OpenPath

Connect place-based learning experiences http://openpath.me

[openpathme]

Place-Shift the Library

Here is our recent proposal to the Knight Foundation News Challenge, focused on libraries. They ask is "How might we leverage libraries as a platform to build more knowledgeable communities?"

We certainly have thought about the library as an important node in a learner's network, as well as an important community resource. Check out the proposal and tell us what you think!

[openpathme]

Updates on the OP

Over the summer, we did a lot of housekeeping, moving the site secure moving to HTTPS. WebRTC only uses SRTP (unlike some other platforms that only have it as an option), so to match that in other places in the app we moved to HTTPS. HTTPS also improves the user experience--users no longer have to click "accept" all the time for the camera and microphone permissions.  

We also refactored the site with AngularJS and made a series of fixes. We prototyped a drawing feature (so you could point out something going on in the video), but haven't put it up yet since we still have some inconsistencies in connecting videos up.

Since in many use cases OpenPath will be used on mobile network, or at an institution with a restrictive NAT, we deployed a TURN server. That will increase the connection success to somewhere around 95%. 

Richard and Jared wrote up two curricula recently for various proposals. More at that later...

[openpathme]

U.S. Ignite 2014 and Visit to Burlington

This year, the U.S. Ignite Summit was at the Juniper Dome in Sunnyvale, CA, and OpenPath was one of the selected apps for the app showcase. While Richard was out in Sunnyvale, Jared went to Burlington, VT, one of the U.S. Ignite Gigabit cities, to get on the Gig there. We were kindly hosted by Champlain College. Faculty member Erik Esckilsen facilitated our meeting with Chris North, the helpful and knowledgeable Director of Infrastructure and Information Services, as well as the dynamic Dr. Kristin Wolf, the faculty member behind the apiary at the college who was leading an environmental education tour that day. We were tagging along, over the Gig, for our demo and field questions to Kristin from Sunnyvale.

Our test went well--there was a noticeable difference in quality, being on the Gig (both at Juniper and from Champlain). To set it up at Champlain, a switch needed to be changed in the ethernet port to make sure I was using the Gig (i.e., the student network, which is capped at 15MB per device, but that is plenty for a one-to-one demo even in HD).

Here we are, setting it all up:

Unfortunately it was a rainy day up there, so my hopes of getting into a beekeeper suit were thwarted. However, I learned a lot from Kristin's fantastic presentation on bees, and the kids had great questions both about bees and about OpenPath too.

And we got to eat some honey:

So, even though we couldn't show beautiful vistas of the lake, or the social lives of bees, live, it still was valuable. We look forward to going up there in the not too distant future to do some more piloting. They are doing lots of interesting and innovative things at the college, and in Burlington.

[openpathme]

More Code Please

Since we began building OpenPath, there have been some notable improvements to WebRTC. Originally we were using webrtc.io as our WebRTC framework. However, PeerJS has gained momentum, and is simple, stable, and supported. It interoperates amongst browsers and browser versions, and is open source and free. This framework is helping us clean up our API calls and rationalize the backend. This will make the app faster, and also easier for us to build on and other coders to understand. 

Recently, Mozilla also made the announcement that Persona, its simple sign-on service, would not be supported with developers in the future. One of the reasons for the move was that it didn't have a big enough footprint. We have to agree; further, people we tested with were confused by it. Sometimes it didn't work (often the user's email provider wasn't supported for instance). They didn't understand how it worked, who was asking for their personal info, and how their data would be stored (though Persona is designed to give users the freedom to decide what they share with services).

We think other authentication methods will present less friction; so we are migrating to passport.js. It is an authentication middleware for Node.js. It supports all common authentication strategies; we will look for input on which ones we implement as we go along. In case you are wondering, we have been working on a different branch of the code, which we have not deployed to the prod server. Of course, this is all in an effort to be able to pilot OpenPath and help people learn face to place!

