 Welcome to the third webinar of the engineering rising to the challenge initiative from Purdue Engineering. My name is Armin Dhraman. I'm the executive associate dean of the faculty and staff here in the college. The engineering rising to the challenge initiative started in May this year in response to the National Academy of Engineers call to action for engineers to respond to the challenges posed by the COVID-19 crisis. Additionally, our initiative also looks into the longer-term challenges that are emerging as we seek to rethink and re-engineer the very systems that our modern society has come to depend on so that they might be more resilient to such shocks in the future and also serve society better. Part of the initiative involves webinars where distinguished panelists and thought leaders really unpack some of these challenges and provide a glimpse into what the future might look like. Today's webinar deals with the future of video conferencing, an industry that we have so come to depend on, especially during this crisis. And I'd like to introduce also the moderator for today's panel. Amy Reidman is a professor in our School of Electrical and Computer Engineering. Professor Reidman is a world-renowned expert in image and video quality assessment, video transmission, and video analytics. She's a fellow of IEEE and joined Purdue in 2015. After 23 years of industrial research at AT&T Labs, where she was a distinguished member of the technical staff and a lead and mentor scientist. Prior to that, she was a faculty member at Princeton University. Over to you, Amy. Great, thank you, Arvind. I'm excited to be here today with our panelists to be discussing the future of video conferencing. As you know, given the recent events of COVID-19, we've all been forced to or chosen to continue our work by connecting to people electronically much more than we connected to them in person. And so you all have much more firsthand experience with the video conferencing than you used to, and I'm sure you have many questions. So if you have any questions for our panelists as the panel evolves, please use the Q&A section so that they're all together in the Q&A. And with that, I'd like to just say a few words about all our panelists. So today we have with us Alex Eleftheriatis, who's a chief scientist at Enghaus. He was also the chief scientist and co-founder of Video, which did video conferencing services. We also have Calvin Hendricks Parker, who is co-founder and CTO of Six Feet Up company in Indianapolis. And also Ed Delp, who is the Charles William Harris and distinguished professor of electrical and computer engineering here at Purdue University. So just to give you some background on what's happened as a result of the video conferencing and all the connectivity that we do, it's really affected our network. So here's a graphic from AT&T that shows relative to a similar day at the end of February. How much did the network behavior change? How much is the network traffic change as a result of all of our retreat to remote work? And so what you see is basically that dramatic increase, statistically significant increase in the core network, in the wireless voice minutes, in the consumer home calling and so forth. And basically the network has been able to pretty much absorb this and still be able to carry on effectively and allow us to work in this new environment that we have. To pose the framework for this panel and what we're talking about here. So you may be very familiar with streaming video where you watch Netflix, Hulu or even YouTube. This is streaming video and it relies primarily only on downstream bandwidth. That is from the network down to your device. Interactive communication is our topic today which requires both upstream from your device up to the network and downstream bandwidth as well. Another difference between streaming and interactive video is that in streaming the delays can be a few seconds or longer. It takes a while to start to play and then it plays and it all goes according to plan. But if you take a couple of seconds for an interactive communication, then the quality of the experience degrades rapidly and in fact, interactive communication requires at least no more than three-tenths of a second. And I've heard numbers that are closer to one-tenth of a second for that. So just a quick brief summary of video conferencing through the years. In 1964, the 1964 New York World's Fair, AT&T presented the picture phone, which is shown here where you would hold a headset and talk to a very blurry picture of someone. This failed after a few years. They actually had a service and it obviously did not survive. In 1992, AT&T developed the videophone 2,500. And here we see the picture. So it actually, this worked on a plain old telephone line that you might have used to call a landline in your home. And it sent video across from one device to another. This device cost $1,500 and you needed two of them. And so it didn't get a whole lot of use. And in 1995, it was disconnected and the discontinued as a product. These days, you can go to Cisco and you can get what's called an immersive video conferencing system where you have for meetings, you have this very high quality video environment as well as video devices and a dedicated network to carry your traffic. This is not something that all of us have been experiencing in the last month or two or three. And instead what we've been doing is using packet video or what you might call network video. So just to pose some of the challenges associated with this, the network capacity in general is variable. So what comes to my house is gonna be different than what comes to other people's houses. And this is gonna vary with time as other applications or other people in my household may choose to use the internet in different ways or people even in my community. It'll change as a function of time. It'll change as a function of time of day. But the video bandwidth requirements, so what I need to be able to send good quality video up is also gonna be variable. And when the capacity of the network is not sufficient and can't carry my video, then the quality of my video may suffer. And I expect that you've all seen some instances of this. And so this kind of sets the stage for some of the topics, some of the technical ideas that we may be talking about today but we'll be talking about much more than just these ideas. So with this, as I have mentioned, we have our three panelists today across industry and academia. And with that, I'd like to get started with Alex Eleftheriatis, who's the chief scientist at Enghaus. Take it away, Alex. Thank you very much, Amy, and thank you very much for the invitation. I think we were very lucky that video conferencing technology was actually ready for COVID-19. If what happened now happened 10, 15 years ago, things would have been very, very different. In fact, with the opportunity of this panel, I looked a little bit in my notes and I found that the very first paper I could find about coding video so that we can adapt to variations in bandwidth and what have you was back to 1978, okay? And in fact, the first company that developed sort of commercial products, the first startup, if you wish, was founded in 1984, a company called PictureTel, that somebody of you might recall. And in fact, it was started by MIT people, Matthew Dertuzos and Greta Papadopoulos with Solé. And in fact, Dertuzos and Papadopoulos happened to be great like myself. And just to give you an idea, they were selling a system in 1986 that did seven and a half frames per second with 11 boards at the rate of two channels of 52 kilobits per second. And you could buy a pair for $150,000, okay? That was a nice gadget to take home. So we came a long way since then, but I want to highlight some key innovations and to appreciate how we go to where we are today. So until the sort of the 2010, the standard for multi-point video was what's called the multi-point control unit. So these were like servers that had a lot of hardware and they were receiving video from multiple participants like we are here today, and they would decode the video, re-encode it and send it back out to the participants. So this happened to be very expensive, very big boxes, like custom design. And what was worse, they also introduced a lot of delay. So you would get in a call with your colleagues or friends, in fact, colleagues typically, because it was so expensive, you would never use this for fun, okay? And the delay was horrendous. Now, we're talking about more than 200 milliseconds when the standard for long distance telephony is about 180. Okay, that's what the company has delivered to you so that you can have a sort of a nice conversation with somebody on the phone. So it was already out of spec. So video was actually founded in 2005 and introduced a new architecture by using scalable or what's called layered video coding, we were able to solve two problems. First, we eliminated all the processing at the server by having the participants send multiple resolutions of the video so that the server didn't have to do any processing. We just select what to forward this participant, okay? So that made them very simple designs, actually a generic computer could do the job, okay? So you didn't have to spend $100,000 for custom hardware, just buy a PC or a new box and it would do just fine. And the delay through these boxes was much lower. We're talking about less than 10 milliseconds. So suddenly the end-to-end delay did get to become much lower than long distance telephony, okay? So suddenly the experience you would get by being in a multi-point call was acceptable. In addition, this allowed us to resolve error-resilience issues. So normally when you start transmitting video on a network that loses package, you start seeing artifacts. In the past, people who are old enough in this webinar will remember cases where entire portions of the video will become severely distorted. You will notice, since as of several years ago, that was completely eliminated. And the reason is that new techniques were introduced, again through scalability, in all standard codecs that allow you to basically get around these problems, okay? So suddenly with this design, which actually called it Selective Forwarding Unit or SFQ, we were able to solve all these problems. In by 2011, all the major providers of video coverage and services switched to this architecture, okay? So that includes Google, Google Hangouts, which