Lessons learned: living with digital media systems in flexible classrooms

The Taylor Institute’s AV systems were designed to be incredibly flexible, able to adapt to changing requirements between (or even during) classes. That meant shifting from hardwired analog systems to fully digital media management to allow for software-controlled mixing and switching of signals. What people assume, when they walk into a classroom, is something basically …

Continue reading “Lessons learned: living with digital media systems in flexible classrooms”

The Taylor Institute’s AV systems were designed to be incredibly flexible, able to adapt to changing requirements between (or even during) classes. That meant shifting from hardwired analog systems to fully digital media management to allow for software-controlled mixing and switching of signals. What people assume, when they walk into a classroom, is something basically like this: You show up, plug your laptop in, and it sends stuff to the projector. And other stuff to the speakers. Simple. What we have is more like this: That’s an extremely oversimplified representation of the flow. There are many many many steps not shown. This kind of design allows instructors and students to have an incredible level of control over the media in the active learning classroom. They can push a button in the software running on the Crestron panel on the podium, and route images and audio from about a dozen input sources to about that many outputs. When it works, it’s absolutely amazing, and it feels like living in a science fiction classroom of the future. When something goes wrong, however, it can be an incredibly frustrating exercise in troubleshooting. Any of a few dozen steps between input and output could have gone awry.  Often, troubleshooting these steps involves running to the separate IT floor (most of the audiovisual gear is installed on “mezzanine” floor, making it possible to work on the equipment without disrupting classes, but making troubleshooting during classes a bit of a pain because the gear isn’t physically in the room) and doing the emergency turn-it-off-and-back-on thing. We had some fun on Friday, when the audio systems in the Forum stopped sending audio to the speakers at the beginning of a 2-hour class that was designed to rely on the microphones (audio), videos on the instructor’s laptop (audio), and a blu-ray movie (audio). So, three strikes right off the top. There had been a firmware update on a network switch a couple of weeks ago, and that apparently caused a problem that exposed a bug in the audio management software. Something about packet corruption, and network data being interpreted as audio and then being dutifully sent to the speakers. Which led to some incredibly loud moments as the giant speakers suddenly maxed out with white noise during a class. Not ideal. Thankfully, we have a great relationship with the company that installed and configured the systems, and they sent their senior tech to try to troubleshoot (while the class was still in session – rapid response times!). After some serious head-scratching, he talked to one of the vendors and determined that it was a bug in the audio management software, and that there was a beta version of the software that solved that particular bug. So, he installed the beta (on our only production audio management hardware – what could go wrong?) and the problem went away. For now. Thankfully.

Lessons learned: AV systems design in the Taylor Institute

We’ve been in the Taylor Institute for Teaching and Learning‘s new building for almost a year now, and it’s time to step back and reflect on what we’ve learned through that first year. The building itself is a marvel of architecture, design and technology. We’re extremely fortunate to be able to go to work there … Continue reading “Lessons learned: AV systems design in the Taylor Institute”

We’ve been in the Taylor Institute for Teaching and Learning‘s new building for almost a year now, and it’s time to step back and reflect on what we’ve learned through that first year.

The building itself is a marvel of architecture, design and technology. We’re extremely fortunate to be able to go to work there every day. It’s been a constant source of inspiration – not the building, but the amazing things that instructors, students and staff are doing within it, together, on a regular basis.

Several key design principles were used to guide the design at every stage of the process – most importantly, transparency and flexibility. The main floor of the TI is a wide open space, with lots of glass, high ceilings, and windows. The light is amazing. It feels like a space that matters, and that instructors and students matter because it’s for them.

But, designing the audiovisual systems that power the learning studios in a way to enable that kind of flexibility and transparency was a real challenge. We couldn’t just stick large displays on the walls – because many of the walls are glass – and the ones that aren’t glass are Skyfold retractable walls that get folded up into the ceiling to combine studios.

We also couldn’t install permanent floor-mounted displays, because the spaces needed to be quickly adapted for different uses and layouts – each class can use the spaces differently, and sometimes they even change the layout of the room on the fly during a class. We needed technology that would support that kind of flexibility.

We (the broader team, led by The Sextant Group) came up with “collaboration carts” – 33 standalone units on the first floor, each with a 50″ touch display mounted to a stand with wheels. They can be plugged into floor boxes throughout the learning studios, and can be easily repositioned (or removed) as needed. We’re able to do everything from presentations to active learning sessions to collaborative project work to academic conference digital poster sessions, changing the layout quickly as needed.

Here’s a basic schematic of the bits of kit that power these collaboration carts. The carts themselves are really just the 50″ display, a webcam, and an ethernet connection that sends input data to the computer that controls that cart and sends back the video signal for display on the screen. All of the computers are on another floor in the building, in a dedicated server room.

