As we saw at the beginning of this module, network computers offer tremendous potential for remote work. In principle, there's the opportunity to collaborate both synchronously and asynchronously, with people all around the globe in the comfort of your own sweatpants. But we have a riddle. Given that over the last 50 years there's, been this tremendous increase, where network computers have become pervasive. There are still many challenges to be able to collaborate effectively, while remote network computing has the potential to decrease distance, to flatten the planet. We see this, I think a great example is in software. Software development teams are increasingly distributed. Often there's a hub and spoke model, where there's one location which is the center. And then you've got lots of other locations, that are also involved. So for Microsoft, that might be Redmond. For Google, that might be Mountain View. But each of them have locations all around the globe, in part, because tech talent is distributed. Back in 2009, a survey for the Institute for Corporate Productivity, found that two-thirds of companies were expecting increased reliance on virtual teams. I wanna know who those other third were, because I think this is gonna be true for almost everybody. This number goes up as you get to larger companies. You know, I have to believe that since then, its become almost 100% as seeing some sort of increased reliance on remote work. Now a third of these companies said that managing virtual teams was their greatest challenge. And nearly everybody said that trust was increasingly important. And so, what we're gonna look at today, is how can you handle a world where people are all distributed? Or rather, can you handle a world where people are all distributed? Where there's effective collaboration and built in trust. Over a century ago the writer Arthur envisioned a world in which, the then new telephone could be extended to a video telephone. And he wondered whether, as he puts it here distance will lose its enchantment by being abolished all together. The interesting riddle for me, is that in the time that computers have gone from a few tens in defense clusters in university research labs, up to all over the place, the amount of air travel has also increased. So with digital technology, you might think that we no longer need to travel at all. Distance disappears. And in fact exactly the opposite has happened. Why is it computers made us jet set around the globe? Well in part, the digital connection that computers offer means that it's easy to collaborate with people all over. That sets up the initial contact. And then you wanna go meet in person to dive deeper. In today's video, we are going to draw a lot on this wonderful book by Judy and Gary Olsen, called Working Together Apart. They are neighbors just up the road at U.C. Irvine, and have spent their careers looking at how teams collaborate both co-located and remotely. People have a lot of questions about remote work. Like what if one person's remote. Could my team still succeed? Or how about, having people work at home for one day? That clearly adds a lot of flexibility. What about when everybody is distributed? Open source software, for example. Does the technology use matter? Will we be better off if we have high quality video conferencing, or is chat or all of these other things sufficient? How much did the organizational practices effect things? And what things should we expect? What kinds of collaboration should we expect, will succeed or fail? At this point we don't have answers to all of these questions. But we can start to look at a few of them. Having something like a factory be distributed may be tough. But software, it's on a computer already. So maybe that's going to be easier to distribute than other things. What happens when software goes virtual? The answer is complex. Jim Herbsleb and his colleagues have been looking at this issue for a number of years. What they've seen is that, when people are physically separated and you can leverage the skills of diverse organizations, productivity can go up. And so here we have a survey of 362 software projects across 4 organizations from 15 countries and 5 continents. Having lopsided collaborations where one place is big and one place is small. Having multiple sites. Having physical separation between people. All of those can increase productivity by leveraging talent. However, one cost of this physical distribution is that you lose the experience spread of having somebody shoulder to shoulder with you. Interestingly, well in some ways people work a little bit quicker when we become more distributed. We lose a little bit in terms of quality, in Herbsclub and Team's finding. And the impact on profit is somewhat mixed. And so, I think this an area where we're gonna continue to see changes happen. Here's another example, different type of knowledge work. Scientific collaboration. Like software talent is distributed around the globe. And there are many reasons to encourage collaboration across universities. One of those, is that it's a way for that expertise to move across universities. And for comraderie and a broader scientific community to evolve. Sara Kiesler and colleagues, looked at whether grants that were given to multiple sites, performed better or worse, than grants that were focused on a single site. What they found is really remarkable. Most multi-university projects were much less successful than single-university projects. In fact, every time a scientific team added a university, the scientific outcomes went down. When projects were successful, they used a variety of specific coordination mechanisms. But these remote teams used fewer of those specific coordination mechanisms. And I think one reason for that is, everybody has different tools and technologies that they like to use, different work and collaboration styles. And getting everybody on the same page to coordinate in the same way gets increasingly hard, especially because academics are famously independent. We all have our own way of doing things and we like doing it that way. So we get two takeaways from this. The first is that, in terms of scientific incomes, our output is going down as the number of sites goes up. However, the use of specific coordination mechanisms was a big mediating factor. And so, if you'd like to be able to have a collaboration with many sites that's successful, attending to how you collaborate and having effective collaboration is really key. So what is it that you're getting from being face to face? I spent nine years on the faculty at Stanford, and one of my favorite parts of being there, was that the system administrator for our lab was about 20 yards from my office. I don't need a lot of system administration help. Most of the time my computer runs just fine, but every now and then something goes wrong. And it's nice to be able to pop in and talk to John, and be able to get a quick fix. Also, there are issues of how infrastructure evolves over time, where the webserver gets hosted, potential security break ins, and all sorts of other other stuff. And when you share a coffee machine with your system administrator, all of that stuff happens by osmosis. You know you talk about each other's families, you hear what's going on in their life. It's, camaraderie that is built naturally over a period of years. And so if something happens and it's an emergency in either direction, because you're friends you understand and can help each other out more easily. When support staff like system administrators or grant administrators, or things like that are not co-located, you lose all of that informality. The same happens in software teams. The same happens in scientific teams. As we think