was actually a customer of video, that includes Skype, Zoom, the open source ZT, everybody today that's doing multi-point video uses this architecture, most of them use single casting, which is a special case of scalability. At any rate, so that was very important because the quality of experience was actually much better. And what's actually very important for COVID is that the servers were genetic. So people that wanted to increase availability of resources, they did not have to go and buy very specialized computers to put on the data centers as demand for video conversing increased. They could use standard servers from Amazon and other Google or other cloud providers and very quickly ramp up support for the users, okay? So the simplicity of the servers and the fact that we today are able to spin up servers at will, I can sit here in my home in Athens and can have thousands of servers operating in application or around the world, okay? That was not possible 10 years ago. So a lot of these factors work together to get us to where we are today, thankfully. Now I actually counted 16 apps on my iPhone that can do multi-point video today, 16. I think I only have to pay for two and I'm not quite sure actually for one of them, okay? But I think that's where we are today. So I think if for people who are asking each video ubiquitous, I think that pretty much says it all, right? That on the phone that I can buy from any sort of consumer electronic store, I have access to all these sort of different types of services. A final, I think very important development was something called WebRTC. Now WebRTC is a technology that was introduced in web browsers to allow them to do video and audio transmission and reception without any plugins. So this browser by itself can encode and transmit and receive video and audio. By the way, using the SFU architecture, in fact it was designed to support that. So this means that you can quickly get on a call using a browser without having to download anything. I mean, you just send a link to your colleagues and friends and boom, they can connect even if they have no relationship with the provider you choose to engage, okay? So I think that was also very important because it allowed people who are not specialists, right? And might be, you know, might have difficulty in installing software, et cetera, to be able to get into calls very quickly. And by the way, without paying any money. So you might have a service where you have to pay, but typically guests don't have to pay. So you can be in a call, invite people to join and they don't have to spend any money. So I think that was also very important. Now, thinking what's next, I mean, I think what's clear now is that we can all talk to each other. And in fact, my workload increased due to COVID. Why? Because I worked with video conferencing, you know, essentially throughout my, most of my career. But now everybody's on video calls. So in the past, I was a little bit shielded, right? Because people might not be used to that. We'll have to have a physical meeting. Not anymore, okay? So that, you know, grew a lot. But also what grew a lot I think is an appreciation that when we connect, unless it's like with friends, usually it is for a purpose, okay? We have a meeting to get something done. So I think right now video conferencing is a ubiquitous tool, but it's just a tool, okay? It's a plumbing that allows us to connect, see and hear each other. But it doesn't help us a lot to get the job done. And I think the best example that I have for this is software which is tailor-made to solve specific problems. One example that I like to give is Quicken. Quicken is a very well-known program for personal finance management that allows you to manage your personal finances. So it models the concept of a bank account or a credit card, et cetera, right? So that you can do a lot of things very easily, like download, credit card transactions, reconcile at the end of the month with your statement. All that just with the push of a button. Now I could do all this work with Excel, okay? I could just sit there and enter data, you know, manually reconcile, et cetera. But good software should model the problem you're trying to solve and provide, you know, first-class workflows to allow you to get it done quickly. So I think right now, in our collaboration experience, you know, I have here, I have Slack open, I have WhatsApp, I have Viber Messenger, SMS, Zoom, you name it, right? But I have to design the way all these things will work together to get my job done. So I think we are at the point right now where desperately need solutions that will make, you know, they'll cross sort of the chasm and allow you to get more into solving absolute problems in collaboration. So things like automatic transcription, retaking, actual item collection, all these things are going in that direction. We are still at the very beginning of that, but I think now that everybody sort of into video conferencing and collaboration, right? As we will be hesitant to do business travel in the near future, I think we'll see more and more of that development sort of a bound, yeah? Thank you. You're muted any. So this is the age-old issue of video conferencing is we try to be courteous, we try to mute ourselves when we're not talking and then we get ourselves into trouble. And usually Zoom is a little bit better to tell me that I muted, but thank you, Alex, I was muted. So next I'd like to pass the mic over to Calvin Hendricks Parker to talk a little bit about his company and how they work with video conferencing for events. Sure, my name's Calvin Hendricks Parker. I am actually a Purdue graduate. I'm happy to be back and helping out with other projects around Purdue. But my company is Six Feet Up. We are a Python application development and cloud consulting firm based out of Fisher's, Indiana, just on the northeast side of Indianapolis. But I guess the real reason why I'm sitting here is because I really care a lot about building communities. I have been involved with building various user groups around programming languages or specific technology groups here in the Indianapolis area. We started the Indianapolis Python user group back in 2007. And since then, you know, they've always been typically in-person events where we all gather together for a meetup. It was growing quite large, you know, to 50, 60, 70, sometimes even over 100 people coming to a single location to hear the speakers that were happening. We started live streaming those events, taking advantage of the technology available to us. And we've been live streaming for about two years and we've been recording the events for about three years to put them onto YouTube and various streaming platforms for people to be able to watch those things. The issue is that they weren't very interactive. They're not very engaging. If you're just watching a stream, it's really a one-way transmission of the content to you. You don't have an easy way to interact with the speaker or ask questions or interact even with the audience. And so we've run three virtual conferences so far. We just finished up our third, which was the Python web conference about two weeks ago. And it's part of the kind of post-mortem from our second conference we did, which was the Cloud Conference in March, which we had to turn into a virtual conference at the very last moment. We were about a week after they did the shelter-in-place order here in Indiana. So there really would have been no way for us to do that conference in person together. We decided to take that thing virtual within a week, used the Zoom technologies, used Slack for chat, but it was really missing kind of like Alex said, it was a set of tools, but it didn't encourage the workflows. It wasn't very opinionated about how you use them. So we decided we actually needed to build a platform to better enable the attendees, to exchange ideas, to ask questions, to engage with the sponsors, to be a part of the community, and really feel like you're a part of a virtual conference or an actual event. I think a lot of times people want to try and recreate the reality, but I think we need to enable not necessarily a full simulation, kind of a skeuomorphic version of a conference where you do a VR, fly through into an expo hall and click through to a theater. Those, I think, get more in the way of the actual goal and goal of what the reality is, which is to actually connect with others, to share ideas, to come together, listen to speakers, give their thoughts on things, and be able to get that interaction afterwards. I mean, one of the most popular activities of a conference for me and for a lot of people who go is after a talk, being able to approach the speaker and ask direct questions that may not be applicable to the whole audience is why you didn't ask them into the whole audience, but there's a typical group of five to 10 people who are looking for that kind of really, really specific knowledge. So we built a platform, a software platform that is hosted in the cloud, uses a content distribution networks to be able to scale, but allows for a single pane of glass and allows for people to come in and consume the content like they normally would in the conference, but also participate in Q and A, also participate in a side, back channel, hallway track style channels, and then also get that face-to-face time that they would want with the speakers through using combination of video broadcasting, using RTMP protocols, using Zoom rooms or Zoom meetings for the breakout sessions, still using Slack for those real-time chats and then keeping it all coming into one spot so that it's easy to access as an attendee who would want to attend these kinds of events. And through that, I think we feel like we've finally given at least a start, a really good head start on recreating that level of value you get for going to a conference because all these conferences aren't inexpensive and people pay a lot of money to go to them, or right now they don't want to travel and they don't want to be stuck inside with a whole group of people where you could still be here at home. Now it's not to say it's without challenges, I think there are still a lot of challenges moving forward. People will definitely get Zoom fatigue or kind of just you're sitting in front of your screen, you really need to consume this content. I've got a lot of tips and tricks for that that I can talk about later as we get into this. Thanks, Amy. Thanks, Calvin. And