[openpathme]

OpenPath.me at UNESCO's Mobile Learning Week

  We were honored to have been invited to UNESCO’s Mobile Learning Week. According to UNESCO, there were “more than 700 participants from over 60 countries” for the third annual conference in Paris. There were hands-on demonstrations of mobile learning content and technology in the Walking Gallery, around 80 breakout presentations and 30+ exhibitions by NGOs, governments and corporations. A fascinating mix of stakeholders.

While I heard some thoughtful speakers both on main stage and in breakout sessions, I found the most value in conversations with delegates from Chad, Sudan, Nigeria, South Korea, China, Egypt, Norway, and many, many other countries. I had naively been considering OpenPath in terms of US audiences, and because of our work with US Ignite projects in gig cities such as Kansas City and Chattanooga, we have been thinking about the sharing of learning paths through low-latency gigabit-speed networks. In describing our work to UNESCO delegates, though, another possibility emerged. These international representatives responded powerfully to the simplicity and ease of the proposal, its democratizing potential. OpenPath is easily accessible, and it gives learners the power to create and share paths immediately.  As result of these conversations, we’re already in discussion with UK and EU partners. I'd like to connect these groups to our partners in Kansas City and Chattanooga. Kansas City shares a path with a Nigerian classroom! Chattanooga’s Hunter Museum shares part of its collection with Egypt.

  As one might expect, MLW2014 delegates were focused predominantly on ways to extend learning to populations that have little or no access to opportunities. As the recent  UNESCO Global Monitoring Report indicates, “250 million students worldwide cannot read, write or count, even after four years of school. Close to 775 million adults – 64% of whom are women – still lack reading and writing skills, with the lowest rates in Sub-Saharan Africa and South and West Asia.” In light of this powerful research, we’re forced to reconsider learning. What are the fundamental ways to open access to educational resources to more people?  It follows then that at MWL2014 many of the pilots and demonstrations concerned content delivery, skills development and instruction--language acquisition, literacy, numeracy. In many ways, this makes good sense, despite a similarity to old-school student = container models. How do we develop the basic literacy skills that are the foundation from which learning grows? While these “fundamentals” are critically essential, there is also clearly room for resources like OpenPath to help learners construct and share knowledge in informal contexts in and throughout their own worlds.

UNESCO's Future of Mobile Learning

http://unesdoc.unesco.org/images/0021/002196/219637e.pdf

[openpathme]

Set Yourself Up for Success

In most of our use cases, including virtual field trips, bandwidth and latency are real issues. In one of our use cases, where a museum docent, or naturalist, or other expert roams untethered, streaming video over wifi, there is a latency floor due to wifi. On our Gigabit call earlier in December, we discussed a "latency" budget.  

For anyone trying to develop WebRTC-related services, Google Chrome provides a look under the hood of all open peer-to-peer connections, SDP (session description protocol) descriptions, and more via chrome://webrtc-internals.

This, of course, is only part of the latency story. Add a peer, and your requirements for bandwidth, as well as CPU, GPU, and processing of memory increase. Maybe one device is a desktop, and one a wearable. Or, though the hardware is HD, the processing of the streams needs to keep up. Should we want to add a WebGL overlay to the getUserMedia stream and get a bit more creative, we tax local system resources, affecting everyone else's experience. Knowing this, I take to heart yogi Dan O'Sullivan's mantra: "set yourself up for success."

[openpathme]

Connected Doing and Being

Team member Richard Scullin is conducting a series this month with thought leaders in mobile learning, "The Possibilities of Mobile: Openly Networked, Inquiry-Based Learning." The discussions have been really great, i.e., Chris Emdin "Mobile devices become a portal for transforming 'place' into a learning space" and Steve Vosloo, "It's not so much that devices are location-aware, but they make the *user* location-aware" and Jack's comment that "Students also need mentors to open doors for youth to see other learning possibilities."

We also are taking a look at research that has been done on "not school"/informal learning experience. Budget and time constraints often limit the amount of time students get to experience museums and parks. Often, there is little time to prepare learners, and again little time to reflect. Orientation videos before visits, clear learning objectives, a combination of structured and unstructured time, and ties to curriculum all help students get the most out of these experiences. We've been taking a look at some research, including Jennifer Dewitt's research on this subject.

