base, relatively high height, and very light weight resulted in an unbalanced base for the GoPro which easily fell over.  Many early experiments resulted in setting the camera up for a 60 minute timelapse only to return after an hour to find the camera lying sideways rotating the timer under its base.  Pretty uneventful video! The other issue was its size.  While it wasn’t huge, it wasn’t small either.  Given that most of my GoPro shooting occurs during travels and adventures like hiking, skiing, etc, it would be ideal to make the panning timer as small as possible.  While the first version of the panning timer worked, I didn’t use it very much.  It just wasn’t good enough for me.  I value it provided didn’t overcome the cost of using it.  My goal of creating awesome panning timelapse videos was blocked.

So one day while aimlessly wandering the Internet, I decided to search for a prebuilt device.  I know – it was a moment of weakness.  How un-DIY of me.  There are not very many panning timelapse devices on the market (at least not under less than a few $100s for professional use), but while browsing through the catalog of Photojojo, I came across the Camalapse.  However, aside from the $30 price tag, it has the same problem as the first version of TinkurLapse – not stable enough.  Camalapse is only 2 ounces, with a small base.  A featherweight.  However, it does support mounting to a tripod to provide a more stable base, so it has potential.

After knocking some sense into myself, I set out to research ways to improve the initial design.  I found a great video from a serious panning timelapser, who posted a great tutorial.  The tutorial suggested using the same Ikea kitchen timer I used in the first version of TinkurLapse – no problem, I bought 3x just for this reason.  The tutorial also had some other great ideas – remove the done bell as to not scare all people and wildlife in 1 mile radius, and mounting the timer to a Gorillapod tripod to provide a more stable base.  Perfect!

Enter the Dremel.  After some drilling, grinding, and gluing I had completed version two of TinkurLapse.  On to testing.  The initial test run resulted in rotation of about 90 degrees before stopping.  After some investigation, I determined that excess glue and some of the timer’s parts were obstructing the rotation of the GoPro mount.  What to do?  Dremel it!  After some precise grinding and sanding, I had trimmed down the responsible obstructions.  The next test passed without issue, even removing the annoying done bell at the end.

TinkurLapse v2 is ready for some more real world testing!  I’m looking forward to a few upcoming skiing trips to use the new TinkurLapse.  I’ll post some videos.  Until then, happy tinkering!

Example Timelapse Video 1

Example Timelapse Video 2

-Adam

 

Tagged , , , ,

“Sprint to Success” and Other Agile Stickers and Swag

Another project completed!  What does it do?  It sticks to things and looks cool.  That’s right – stickers!

I haven’t done much design work lately, so I recently fired up my new Adobe CS6 Suite tools to create some cool stickers to promote agile and agile practices.  In addition to working on Tinkurlab projects, I happen to be a Scrum Master for agile software development by day – working on projects to create websites, mobile websites, mobile apps, other interesting things.  And I’m a huge believer in agile principals, methodologies, and tools – from Scrum to Extreme Programming to Kanban to the collective knowledge and experiences of all the great agile doers out there who get stuff done done with agile.  Anyway, here are a few of the stickers; you can check out the full collection in the Cafe Press store (and buy them as well).  Stick them everywhere you can!  Enjoy!

Check out the full collection at http://www.cafepress.com/agileswag.

 

-Adam

 

Tagged , , , , , , , , , , , ,

Das Bot 2.0 – Now with more beer!

How quickly time goes by.  It seems like we only just completed Das Bot v1.0, but with Oktoberfest 2012 quickly approaching we started one of the biggest projects we’ve completed to date – Das Bot 2.0.  Instrumented beer on a grand scale!

The final setup, ready for the party to start.

 

Das Bot is a system for monitoring and controlling access to beer through RFID tags. Built on top of an internet-connected Arduino, the system accepts an RFID tag, checks to make sure you’re registered, opens the solenoid valves in the beer lines, lets you pour a beer and monitors how much you’ve poured, saves the data to a database, and prints out a receipt. Oh, and there’s prize badges you can win as well. There were three beers on tap this year (a homebrewed Dunkelweizen, a homebrewed Hefeweizen, and a Hofbrauhaus Oktoberfest) and each beer was tracked independently.