next up is Ed Delp, who's a professor here at Purdue Electrical and Computer Engineering. Ed. Thank you. One of the things I'd like to comment on is just the attendees list. I'm seeing some old names there of some of our former graduates. In particular, I see Karl Crawford and David Tumanuska. I haven't seen you guys in a while, so I just thought I'd say hello. I'd like to comment, I've done work in video and image processing for a long time. And we did some early work with PictureTel and some of the other companies. But I'm not gonna really talk about the technology, I'm gonna talk about some other aspects. And I wanna mention something that Alex said. That I agree with him, but then I also disagree with him. So at the same time, Alex, okay? So Alex said, it's amazing how well this worked. In 10 years ago, we would have been in trouble. And I agree absolutely with Alex, but I think there are a large number of people out there that say it's amazing how much this didn't work. Okay? And I think that there's been some sort of interesting implications when you put this on top of COVID. Number one, I know because Amy and I talk a lot, that we complain how much at the end of the day we're completely wiped out from doing four hours, five hours. I think Alex, you mentioned this too. From doing four or five hours of this, okay? And we gotta do it to interact with our students and run our research programs and interact with our colleagues. And I think we gotta think about that because I think some people are really getting worn out. I also think that whereas Zoom, and we're doing pretty well today on this, I think if you talk to a lot of people that have young children, who they were having to work at home and then helping their young children get through their education program, particularly in the March timeframe, at least here in the US, the schools closed, the K through 12 schools closed for the rest of the year. I think that they would argue that maybe it didn't work as well as we'd like it to work. So I think this idea of what can we do? And I'd like to hear from the audience too. I don't have a particular solution for this. I also would like to just add one other thing. And I put it on my list of things that we sent to Amy ahead of time. There was a recent study that was published by the British Health Service. And they said that they thought one of the leading causes of death of older people in the UK was loneliness. So the question is now that we're all sort of shut up and things are opening up but there's some issues here in the US with that opening up. You know, is this type of technology, can it help with these issues of loneliness and people still maintaining connections and things like that? Because I think this COVID-19 definitely has a mental health component that I think we also need to think about. Purdue is gonna be doing a grand experiment in about six weeks, where we're gonna be having 40,000 students or so come on campus and we'll see how we're gonna do all that and how we're gonna use some of these technologies to augment and replace some of our educational experiences. But I think it's really interesting. It to me, knowing, going back again, Alex using one of the old H.261 systems and going back and how much money you had to pay to get access to it and four, five, six, eight, 10 ISDN lines, it's pretty amazing what you get for nothing nowadays. And I think people have exploited that. I think having said that, I think I'll stop and maybe we'll move into the next phase, Amy. So I'll turn it back to you. Okay, so on that note, so we really have come a long way and the fact that it did just almost seamlessly start working and we could all transition into this situation was pretty astonishing to most of us. But like, what's missing? What do we need to reduce the distinction between video communications and the face interactions at work that we were expecting? Or do we need better video quality, more functionality to our apps, better audio quality, immersive video? What are your thoughts as to what might help actually make this better than we currently have it? Do you have Calvin? Oh, you're asking us? Okay. Yeah, there's a lot of things that need to happen there. And I think Ed has a lot of opinions on like rural broadband and getting access for certain kinds of folks to be able to access this. I mean, from my standpoint, when we talked a little bit about this prior, audio quality is so important. Getting that message through to people, I think people can put up with a little bit of grainy video or it's a little blurry here and there, but your brain can get very distracted when it has to process all that noise around your own audio, around the audio coming into your ears. You do it transparently, but obviously it takes energy and brain power to do that as well. And the more crystal clear and the audio is coming to the end users. I mean, we're getting very good. The microphones on the MacBook Pros and the latest laptops and latest phones are all very excellent, except if you start using Zoom and it spins up your fans, and all of a sudden it's trying to battle the fans with the audio processing and everything else that's going on in there, I think there's still a ways to go with the microphone technology. Yeah, in fact, I would doubt, I think that the audio we cannot emphasize it enough. I have a friend who's an audio engineer and he always complains about it. And I think nowadays, I actually refuse to take calls on my regular phone. Me too. On the regular sort of phone service. Why? Because it's so low bandwidth and the contrast with the actual, voice over IP connection is so big to me it's like there's something broken. Like I'm like, what's happening? What's wrong with the phone? So for better or for worse, this is completely unacceptable. So what, and unfortunately, setting up the audio by the way in a room is not easy because there are aspects that have to do even with architecture. So it's not just what equipment you'll buy. You can buy the best speakers, the best microphones, but if you have a room with boxy with like, you know, flat surfaces where you have reflections, that can be problematic. A lot of small conference rooms are like that. They're unusable for, you know, the office in course. So I think people have to be a little bit more, so to say, trained about these aspects. Have the right equipment like, you know, the least you could have is like a USB headset, right? So this way you isolate acoustically from the room. You know, you wear, okay, you wear the headphones and a microphone, fine. It doesn't be beautiful, but at least, you know, it's a reasonable compromise. Ideally, what you have is a dedicated system with a very high quality mic and speakers that do echo cancellation in hardware, so that's like, you know, rock solid so that you never get sort of glitches, okay? But of course that costs money. Hopefully, you know, these days you can buy for a small room, I don't like a thousand bucks maybe. In the past it was 10,000, right? So hopefully slowly, slowly, you know, equipment gets more sophisticated. You can buy for 200 bucks, like webcams with like a microphone sort of array and an HD camera, so things are improving. And I think as people are more, you know, tuned to what they have to pay attention to, right? I think things will get better. But to the point about being tired, I think what I realized is that when we're in a conference call, what happens is because of our positioning, our posture, right? With our, you know, instead of our neck being like in a relaxed position, we all tend to sort of look up to the screen. So that creates a lot of tension. And of course after four hours, it's like you took a beating, okay? And unfortunately, even though I know the course, trust me, I don't, I never have the right posture, right? Someone was like that, okay? So I don't know. We will have to get used to doing that, to fixing that. I mean, I agree with the audio. And I think it's interesting that three of the four people in here have spent a lot of their professional life working on pixels. And we're all saying audio is important. And, you know, I took the time and before we started, the panel was talking, you know, and I spent a little bit of money to get a much better mic. And that's what I'm using right now. And also I spent, I also am using an external camera, not something built into my computer. And I think that also makes a big difference. Just Amy, if you just give me a second, Calvin did bring it up. And I forgot to mention this. I think one of the other concerns I have is that, you know, I am concerned particularly in the US about this issue of the sort of the have and have nots. We live in Indiana and, you know, there is issues with people out in the rural areas, you know, getting broadband. Now, 5G is supposed to help this, but it's not quite here yet. And I think this has been a real problem for K through 12 education and a lot of states, you know, you wouldn't think of like California and New York and things like that have had issues in the rural community. And I think we really need to make sure we take care of this. And also those people working out there who maybe would like to work from home, I think it could be a problem. So, you know, we're gonna be dealing with this COVID thing for probably two years and at least, and I think we need to sort of worry about this as another component. Looks like there's a lot of questions in there. There are a lot of questions. So, look at that, there are. So, one of the questions ties on with the discussion about the audio, which is some people's computers have a lot of echo problems and others don't. What technologies do we need to eliminate echo? And, but what I think I heard you all saying was we just need to pay a little bit better money for a better system, a better detached microphone for your computer or something, is that true? Yeah, you need what's called echo counselor, right? So, that is a device which basically makes sure that the microphone doesn't pick up the signal that the speakers produce, right? So, this way you don't get a return of the sound. Let's say, you know, my sound goes to Ed and then if there's no echo cancellation, okay, I'm gonna get a copy of my sound back to me. So, then I will hear it as, you know. Yeah, and those echo canceling mics are available and not that expensive. You know, I have a particular mic called a Yeti, Nano Yeti mic. And I think I paid less than $100 for it and it's got a really good echo cancellation on it. The fun part is that if you don't hear echo, you are causing it. Okay? So, keep that in mind, okay? So, unfortunately, right? Hopefully I'm not causing echo, I'll tell you. But that's, you know, usually. So, the corporate, they're not aware that they're causing a problem, okay? So, I guess my, I'm sorry, Calvin, but my thing would say, you know, don't use the built-in stuff, okay? If you use it, if you need to, but some of it has issues. So, some companies, and you know, I don't wanna make any product placements here, but some companies are embedded in the acoustic design of their hardware, okay? And I have very positive experiences with built-in mics and speakers from very well-known and very big companies, okay? So, but again, you know, you can try for yourself, okay? Also, what I have found is like tablets are perfect for video conferencing. Why? Because, you know, they don't occupy your computer in terms of CPU, et cetera. They're more properly placed downward, right? So, you know, the tablet will be in a lower position than your main screen, right? And so, unless you need to set a document, right? It just need to be in a call with video and audio. It's actually a very good alternative to your full computer. And typically, they have good combination of mics and then speakers within the cancellation. Well, and very good cameras. Those tablets and phones today have better cameras than what are in your computer. True. That is true. But again, I chose to go all external, particularly in this particular setup I made here, and it's been a tremendous improvement, and I think I've invested about $200. Yeah, cool. So, I want to switch gears a little bit and pick up some concepts that are in some of these questions. So, Alex, you mentioned early in your discussion that when we create our video calls, now we're primarily, we have a plan to get something done. And there's a lot of discussion in the news about companies who are saying they're just going to go all remote work. In my experience, there's a lot of value to be had by those spontaneous hallway discussions that you have where you just run into somebody that you don't work with, that you don't have a plan to work with, that you just, and sometimes the best collaborations form at those exact moments. And I think my question is going to start with Calvin. Calvin, how can we, but all of you can chime in. How can we create environments in which the video will allow us to have these spontaneous connections that allow us to just chat with the person that we might meet in the hallway? Yeah, I've got a couple of takes on that, because one, we're all fully remote right now on our company. So those hallway kind of water cooler discussions, we had already transitioned a lot of those things into Slack or some kind of chat or text mode. But then we're very, very fast to move that into a Zoom meeting. And actually, there's plugins for the various chat apps that you're always one click away from seeing the person who you're next to or kind of having that run-in with or want to have a discussion with. And we do encourage that. And again, it's going to be a mindset shift. It's a company culture type thing you need to establish from the top down for sure. Everyone needs to be just ready and willing to jump on a video call. And in our company, everyone's expected to be on video anytime we have a phone call. We never have a phone call. And I'm like, Alex, I never answer the phone if it's actually someone calling my mobile. I will cancel the call and call them back on a video call of some kind because I want to see them and have that interaction. Now, the second aspect I wanted to talk about was we talk about like networking at real events like conferences and running into people in the hall. Again, a big reason that I even go to conferences and that's harder to simulate because there's a lot of people in a chat space who don't necessarily know each other and you're all kind of trying to feel out what's going on and what's the appropriate etiquette. And I think that just takes a lot of leadership from the conference organizers and also getting people who you, as a conference organizer, getting people coached into how to help bubble those conversations to the surface. For example, we leverage all of our speakers, have them be in the conference rooms and all the chat rooms. We also encourage the speakers to spawn separate chat rooms for their topic if they know that a lot of people who are interested in software testing make a software best practices room in the Slack and let people go nuts there. And I also do a lot of things during the events to try and get people to interact face to face. You can do Zoom meetings with 300 plus people on them like face to face style like this but obviously you can't have 300 people all talking at once. So what we'll do is typically organize some kind of events like a board game night or a virtual cocktail hour. And then I use the breakout room features of the platform. So for example, Zoom, you can actually randomize so many people per breakout room. So if we've got 50 people on the call, I'll put 10 rooms together, send five people out with a mission. What's your name? Where are you from? And I give them usually a question. What was your first computer? What was your first program language? Some bit of trivia that usually sparks a bunch of conversation. I give them 15 minutes and then I bring them all back in and we do it, we randomize it again and go after it again. And the next thing I know is that the next day people are much more active in the Slack channels having discussions because they got to meet the night before and actually see a face-to-face conversation with people. We had one of these cocktail hours that spanned a 13 time zone region. We had people in Bangladesh and people in California. Yeah, if I can jump in here, I've been on good ones and bad ones in the last two months. I'll give you an example of a bad one. I won't be very specific, but it was a workshop associated with a large conference. And what the workshop organizers did is they had all the presentations recorded and submitted ahead of times and they just edited them into one big file. And at nine a.m. West Coast time, they just started playing that file and they stopped it for the lunch break and restarted it and that was it. And then there was some interactive chats but it didn't do very well. I was in some, at that same conference, I was in some other workshops that essentially did some of the things that Calvin mentioned. I was one of the speakers at that workshop and I made myself available, the whole workshop was a one day workshop. So it wasn't, you know, but I made myself available and there was a Slack channel going on. We did that. I recently attended a DARPA principal investigators meeting. There was 350 people on the Zoom and it worked really well. There was lots of inactivity. So it takes a different way of thinking and I think it's a lot more work than a live event in many ways, I think. So I think it can be done, but you gotta think about it. You don't wanna end up just recording the session and then just streaming it. There's a certain loss of like energy, I think, in the session and in the speaker and everything else going around. If it's pre-recorded as opposed to, I expect my speakers to give their talk on the spot and actually take questions and interact. There's a certain energy. It's like going to a live performance versus like just getting in the CD and listening to the CD. I think, but if I may chime in, I think that we're social animals and I think that the physical aspect of our being together is very important. And so I think for better or for worse, there are certain things that we do that require this physical presence. For example, if you are like improvising in this product design, for example, right? So you need to have people interacting very intensely in a maybe in front of a whiteboard or whatever, then the physical presence benefits. If you're just managing things and you need to touch base on a standard meeting, sure, you can do that remotely, right? And go through your issues and inform your colleagues, right? So I think depending on what it is that you're doing with your coverage may or may not eliminate the need for the physical presence. But I think the best place where you can see the need is in through K through 12. But in fact, being there is part of education. Education is to be there. There's no alternative. So there I think, I don't think it's realistic to expect that we could replace high school by remote education because the whole point is for the kids to be with other kids and start building a social identity, learn how to interact with others, et cetera, right? And I think going up in a student environment, undergraduate students, graduate students, again, as you grow older, maybe you need less of social training for lack of a better word, right? And maybe you're more focused on getting things done. But I think everybody would agree that even in a company having social interactions as you go through your career, as let's say different groups form, different allegiances, whatever, you know? So there is a lot of commotion within companies. So I think it's important to have that presence, right? And I think personally, having lived through, again, most of my professional career as a remote worker, I certainly feel that, yeah, I missed some of that activity for sure. But again, you know, you cannot have it both ways. I mean, there's always a trade-off. If I can jump in here, Amy, for some of the questions that are very produce-centric about is there discounts available and things like that? Amy and I really don't know the answer to that question. I think if you're a Purdue employee, you better, you should talk to your management. And there was one question about, is there a centralized space? I think what you mean is there a centralized space where maybe you can go for a Zoom meeting. Some departments have this. Many of the classrooms at Purdue will have boiler cast, which is somewhat similar to a Zoom that's available. But questions about equipment and discounts and things like that, we really can't provide any answers. You need to talk to your department head. So... So I'm gonna switch the topic a little bit here. So, and this is triggered in part by one of the questions. So I'll just read the question. What I have experienced is people play videos in Zoom calls. 