Crosby Kemper at Kansas City Public Library mentioned the Crystal Bridges study that links improved critical thinking (and more) with field trips to the art museum. Crosby also mentioned these interviews, which are really fun! Imagine being able to do this with a classroom or classrooms, make it interactive, and have students in a rural school district ask questions. Take the logistical friction out of the equation. Then take the recording and remix it and apply it to another visit they made with a docent at the Museum of Flight, then study lift at this interactive of lift.

What happens when learners get a steady mix of in- and out-of-school experience, both real and virtual? How can those experiences build over time? What happens when those experiences are recorded, and able to be reviewed and remixed? What are the learning outcomes/how are they different? We are looking for some research on that, and to conduct our own...if anyone has some references to check out, please send them our way. Also, we continue to build and are doing some debugging. If you are interested in trying out the latest in an alpha pilot, let's talk!

[openpathme]

Mozilla PopcornMaker and Glass

One of the first prototype paths we constructed was a very literal one, the Fort Greene Tree Trail in Brooklyn. 39 varieties of trees live along the trail, which undulates along the hill that once supported forts for the Revolutionary War and the War of 1812.

With most field trips, learners get some background information before they arrive at the site. Often, a teacher or facilitator will link that information to curriculum they are currently studying. While at the site, the experience is structured and facilitated with combinations of open-ended prompts, worksheets and writing activities, group work, and free exploration. After a field trip, reflection, and the development of each learner's personal take on the experience, makes it more meaningful.

We think Mozilla PopcornMaker is a fun tool for reflecting and building upon experiences documented in video: learners can take snippets of a video tour and remix them to answer a question, go deeper into the content presented, restructure an experience, or synthesize experiences and compare them.

So without further ado, here is a prototype Fort Greene Tree Trail "make" using Glass--Gabriel Willow wore the Google Glass device, and we only did the beginning of the trail (we have a lot more video).

[openpathme]

Glass Path Experiments

Since July, I've been experimenting with Google Glass applied to experiential learning. I have had many of the same experiences as others, and find the camera the most useful feature of Glass. If you are interested in showing others what you're passionate about, expert in, or thrilled by, you might find it useful.

We've found that it requires a different mindset. Instead of a photo of a street in Paris, you can inhabit that street and share it as you walk through it: imagine streetview becoming real-time, or being able to choose among views nearby.  It reminds me of the experience old stereoscopic images give (it would be interesting to have 2 cameras, like Vannevar Bush describes in 1945). 

Its connectivity to the Internet and social networks is also promising. However, my hangouts so far suffer from poor quality (mostly bandwidth related--hello Gig?). Another challenge with Glass is getting content off the device; I wanted to share a video I recorded immediately, but it wasn't possible because the device needed to dump it into my Google+ account first. I also would prefer to get it into Youtube directly; I wanted to remix some videos in popcorn.js, but had to download the video from Google+ and then upload to Youtube (both are owned by Google--why do it this way?). There are ways to do a hangout and archive it in Youtube, so I imagine this will get fixed.

With respect to the OpenPath project, Glass integrates well with our philosophy of focusing on content at locations, rather than on the typical headshots in so much educational video that replicates lecturing. Glass videos, to me, have more in common with the dérive and flâneurie. There is great potential for educators and learners in a variety of contexts. We are working on some code to get the video into the browser from Glass, and in the next post I'll talk a little about pathmaking.

[openpathme]

Chattanooga Recap

Chattanooga really rolled out the red carpet for us on demo day! Everyone was extraordinarily warm and enthusiastic. A special thanks to Ken Hays for making introductions for us and coordinating our meetings.

Richard and I co-presented OpenPath at the demo day hosted by the Company Lab (Co.lab) at the Chattanooga Theater Center. We used the Gig connection to present a live demo, with Gabriel Willow and Mr. Turtle (an Eastern Box Turtle), live from Brooklyn. Four Mozilla/Ignite teams presented at demo day, alongside the graduates of the Gigtank incubator program. It was great to see Sheldon again and see all the hard work the team puts into the companies in the program.