This design gives a level of flexibility I’ve never seen anywhere. Each of the 33 collaboration carts can be positioned anywhere in the learning studios on the first floor, and the system discovers where they are and routes data accordingly. The instructor (or students) can control what happens to the room from a simple Crestron panel on the podium – switching the room from “presentation mode” where the facilitator can display content from the studio PC or their own laptop, and push it to all displays in the room (including the projector and any collaboration carts that have been deployed in that studio). Participants can see the presented content up close on the nearest collaboration cart display – and this has turned out to be a wonderful benefit in line with Universal Design for Learning.

From that Crestron panel, the studio can be switched over into “active learning mode”, turning each of the collaboration carts into a standalone unit to support small group collaboration. In this mode, participants can use the carts for Skype for Business calls (although that hasn’t actually been used, aside from demos), or launch Chrome or Firefox browsers to access any web-based content or software (including Office 365, Google Docs, Prezi, Padlet, Top Hat, and various domain-specific tools including chemical molecule viewers).

The carts also have a “present media” option, that activates a Mersive Solstice Pod for wireless presentation of media (or screen sharing) from participants’ own devices. it works on macOS, Windows, iOS and Android devices, and doesn’t require a dongle or cable for people to share their work. It’s been a really powerful tool, used pretty regularly by students.

We’ve also discovered that students love having access to the studios when there aren’t classes in session. Students have colonized every square inch of the TI, and are regularly working together in the learning studios (and occasionally watching Netflix or even dragging in a PS4 to play on the big screens…) – but by and large, they actually come together to use these amazing technologies to work on projects together. I’m really glad we decided to take the risk of leaving the studios unlocked during operating hours for the TI – our initial plan had been to leave them locked when not in use by a class. We took a risk by changing that, and so far it’s been extremely successful by letting students adapt the spaces and technologies to support their own learning.

So, what have we learned about the design of the AV systems in the first year? Mostly, everything has worked as designed. With the sheer number of units in the building, of course there were failures. Thankfully, that was covered under warranty, so we were able to get replacements quickly. But, the warranties all expire at the end of this month. Gulp.

1. Evergreening. We were in the honeymoon period during the first year, with warranty periods covering everything. That’s about to expire, and we need to plan to evergreen – replacing and upgrading portions of the system each year to make sure it keeps working as needed and we’re able to incorporate new technologies.

2. Fragility. The flexibility built into the system has been absolutely incredible – empowering people to do cool things without needing much more than a quick orientation – but, the state of the tech when the building was designed meant that the level of flexibility came with a cost. There are many points of failure – often, if a collaboration cart goes down or acts up, there are literally 15 different things to check, several trips to the mezzanine floor to reboot things and tweak configurations, and eventually the tech comes back online. But which of those 15 things did the trick? When there’s a class in session, we’re not about to methodically go through 15 trips upstairs and back downstairs to verify each step. We have time to make sure cables are seated, and reboot the smallest number of devices without bringing the rest of the studio or first floor offline in the process.

In the 2 years since the systems were designed, there are already new technologies that have the potential to greatly simplify this design, reducing the reliance on interconnected systems from different vendors (and hoping the interconnection keeps working).

3. Support. We’ve come up with what I think is an ideal support model. Instructors who teach in the building have to apply for space on the TI website, and a committee reviews applications to make sure what they’re wanting to do is feasible. Once they’re approved, we assign a staff member to be their designated first contact, who then liases with the rest of the team as needed. It’s meant we can learn much more about the courses and what instructors and students are dong, while still pulling support from the various groups in the TI as needed.

We’ve also launched a “TI Learning Technologies Coaches” program – 2 undergraduate students who are available to work with instructors and students to help brainstorm strategies before or during classes, and then pull in other team members to expand that support as needed. Drawing on the experience of students has been fantastic, helping instructors to think through some of the things they’re wanting to try from the perspective of students before trying them in an actual class.

4. Stewards. Because of the extreme flexibility of the learning studios, they can be kind of a mess at the end of each day. We have 2 student “TI Stewards” who take care of the studios, resetting technologies, clearing whiteboards, moving the tables and chairs into a common default position, and helping students clear the building before it closes. This has been a huge success.

5. Letting go of control. We had initially planned to leave the learning studios locked when there was no class in session. We let go of that, and it was a huge success. We have instructors doing weird things in the building. Our initial reaction was “is that allowed?” but we’ve always wound up responding “sure! that sounds awesome! how can we help!” and as a result, we’ve had life-sized african elephants made of newspaper, or volumetric data visualizations with a raspberry pi-powered fog machine. And lots of other weirdnesses. Let go. Amazing things happen.