90% of the time it didn't work well. Any suggestions to make it better? So the idea here, and my students have had this too, is we're doing research into video and we wanna show the results of our tracking algorithm. So we play, they share the screen with the video and it doesn't come through very well. And I think there's a broader question associated with all of this is it's not really just about face-to-face communication. It's also about getting stuff done. So... Yeah, what happens, what happens, let me tell you, what happens is this. Usually the way content setting works is by grabbing a portion of your screen and using the same video encoder to transmit it. However, because of the very high resolution and the need for very high fidelity, what happens is that video encoding is tuned to happen at the very low frame rate. Because typically when you're looking at the content share, let's say a PowerPoint, right? It doesn't change a lot. I mean, it might be flipping slides, but then for several seconds, it remains stationary. So the whole process of encoding that content is tuned for stuff that doesn't change a lot. So if you do the content setting of putting, let's say a browser playing a YouTube video, okay? You're breaking all the assumptions that the designer did for that. So unless the designer has a detector to figure out, oops, he's not setting or she's not setting PowerPoint, this is actually video. So I better use the same algorithm I use for the video camera for the content. Like if I don't do that, it will be, you know, it will be awful. Okay. In this case, it's probably better off setting the link to the video so that everybody can see it. I mean, if it's possible, right? But that's an impediment from a design of the store. All right. But Zoom does have a setting for that if you're using Zoom. When you share screen, you can actually say optimize for playback video. So it does change the encoder for you. Also, again, I'm just looking at the questions. There was a question about having private side conversations. Many applications like Zoom will allow you to set up breakout rooms. And in those breakout rooms, you have a private conversation with the people in the breakout rooms. You sort of have to coordinate that with the host. I have used this on several meetings and it works quite well for having side conversations. I don't know about some of the other apps, whether WebEx will do it or I know, but many of them will allow you to do that. Another way of doing it, as Calvin mentioned, for a lot of these large meetings, there's usually something. Lately, it's been Slack, okay? I'm not a great fan of Slack, but I use Slack. You could also use Microsoft Teams to have a separate channel going and Slack allows you to have private conversations. You could then, even Slack has some video chatting capability. So that's another way where you could have private meetings. So those things are possible. You just got to learn how to use them and play around with the technology a little bit. But the breakout rooms in Zoom has worked out very well. Actually, but back to the first question on the video, like being able to stream video or streaming content that's moving, do either of you know of ways you can actually do kind of a watch together where it synchronizes it across a whole group of people who would be sent a link and say, let's all watch this together? Ah. That's, the problem is getting the perfect synchronization is what you're talking about here. You cannot be interactive, of course, but I think it's convenient, but I cannot give you specific points that's on the top of my head. Yeah, I mean, that would be a good feature to be able to have. Oh, by the way, again, I'm sorry, I mean, these things are popping in my head tips. Another set of tips. If you can connect your computer directly via ethernet to your router that you have in your house, you will also find, and don't use Wi-Fi. Some of you may not be able to do that. You get a lot better results when you're doing video conferencing. So this machine I have right here is connected to by ethernet cable to the cable modem router that's right over there in my house. And it's much better than using Wi-Fi. I realized some of you can't do that, but Wi-Fi wasn't exactly made for streaming and that it has some problems relative to that. Now some of the later ones are gonna be doing a little bit better. But yeah, if you can do that, that's a tip. Let me ask you guys some video questions. So the H266, the versatile video coding was just finalized this last week. Within a week from now, it was finalized. Many of you may know that for the most part, video these days has been transmitted with H264, even though we started with 261. What happened to H265? We're just starting to see a few things where people are sharing 265 video. What does H266 bring to the table? Are there any features that are must have in terms of video conferencing or other areas that will drive adoption of this in a way that was faster than 264 and who remembers 263? And so yeah, do you expect this to have rapid market penetration, be a market flop, something in between? I'm gonna let Alex jump in. That's all your question for Alex. I spent most of my life doing this stuff, so maybe I should say a few things. So look, for better or for worse, it's been a very complicated issue. The fact is that most of the video was H264, which was developed back in the early 2000s. Seats were developed around it. It was developed by the joint committee between the International Standards Organization and the International Telecommunications Union. So things were fine. Now there was a patent pool associated with that. So if you make equipment that uses this codec, you have to pay money to this patent pool and everybody was doing it and everybody was happy. Now what happened is because this is big business, just to give you an idea, DVDs were using MPEG-2, an older codec, essentially H262. The royalties for that business were more than a billion dollars a year. Just to give you a sense of how important patent royalties were in video. And of course, this number just kept increasing. I should say I have patents in the patent pool, so I'm not even a partial party to this, but so it was big business. What happened is when, with the next generation of codecs, and by the way, H264 was the first codec design that was able to do both communications and entertainment. Before that, you had separate codecs for PV, and DVDs and what have you, and separate codecs for video conferencing. So these, for the first time, were merged in 264, okay? So big success because for us consumers, that was a big improvement. Unfortunately, it became the victim of its own success in that the next generation, which was H265, or HCBC as it's called, instead of having one patent pool, there were three. And not only that, instead of having a royalty cap where its company would pay a maximum, let's say, okay, for patent royalties, right? Some of these pools did not have a cap. So at any rate, this business environment got very complicated, right? And companies started to become reluctant to embrace it, okay? And in fact, Apple is using HCBC in all its products, but aside from Apple, very few other companies, I think, are using it. Obviously, all TVs support it, right? So- But it was really complicated. Yeah, the problem is with software. What happened is this, the more videos started to become used on the internet, right? And you were actually downloading video players, right? The business model of a free player did not allow for payment of royalties because I downloaded a free player who's gonna pay the 20 cents, let's say, due to the patent owners of the codec. So either the developer of the software pays it out of pocket, has some other business to pay, right? Or find something other solution. That's why in parallel with this, you know, this codec that came out from ITU and ISO, there were royalty-free open source codecs developed. VP8 was one, VP9 is another. So that was pushed by Google, okay? And in fact, we also, you know, code develop parts of VP9. And most recently, because of HCBC, a new royalty-free codec was designed called AD1. So the alliance for open media was formed by the likes of Facebook, Google, Microsoft, you know, you can look them up online. I actually turned the group there. And the goal was to design a royalty-free open source codec that'll be as good as HCBC, okay? So that companies that are offering, you know, software-based services, but they don't sell hardware, they don't get money necessarily from the playback of the video, they could do, they could do their own. So that's how it's available. In fact, already you can TV support it. Already YouTube allows you to encode every one. So it's slowly making its way into the marketplace. And of course, to make matters way even more complicated, the next generation codec for my new comes out, 266, as Amy pointed out, okay, to try to find its own place in the marketplace. So now everything is very upset and confused. And it has been like that for many, many years. In fact, more than I care to remember. Yes, Amy, sorry. Standards are supposed to make interoperability easier. And that's why we do these standards so that, you know, the video MCUs talk to each other and whatnot. If there is this proliferation of standards, how does that help the people who might be on the call who just wanna be users of video? Two things. First of all, for better or for worse, we don't use equipment anymore, we only use computers. So as a result, if I download the app you are using, I'm done. I don't need to care about interoperability because my equipment does have to talk your codec, right? The software I will download will include the implementation of the codec. So the cloud, the cloud services partially solve interoperability problem, right? It was no longer true that I was buying a separate box, but another box and these two boxes had to talk to each other. We both use computers and we just have to download the same software from the same provider. So that's one part. The other aspect is that, look, you always need to progress, right? So every codec generation improves compression efficiency. In fact, since 264, we all have whatever is needed to do good interactive multi-point video. So after the sort of people figure out how to do it, all generations after that embrace all the same tools. And the goal then was just to keep improving compression