Following the presentation, we moved to the Chattanooga Public Library to hold an informal workshop with representatives of local cultural and learning institutions. Representatives from the Library, Engage3D, the Hunter Museum, the Tennessee Aquarium, the Children's Discovery Museum, the Chattanooga Zoo, Normal Park Elementary, Bessie Smith Cultural Center, the Kansas City Public Library, Mozilla, and the Hive Learning Network were in attendance. We wanted to find out how we might leverage what is already happening at institutions, fulfill a need, or otherwise augment programs they have. We also wanted to see if what we have built is useful, and get input on what to prioritize in future builds.

We engaged in a broad discussion about place-based activities and online learning. Some of the issues brought up included the difference between what we are doing and videoconferencing. Is it just that it is a better UI, with more modern technology? How would it change learning and pedagogy? This led to some good discussion. Where is the fun? What are paths? What could they be? What do we do when there is no Internet? We tried to learn as much as we could about their current practices and challenges. We were not able to have them fill out our informal surveys, but will do that soon. The hour went really fast, so we barely scratched the surface.

One thing that we kept running into is that, even in a Gig city, a fast wifi connection is still hard to find. It is a concern for some of our use cases, i.e., tours of exhibits and collections, outdoor walking tours, etc. We also heard that there isn't a lot of technical support at some institutions, and obviously everyone is time-constrained! 

After all of that, it was time to try out the Tennessee whiskey! (Bob Metcalfe above) The next day we jogged along the revitalized riverfront, visited the amazing Tennessee Aquarium, and caught the amazing Whitfield Lovell exhibit at the Hunter Museum. There was some wifi signal in both museums, so I tried out OpenPath when possible, looking at what kind of wifi signal strength existed. Looking forward to designing some pilots with the wonderful people we met! (some great turtles at the TN Aquarium BTW)

[openpathme]

Prepping for the Gig in Chattanooga

We are really excited to be demoing OpenPath in Chattanooga, Tennessee on August 6. We'll be part of demo day at CO.LAB (the Company Lab), presenting OpenPath on the amazing Gigabit network they have in Chattanooga going to every one of the community's 150,000 homes and businesses. We are looking forward to talking to representatives of the city's cultural and educational institutions as well, including the fabulous Chattanooga public library, the Tennessee Aquarium, the Hunter Museum of American Art, and more.

We're hoping to share our excitement for OpenPath as well as learn from representatives of these institutions and get feedback from the investors, creators, and mentors in the larger ecosystem there. We are going to explore with them how they might envision creating paths through their city, or link to institutions in other cities, via paths around topics.

Recently, I participated in the Connected Learning MOOC (#clmooc), and ran across Tube Map Me, which has been turned into a cool Thimble project by Chad Sansing. This might be a fun thing to do with some of the representatives of educational/cultural institutions as well. Lots of work to do in the interim!

[openpathme]

OpenPath at Mozilla/U.S. Ignite 2013 App Expo

The last few weeks have been a lot of work, but it all came together on Tuesday when we demonstrated OpenPath in Chicago at the U.S. Ignite application summit. The ideas of virtual field trips and facilitated place-based learning resonated with a lot of people. The ConnectED initiative is certainly paving the way for the high-quality, low-latency interactions we envision. Imagine you are a student in a rural school, taking a tour of the Louvre Museum led been a docent--or a student writing about turtle species, who gets remote facilitation over a mobile device with your teacher while in the local conservation area. 

Speaking of turtles, we had Frodo the Eastern Box Turtle at the Prospect Park Audubon Center hang out at the App Expo live, via WebRTC. Users could see the map of the location of the feed, and interact with Gabriel Willow a staff naturalist at the Center. In 2011, the Eastern box turtle was downgraded to "vulnerable" from "near-threatened." 

Lara Kohl, video artist and adjunct professor at Pratt Institute, showed Marina Zurkow's Slurb, which demonstrated that when there is bandwidth, even highly detailed and motion-filled subject matter can be streamed.

We were able to use OpenPath on the mobile phone as well--WebRTC video sessions now work on the Android mobile browser (that grass in the main window is coming from the Android phone). It was rather amazing when we turned the flag on and it started working on our Nexus 4s!