The Hardware

Das Bot contains the following pieces of hardware:

  • Arduino (with Ethernet shield) – the brains of the operation. This micro controller is what reads the RFID tags, talks to the PHP/MySQL site, controls the solenoid valves, prints to the thermal printer, and reads the data from the flow meters.
  • RFID Reader - reads the values from the RFID chips and sends the value to the Arduino.
  • 3 Solenoid Valves – similar to the ones on Adafruit’s site, these valves are normally closed (restricting access to vital beer) and are only opened when the Arduino gives the go ahead. After 10 seconds of no beer flowing, they close again. Since these draw more current and run at 12v, we used an old 12v wall wart we had laying around to power these (through the relay).
  • RelayThis relay controls the power to the solenoids.
  • 3 Flow Meters – we used the ones from Adafruit to send the flow data to the Arduino.
  • Thermal Printer – printed a welcome message, current stats, and final pour total for each pour.
  • Electrical Box – from the big box home improvement store, this water (and beer-proof) project enclosure was perfect for housing the Arduino, printer, and other electrical components.
  • RGB LED – when the light turns blue, pour your beer
  • Piezzo Buzzer – something this complicated needs to beep. Ours beeps when an RFID is successfully read.
  • Jockey Box – this converted cooler has a 7-pass plate chiller inside and once covered with ice, chills the beer as it flows through. This means no need to keep the kegs in giant plastic tubs with 800lbs of ice.
  • Tap Handles – no jockey box is complete without custom milled tap handles.
  • The Override Switch – this can not be over-appreciated. Buried in an undisclosed location was a switch that would override the arduino and open the solenoid valves. In the event of catastrophic failure, this is the switch represented the difference between beer and no beer. I’m always going to support the side with beer.

Closeup shot of the solenoids

The override switch. Luckily, this wasn’t needed this year. But it’s nice to know it’s there.

 

The Software

The hardware was almost the easy part of this project. We had experience with Das Bot v1.0, so adding some additional sensors wasn’t too difficult. The hard part was getting everything to talk to each other at the right times, with the right data.
Here’s what the software side looked like:
  • Arduino code – this is the air traffic control system. When an RFID chip is read, it sends the data to a web service and gets back the user’s current status. If they’re not registered, they get one (and ONLY one) free beer before they’re required to register. The flow meters are opened and activity from the meters is recorded. After 10 seconds of no flow, the system closes the valves and sends the pour volume data to the web service. The Arduino then prints out a receipt listing pour volumes and any badges that were won on this pour.
  • PHP/MySQL – a website used to store and present data during the party. The web service components were built using some simple PHP with a MySQL back end. The dashboard (running on an iPad taped inside the kitchen window during the party) presented current keg stats, leaderboards, badge summary, and other interesting bits of info. We also included the calibration for the flow meters in the database to allow for different keg sizes and to help more accurately record the data.

The welcome message displayed once an RFID tag is scanned

 The Dashboard

Want to become mayor of the Hofbrauhaus? Brag about having the largest drinking vessel? Why not turn an entire party into a drinking game? With all of the data we were collecting, it seemed natural to present this information back to the user. The thermal printer can only present so much data. The kegs were also under the table, making the “just lift it to see how much is left” method rather difficult. So, we created the dashboard to give a quick way to see how much fun we were having (note to self: next year, add a “fun” meter).

The dashboard presented the current leaders, the amount of beer consumed for each keg, the 5 most recent pours, and some other random stats. Some badges were mysteries, only visible once unlocked. Others were hinted at by the icon. Either way, the best way to win a badge was to drink beer. Actually, that was the only way.

How much fun are we having right now? Oh, THAT much. This was displayed on an iPad safely taped to the inside of the kitchen window. It was visible from the taps.

Future Enhancements

Some ideas for the 2013 Oktoberfest:

  • Integrate the photobooth – haven’t been inside the booth in the last hour? No beer for you!
  • More badges to win – need target different needs in people’s psyche, not just the “I drank more than you did” desire
  • Allow “offline” badges – someone won the beer stein race, but we’ll never remember who it was. We’ll need a way to award spot badges from mobile devices.
  • More notifications when a badge is won – I’m thinking car horns, disco balls, or fireworks. Perhaps all three.
  • Breathalyzer integration – imagine tracking this against the volume consumed overlayed with the photobooth pictures. None of our friends will ever be politicians.

Images

Additional images from the project:

 

RFID chips. Everyone got one. There were also zip ties to attach the chips to your beer stein.

The custom tap handles came out well.

The system, all condensed.

Water and electricity don’t mix, so we taped the electronics inside the cooler to the lid.

Additional photos can be found on Flickr

 

-Matt

 

*Disclaimer – Please note that this system was built for responsible adults. EVERY person in the “leader boards” either spent the night (and likely much of the next day) at the location or were driven home by a designated driver. We might mix water, beer, and electricity together from time to time, but we certainly don’t drink and drive.

Tagged , , , , ,