efficiency, trying to have whatever the previous generation did. So that's the progression, okay? But unfortunately, again, for better or worse, for worse, we have two camps, right? One is the royalty bedding, one is the royalty free. Both are very strong. Both have arguments, right? Of course, electronics, what have you, right? And so I'm completely with you on the issue of standards, but just to drive the point home, when the W3C, which defines the web standards, including HTML decided, had to decide a standard codec for WebRTC, what did they choose as a mandatory deployment codec? Both would be eight and eight and six four. So instead of saying one, we said two. So neither VHS nor Betamax wins in this case. Everybody gets a seat at the table. So I'd like to answer a couple more questions from there. So Kartik has asked a question about Zoom, playing videos and Zoom calls. Yeah, we talked about that one. Did we? Well, one of the things you can do is there's, if you click on the share button, you'll see two things at the bottom. One shares computer sound, the other one optimized for video clip. You might wanna try that. Brian asked a question about looking at sort of wireless congestion and things like that. Yes, there are people at Purdue and the school electrical and computer engineering that are looking at problems like this, but remember Wi-Fi was not made for interactive video sessions. At best it was made for doing streaming, one way streaming, like you watching a YouTube video or something like that. But people are looking at that. It's not gonna happen tomorrow, but there are definitely people in our wireless group at Purdue looking at this. Somebody, Young Lou asked a question about what I pay $100 for a haptic device so I could do an actual handshake of somebody. And the answer is, yeah, I think haptic devices are gonna be interesting, okay? And so what I pay $100, yeah, I'd probably pay 100. I'm a first adopter, so I'd probably pay a little bit more than $100. They have something like that. I don't know, Ed, do you remember the application called Bump? Yes, yeah, I remember Bump. That was a core example of this. It ended up being basically data sucking where. Yeah, yeah, I agree. So, yeah, so okay. So I answered a couple of questions. I just thought I'd jump in, so. So, Dr., I have a complicated question inspired by the questions in the Q&A, but basically like Zoom has a number of features, WebEx has other features. You know, we've talked about some, even third set of features that we might wanna have. Is there any way that we can hope to like pick and choose the features that we want for our particular thing and maybe Zoom can talk to WebEx or you talked about downloading software each time? Is this even plausible or do we just need to push the manufacturer? Go ahead. It works today. So for example, at our office building, we have, we use Zoom and we have Zoom rooms. Zoom rooms can call to a SIP phone or an H.263 conference bridge. We regularly will dial into a WebEx room because maybe a customer we're working with doesn't like to use Zoom or doesn't wanna use Zoom and they're a WebEx customer. The only issue is that it's not that easy to do it from your desktop. It only seems to work best when you're in one of these rooms. It's been specifically set up and has a controller there and you can type in the address and then it does interoperate and work quite well. But that's been my experience with it. Yeah, all the, look, all the different, including videos rooms, for example, have support for all these older, let's say, standards of interoperability which do get basic video and audio communication flowing but it's nothing close to what you have available on your computer, okay? So functionally, they're much more limited. And again, maybe also because they were like special purpose devices themselves, they don't have the flexibility of the software. But look, in theory with environments like WebRTC, where you can have JavaScript running on your browser, I can certainly see companies developing components that all work together on a webpage. Already by the way, if you go to the website, a lot of the components are made by different developers, right? So it's not like the whole website was made by one person, right? So they might include JavaScript from different libraries, the whole components coming in from different resources. So what you see in the end on your screen is really a mashup from different services, right? So I can certainly imagine a situation where people might be combining, right? Using WebRTC and JavaScript, application-level functionality with video nodules. So you have a server doing this, a few stuff for video nodules communication. And in the same page, you have components for like document setting, you know, what have you, right? So I can certainly see that. And I think a technology like WebRTC that make it available on the web browser make it easy for any developer to do it. You don't have to build everything else like the video conferencing component. You can take that ready made by somebody who supports WebRTC, okay? And a lot of other companies support that. And then you can start building your own application components on top of that. And they can all live together on the same page. So in terms of the bandwidth, the have-nots, the rural communities, the senior citizens who may not have the latest technology sitting in their homes. Is it in the fact that the royalty, you know, the companies don't wanna be paying or you can just download this stuff? Is there any incentive for anyone to address these sort of lost communities? Do we need some sort of legislation or? No, I think the kids, their children are going to take care of like you're mentioning kind of older people maybe in rural areas. I think their children are going to insist that they get a smartphone, which is what we've done with like our parents or get a Alexa or some kind of video enabled voice assistant device in their home because that gives us an instant way of talking to them and reaching out to them or having them reach out to us and see us face to face, even when we can't travel to their homes. And so I know we have done that as a family. We've invested the money we've made ourselves to buy equipment for our parents. I once offered my sister to buy a video phone 2500 for $1,500 for her. And she said, well, she would use it only when she was talking to me. Go ahead, Ed. Well, I mean, that is the solution. There's no doubt about that, but that implies that you buying a smartphone, you're going to have a 3G, 3G, 4G type connection. There are places in the US that don't have that. There are places in the US where people are still using dial up. So I think we might have to have an infrastructure, some sort of, it might require some sort of legislation, which I know in the US, we're sort of reluctant to do these types of things. You know, there was a question there, it says, aren't we always going to have not communities? Well, yeah, but that doesn't mean that's a good thing as far as I'm concerned. I mean, particularly when we have seen that the value of having an internet connection is so important, particularly for young people and their educational experience. So I think we need to be able to address this. Now, is 5G going to do it? Well, I hope it is, but you know, there are some deployment issues in the US with 5G. And you know, the other question is when you, you know, you're going to put a cell phone tower up when you know, the population density is so low. And I think those are issues. I'm aware of a situation of American Indian community on the far west coast of Washington state. And what they did is from their hunting licenses that they sell people to hunt on their land, they saved up over two years and bought their own cell phone tower and actually paid a provider to come light the cell phone tower up and provide service. So I mean, things like this can happen, but it can be difficult. So There's no doubt. Look, internet service is as important as electricity and running water. I, you know, just ask yourselves how long you would survive at home without, you know, one of those three. And so, and in fact, you know, I appreciate that, you know, from a cost benefit analysis, you know, sending a tower to a underrepresented, you know, it's a small community might not be cost efficient initially, but if you think the benefits of educating that community, in other words, the kids will have access to the internet that will learn faster, they will learn more, they'll be able to be, you know, big citizens that make money, et cetera, rather than, let's say, expect the subsidy from the government. So if you look at it as a total cost, in fact, you know, early, I think everybody knows, and it's been proven time and time again, early intervention in communities, meaning, you know, the younger the kids are, the more access they have to training information, the better it becomes afterwards. So I would say is that, you know, it's a great investment, right, to actually require this sort of deployment of equipment, because you save a lot of other money you will have to spend down the road, right, to deal with other types of problems that will arise in the lack of this environment. Now, we may get some solutions with this, you know, Elon Musk's company, I can't remember the name of it, that they're launching all those satellites. And, you know, I've actually volunteered to be an output tester whenever they start allowing people to test. That might provide some help. And there were other companies that looking at this satellite internet, the experience though that we've had in the US with satellite-based internet services has not been very good up to this point. So we'll see. The US is an interesting case, guys, because of cable, which I don't think translates to any other company in the world, like, you know, so in the US, there was so much proliferation of cable due to TV that it became the predominant way to get internet access, right? Whereas in Europe, for example, it was DSL for better or for worse, right? So unfortunately, I have a very poor, you know, corporate-based connection that happened to be in downtown Athens, very old circuits, you know, a lot of noise in the circuits, you know, I'm waiting for optical fiber to reach my building. Again, I'm keeping my fingers crossed. Right. But you guys are like, in my New York apartment, I had 100 megabits in and out. Not related, actually. I