We still have a ton of work to do, and audio especially needs to be sorted out, but we are excited to do some upcoming pilots with users. We are also planning a path-making workshop for educators and learners in which they will connect learning around topics they are interested in. We have had a ton of positive responses from museums, libraries, and educators, and will give a brief update on those pilots soon.

[openpathme]

Ignition!

We made if to the final round of the Mozilla Ignite competition! We will be demoing OpenPath at the U.S. Ignite Summit in Chicago on June 25. We'll be among the other really exciting projects by the super talented teams that have been participating. 

WebRTC has also been getting more sophisticated; this video from Google 's I/O 2013 conference highlights some of the advanced that have been made since we started with it. Time to get coding!

[openpathme]

OpenPath at the Organ

Part of our Round 3 proposal was using OpenPath in a number of pilots. One of them was to stream a concert of a historic pipe organ as part of the New Museum Ideas City Festival from New York to some people in Massachusetts. The multimedia performance turned the 1868 pipe organ inside out, using HD cameras and mobile phones located inside the 2500-pipe organ to record the movements on the instrument's sculptural mechanisms. These video feeds were then mixed and projected onto the walls of the 204-year-old structure. We wanted to test the limitations of using a laptop in such a scenario, and see what issues would come up under these less than ideal conditions.

Our set up: we found it hard to pick up the projections in the dark room

Pedal squares of organ and mobile phone streaming via a router

Projection of Great division trackers

Additional computation for video feeds via MaxMSP/Jitter

We also had some feedback in the set up we had. This noise cancellation has been an issue, though WebRTC has improved this recently and we it has improved in the latest update. We also have taken a look at adapter.js to help manage the updates as WebRTC improves. As with all real-life pilots, we learned we have some more work to do, and under what conditions (good light, audio close to computer rather than distant) we can expect things to work and the experience to be satisfying. With respect to bandwidth, it was good, since we were only streaming peer-to-peer. We are doing other tests with that in mind. One thing we could have done is take a separate audio feed nearer the organ, choir, and instrumentalists and sent that through a good A/D to the computer and Chrome to OpenPath. This article explains that a bit--good stuff.

[openpathme]

This is an interesting non-scientific study of how museums are currently using technology in their programs. It would be great to find out over the last 3-5 years how their approaches using mobile apps, games, and badging in their programs have changed, and also where they are putting resources in the future. What other ways could their programming be leveraged to a broader audience? What content and learning goals do they establish to measure the success of a program? In what ways has technology benefitted learners and helped fulfill the core mission of the museum? How could programs offered in one museum build on programs in another (and are they doing that already)?

[openpathme]

Need for Speed

Last week, we worked on fleshing out the API and requirements to start building "paths." So along with Users, Sessions, and Events, we now have "Paths." To build some demo paths, we are teaming up with Hive members in a series of workshops (more on this later). The paths will be included in our Web app.

We had a hangout scheduled to talk to Chris Lawrence at the Hive regarding all of this. Since I was finishing up a WildLab class in Central Park with students from City as School, I took out the Google+ app. I had seemingly great 4G reception in the park. However, the truth was they could all hear me about 33% of the time, and the visuals were worthless (and I turned myself into an icon to save already constrained bandwidth for audio). Luckily, we all had a sense of humor about it! Onward network programmability...

[openpathme]

Integrating Frontend and Backend

This past week we continued to build out the API and database, while constructing the frontend. We are demoing OpenPath this weekend for some educators and learners, so we are churning out the code and debugging as fast as possible. 

We migrated the code to an Amazon EC2 instance. The current hosting service had issues with MongoDB that resulted in unacceptable lags using the Persona login. We are very concerned about performance based on our interviews with learners: they don't use videochat because it is generally clunky, slow, and of poor quality.

became:

    Node.js is our server, which can handle up to 10,000 concurrent sessions, making it highly scalable. Continuing with the Javascript, we are using the Express Web application framework, as well as Swig, a browser-based template engine. It took a little time to get used to, but we're making progress.