was measuring 100 megabits in and out, okay? So, you know. I mean, I'm sitting in Fisher's, Indiana, and I have a gigabit fiber line into my house. Yes. For under $100 a month. I mean, it's insane. It's insane, yeah, insane. So, there's a question about HLS versus Dash. Ed, do you have any thoughts about HLS and Dash? I understand what the person was asking about. That's not interactive video, is it? That's great. Well, HLS was invented by Apple when everybody else was switching to Dash. I think that Dash is probably gonna win out on this one. I don't know, Alex, I'm sure this has popped up with your companies. What do you think? Is this video conferencing? No, no, it's streaming. It's streaming. It's streaming. They're both streaming, yeah. Okay, so let's talk then about internet musicians and the requirements of latency and so forth to be able to synchronize and so forth. Any thoughts on that? A lot of thoughts, actually, because I happen to be a practicing musician. So actually, you know, playing in professional gigs. And in fact, I also did some research, because of my involvement in communication or in video communication, a national question is, okay, what could you do for music performers, okay? So we explored this question several years ago. And in fact, a lot of work happened, I've done for many years ago. I said, I mean, this is not a new problem. But the reality is that the demands for latency are unfortunately in order of magnitude lower than it is for regular communication. So we're talking about less than 20 milliseconds and 20, okay, and even lower, okay? So to this day, unfortunately, I have not seen something which would satisfy regular professionals. Now, if you train, if you train to play with somebody so that the third person can hear you, you can probably do it. But not because you will enjoy playing with your fellow partners, you will change yourself to play out of sync, but it will be sort of in sync with the listener. You know what I mean? Isn't 20 milliseconds a challenge based on the distance between the U.S. and Greece? I mean, it's not just the technology in the middle, it's the speed of... The 20 millisecond is a physical limitation, right? But of course, if we were to collaborate between U.S. and Greece, yes, the speed of light would prohibit sort of collaboration and the story. Unfortunately, again, there's no way around that. By the way, I think that's why, by the way, in big orchestras, you also need the conductor. Turns out if you calculate delays between different parts of the orchestra, right? Without the conductor, it wouldn't work, right? Let's say for the far out violins to listen to the... On the other side, where the bass players are. The delay would be so high that it wouldn't be in sync. They would be racing each other. So that's why the conductor was used as soon as the orchestra became big enough to become acoustically... The delays to become acoustically significant. Anyway, that's my experience. Again, I don't consider myself an expert, but due to COVID, I actually played a little bit. I played with Zici and other systems, but nothing really helped out. So if anybody from the audience knows any better, I'm eager to... There was a project that was being funded by the National Science Foundation. There was some work being done at USC, was part of the engineering school and also the cinema school. And they were trying to do this synchronized playing. And they came up with some interesting tricks to be able to do it. They published a lot of papers in this area, but I don't know if that evolved into any technologies or standards or anything like that. But I know people have tried to do it. I'm not aware of any good solution. I think this kind of goes back to some of the things I want to talk about, which is like the delay in audio, and that there's a couple of things here. Like there's a fatigue because the delays between people speaking and their reactions is just enough that you have to think hard about when you're gonna interject yourself. And it's also very hard to talk over each other because the software will start muting down one person and bringing another person up if they think that maybe you're just background noise and not actually the primary speaker. There's different other platforms that do the treat this a little better. For example, if anyone's used Discord, it's kind of popping among the gamers, but the Discord servers are really good about having multiple people talk simultaneously and you can hear it. And you could actually kind of make it out as if you were friends in a room. I have a friend from back when I was in San Francisco who had a startup got bought by Dolby Labs that specialized in spatial audio on a call like this. So we could both be talking and a third person could start talking simultaneously and you would hear them as if they were in a different space in the audio sound stage in front of you, which is I think it's almost needed for a big group of people to be on a call and talk together where maybe you want to interrupt someone. I mean, maybe it's not socially nice to interrupt someone but you'd want to do it in an effective way that just wasn't jarring. So we've come up with a lot of things that are wrong with video conferencing here. How about are there any laws or regulations that limit a provider's ability to harvest the user data or to do screen grabs or to otherwise grab information from a call that might be private? I know that Ed and I have had discussions where he's concerned that maybe people will figure out how to do financial crimes via video conferencing. Any thoughts on what might be necessary to try to address these issues? Let me, from a technology point of view, let me point out that the W3C, there was a lot of work that went into designing how you can do content setting. In other words, I'm on a call with my browser and I want to set one of my windows, right? Unless you are very careful, you can open the door for somebody that is in a call with you to behind your back, read another window, not related to the call you're having, any other window on your computer, right? And just start scraping data without you knowing it, right? So W3C went an extra mile, so to speak, right? To make sure that this can absolutely not happen, okay? So that's a no-no, okay? I don't know about laws, by the way, if there are any laws that prohibit you, but certainly from a technical point of view, you have to give your positive acknowledgement to any application actually on the computer, certainly with W3C, right? But I think on any computer these days, you need to be given affirmative approval. Now, you might ask, where is everyone? I get all these pop-ups, let's say yes, no, yes, no, and I keep clicking yes, without really thinking what I'm clicking, okay? Like a lot of people never read any of the user agreements, right? You might be signing off your home for all you care, right, you're just clicking yes, yes, yes. Anyway, that is an issue, right? But I think there is sensitivity in the technical community for this matters, and I think, you know, just, sorry, another thing, end-to-end encryption is along the same path. In other words, encrypting the data, not just from myself to the server and then from the server to you, but encrypting it all the way from me to you, so that the server in between cannot see a video or hear a audio. That would be the ultimate encryption because nobody in the middle can read it, right? So some apps like WhatsApp have it, some others do not, okay? And Zoom apps was criticized for not supporting some of this encryption, right? And sort of suggesting maybe they might. Anyway, I don't want to go into the details of that, but there is sensitivity these days in people about who has access to their data, okay? So that's getting better and better, we're not quite there yet. It is, Zoom, I mean, this call is encrypted and Zoom has done a better job. I mean, and that also turns on policy. So- The question- Hey, but- What encrypted? Encrypted, if it's hop-to-hop, it's still encrypted. Yeah. So encryption is not the same as end-to-end encryption. I understand, but I think Zoom is not using end-to-end. I think they're using end-to-end encryption now. Along that same line with Zoom, if you remember they had announced on their free service, they were not gonna allow encryption because they were concerned about people committing crimes, particularly child pornography issues. And there were such a big outcry that now even the supposedly the free version of Zoom also has encryption now turned on. So I mean, I think this is gonna be an issue. Look at, we're all staying at home, criminals still wanna get their hands in our pockets. So the question is gonna be, they're gonna try to use this technology to commit crimes, it happens, okay? And I think we need to be cognizant of that and to make sure that the people who are inventing support in these apps, it's not all gonna be 14-year-old boys wanting to show a video of a dog chasing a cat or something like that. And I think we need to be careful about that. This has been happening for years. There's very rich princes in certain countries that want to give me their money, and they've been doing it over email or phone call now that they just have a new medium. That's correct. And I think that, but I think people need to be careful about it. And we have run into issues. There's a particular type of video conferencing system, I'm not gonna name them. That here at Purdue, when we do work with some companies, they will not allow us to use that particular system because they are worried about security issues. I'm not gonna mention the system. And so the fact that WebRTC exists in the browser allows for us to build those platforms that we can say we trust this piece that we built, we know everything is going on behind the scenes in it. But WebRTC is not completely flawless. And there's issues associated with that. I mean, WebRTC is great. And I think some people have got some really interesting applications that they were able to stand up very quickly by using the WebRTC toolbox. So in terms of, we talked about education and I think that's, in terms of K through 12 college education, I think that's a worm that we can't really eat at this stage. But in terms of grad students and training them for the next generation of engineers that are gonna help improve our interactivity as humans, as our face-to-face, our connectivity, community building. What do you believe are the skills and knowledge that students should be training themselves for for the future? Well, I mean, I think it, you know, as far as this technology is concerned, there's certain etiquette rules that we have probably all violated, okay? But we ought to make sure we train our students. And it's also applies to undergraduate students. What are the, you know, appropriate etiquette rules? I remember when the four of us met to talk about, with Shelby, to talk about this presentation, we spent some time talking about what was gonna be behind us, okay? All of us did, remember? Because we wanted to show something that, you know, that wasn't distracting, that wasn't cute. You know, Alex and I chose our libraries, but I think those are types of things you need to be cognizant of. And you need to, you know, sort of be careful of looking at the camera. And sometimes if you really gotta look away or do something else, I think it's appropriate to shut your camera off and things along that line. But I think that's one thing we need to talk with them. You know, I find, at least me, giving presentations are, I like to give live presentations on situations like that. I will record them if it's required. But I like to give live presentations. And I find that the presentation, I give it the same way I gave it in the past, but it's much more stressful for me when I'm giving it, okay? Is that because there's no audience feedback of any sort? I mean, after I give the presentation, there's questions and answers. I don't even mind people interrupting me when I'm giving a presentation. As a matter of fact, I would prefer that. So some venues where I've given presentations, they allow that. Other people use the hand up feature in many of these systems, Zoom has one, where you can then answer questions. But I think we need to teach our students about using this. And the fact that, you know, in many ways it's different. People behave on this some ways that they wouldn't probably in a conference room. Yeah, and for me, for example, giving presentations, I go so far as to actually stand up when giving the presentation, because I really want my brain to be engaged in present mode. And so I think teaching them, you know, you need to be in the right mindset, you need to have the right environment around you. It needs to be, you know, non-distracting. And go so far as to try and simulate the experience that you want to give, like the, for example, standing up when you're giving presentation. I'm actually sitting in a chair, like you would be sitting in on a panel right now, like a bar stool height chair in my home studio, because I wanted to simulate, you know, being in a panel with like, you know, these amazing people who I get to sit here and chit chat with for the hour and a half is amazing. I guess I was actually, those are really interesting points. I was actually interested in technology things. Yeah, I was about to say, I was about to say, but look, if I may chime in, you know, and I spent like 15 years over my life at Columbia as a professor, and I want to point out the following, that there was no course on antennas at Columbia, okay? And it was funny because, you know, I guess people thought that antennas is all function. I mean, who designs antennas these days, right? Guess what? You know, mobile phones happened, Wi-Fi happened, and companies were scrambling to find, you know, antenna and RF electronics engineers, okay? You know, there were very, very few. Thankfully at Columbia, you know, Giannis Sividis was there and, you know, he saved the day. But what the point I want to make is a following that you should never forget the fundamentals. And I think especially in an environment of early education, and I would say undergraduate and graduate studies is early education, okay? I will still stick to the fundamentals. In fact, do the hard math then, you know, so, you know, we're talking about the latest and greatest, but I think when you're starting out, immerse yourself in the fundamentals because nobody has a clue what will happen 10, 20 years down the road, technologically, right? What skills you might need, et cetera, et cetera. So I would build up on a, you know, on a set of knowledge which has universal value, okay? And then as I grow, I will adapt that to whatever, you know, I happen to be engaged with at the time. Okay, so one last question to wrap this up as we're reaching the end of time. So I'd like you each to kind of tell me, what do you think is missing from video conferencing now? And this goes back to things like haptics and just being able to share other content more effectively. What's missing? What do you see as what video conferencing is going to be like 10 years from now? Let Calvin go first. Wait a minute. Why is that going to go first? I mean, I'm happy to give it from a community and human engagement perspective because I'm not nearly the technologist that the other two here on the panel are. For me, you know, what's missing and what's coming next is going to be that the protocols for online engagement between people are going to enable those moments of like serendipitous meaning, I think, better. We're going to have that ability to have clearer audio, to be able to talk amongst each other in a way that just it makes it smoother and more natural. I don't think we want to exactly try to simulate or replicate the actual in-person experience, but we're looking to replace it with something that's even more interesting to the participants in the conversation. So leveraging tools like augmented reality, I think will help actually, those are things we've not even talked about or thought about, but going into a full, not necessarily VR, but maybe more AR where we're leveraging these tools to augment our experience, to make it better, to make us better humans. So we're not trying to replace the experience for not trying to have artificial intelligence do our thinking for us, but we're trying to have things like that enhance the experience, give us additional data, give us additional context and additional environment, allow us to choose our path through these social events in a way that makes it easier to basically tap someone on the shoulder and ask them a quick question. So I think those are the things that are missing now that'll be coming in the future. Any other thoughts? Alex? If I may, there's no doubt that we have a lot of tools, we have a lot of tools, but again, I find myself doing a lot of hard labor in front of my computer. I find myself typing stuff, copy, pasting stuff, moving files from one place to another. I have certain ways of organizing my data that I particularly like, which I think makes sense. In fact, I can find notes, meeting notes from a meeting like 10 years ago now. So, but I designed that on my own and figured it out on my own. That's hard work. And it's the result of a lot of work that happened over many, many, many years of struggling with all these things. So I think we're at the point where we're hoping that tools will come around to help us solve that. So, so innovations like Slack, for example, are in that direction, right? I think that, you know, to cut this point, we need a lot of innovations in human-computer interaction like the iPhone, for example, was, right? So forget the keyboard. Everybody thought you absolutely need a keyboard, okay? Nobody would sell a device without a keyboard. And then Apple comes and says, no, you don't need a keyboard. And of course, nobody has a keyboard since, right? And we are so much better for this, right? Because it's so easy to interact with this, right? You just grab. So right now, guys, I'm sitting in front of a screen which is 49 inches wide, okay? So I have my laptop in front of me and then there's a big screen. And in fact, with all those video windows, a real screen real estate is very important, right? So my ability to interact with it, maybe if I could touch it and move it and shape it, that would make my life so much easier. So I think application functionality is missing, for sure. At least for me, maybe I'm most in view with my data and that's why I feel the pain, but I think in general it's missing. And as Calvin said, the human computer interaction piece means a big boost, so it makes it more easy for us to interact. I'm always with the keyboard typing this and that, okay? And you know, I used to use Emacs for the older guys in the- Oh yeah, oh yeah, yeah, yeah, yeah. And we were proud that we were able to use the computer without lifting a hand to touch the mouse, okay? Everything was happening on the computer. Yes. Alex, I still use VI. I'm a VI user. Sorry, Alex. Anyway, so is Amy. Okay, I mean, I think I agree with Calvin and Alex. Particularly, Calvin indicating we should not try to make it be like we were in the room together. We should exploit other things to in some ways make it better, okay? And I agree that I think we're gonna have to have better interactivity tools. Maybe these haptic things are one thing, larger screens. I think with the advent of some of these newer technologies, both on the audio and video side, we're gonna be able to do a lot more with maybe a lot less bandwidth. And so we might be able to have two of these conferences going on, two of these streams, interactive streams or multiple of these interactive streams going at the same time. And that would be interesting to see how that all works out. So I think we're gonna see this get a little bit, it's gonna get better. And I think we're, for the people out there, I think the stuff that's free is gonna get better, okay? And I think we're gonna see more and more of that. I think many companies were surprised how well this worked. And so it'll be interesting to see what's gonna happen in the next year from now. I hope we don't evolve into that. What was that movie with Bruce, was it? Replicant? Yeah, Replicant, that movie Replicant, where he sat at home and he sent his Replicant one out in the world and did his job for him. I hope it doesn't devolve into something like that. So yes, hopefully not. Well, with that, I'd like to thank everybody. I'd like to thank our three panelists. I'd like to thank the audience for their questions. And I recently read some article online that said, why do we wave at the end of a Zoom call? We don't wave when we get up out of the room. So in any case, thanks. I'll really appreciate the conversation with all of you and take care. We'll see you later. Bye. Thank you very much. Thank you, Amy. Thank you.