about how distance collaboration can be supported by technology we need to find some way of being able to have all of that informal stuff happen. And actually think things like the Facebook News feed, or Twitter are very good examples of a different kind fo way that you can hear about what's going on in your colleagues lives, follow things that they're interested in, share stories, that kind of thing. So while the current situation may be that co-located work is preferable, it's possible that thiswill change over time. One of the most dramatic examples of successful distance collaboration for me Is Wikipedia. If you ask yourself a dozen years ago, did you think that Wikipedia would be as successful as it was. I think a lot of us being honest with ourselves, would have to say maybe not. I certainly would have bet against Wikipedia and truthfully, for every Wikipedia that succeeds. There are many, many, many similar systems in the graveyard. But Wikipedia has been tremendously successful. There are tens, hundreds, even sometimes thousands of editors for large articles. And this coordination that happens through the history and talk channels, enables this ability to have articles that are evolving constantly fresh peer reviewed. All of the things we would want in an encyclopedia. Open source software at its best. Things like the Apache system, are another example of when distance work can be effective. And there is no co-located team that can write Wikipedia. I'm at a University where there is a tremendous amount of expertise. Maybe U.C. San Diego could come close to pulling that off, or another major research university could come close to pulling off of Wikipedia. But then you look at it and you see the diversity and there is just no way. There's so much stuff out there, no one location, even a great research university can have all of that happen at one place. And so while some tasks you really do have a choice, between whether you want to be co located, or whether you wanna be distributed for something like Wikipedia. You don't really have an option for something like Apache where software developers are donating their time. There aren't that many people willing to donate their time, who all work in the same environment. And so for many distance collaboration tasks, the question isn't should we be better off co-located or remote? But what are all the new awesome things that remote collaboration enables, that could never be done in a face-to-face world? So what's successful? Well what Judy and Gary Olson have found in their surveys of global teams collaborating, is that when people agree about processes and tools, then you have a much greater chance of success. Also, by and large, co-location is a much safer bet for tightly coupled work. If you need to be back and forth all the time. It's gonna be a lot easier when you're co-located. For loosely coupled work, then remote work can be just fine. You get to leverage the expertise that's globally distributed. You don't need to touch base that much. It is valuable, they found, to be able to use video and other high bandwidth channels for creating common ground. And also it's extremely important to travel often. In the research on distributed software development, one great example of this is a site visit, where part of one team travels to another location. Ideally you want to do this bi-directionally, so it's not just the people at the spoke always traveling to the hub. It's valuable for the people at the hub to travel to the spoke, to see what the local context is there too. When these things are missing, projects tend to fail. For example, if you have a project team that has never worked together before, that doesn't know who each other is. Where they come from different time zones, different cultures, different backgrounds and they're thrown in the mix to work on something right away. The odds of that one winning are pretty darn small. If you want a team to work well together, it works best if they knew each other in person before. So many organizations, for example, will require that all new employees spend some time at the home base, before moving out to their satellite. Or, for example, I still have a lot of research collaborations with Stanford, and because I worked with my colleagues there for a number of years, and I know the organizational setting there. The cadence of the calendar year, the value of the institution, how people's lives generally work. It works much better than if I were to strike up a brand new collaboration, with new colleagues at a new university. In a place that I'd never been with people that I'd never met. So co-location tends to work better for tightly coupled work. Because you get all of that ambient awareness awareness. Will it alway need to be that way? Fifty years from now will we still need to walk down the hall, or share a coffee machinem or a water cooler, or a lunch spot, or bump into each other as we walk in and out of talks and meetings and other things? Could we take some of this social transluncence. Which is a term coined by Tom Erickson and Wendy Kellogg, to see what's visible about people that are at a distance. Can we somehow make that naturally available to others? And we saw that things like Twitter, Facebook, other online collaboration tools can give you some of that. But there's a challenge too, and there was this great New York Times article a couple of years ago that, I think, put it really nicely. That the paradox of ambient awareness, is that each little update is insignificant on it's own. And taken together, when you see these Facebook updates and Twitter feeds, and other kinds of photos that you're sharing, and things like that. Over time it becomes a really rich portrait of your friends, and their family members. Like thousands of dots making a pointillist painting. I think that's a really wonderful way of summarizing it that this article does. The flip side of course, Is that if you have a global team, where we are making many updates available all the time. As the time between interruption goes from hours to minutes to zero, I mean that's what we seem to be approaching in the modern world. There is no more time for quiet reflection. It's a constant beeping and updating and all sorts of other stuff. Physical co-location naturally limits this, because you can only be physically co located with so many people. When you make a globe connected, then the interruption challenge becomes ever more present. Here's one example of a nice social translucence interface that I like a lot. This is the Google Docs interface. One simple thing that it does, is if somebody else is editing the document at the same time, it just puts a little icon up at the top of the screen with an image or their name. And so you can see in a very light weight way, unobtrusively. That somebody else is editing the document at the same time. And, if you wanted to get information from them, then you can click on that and start a chat or have a conversation. Even just pick up the phone and give them a call. And this is a great example of what my colleague Jim Holland calls, beyond being there, that often, when we create digital technologies, we're tempted to try to recreate the physical world, exactly. But man, you've got computers, you can do anything you want, why try and re-create the physical world? Because it will always fall short, any fact simile will always fall short. Maybe by less and less every year and we get the virtual reality. And the higher resolution. But you can do anything. For a creative designer in this space, for all of you that are going to create collaboration technologies. Whether as developers or as. Gluing things together to make your own lives work. You can have things be whatever you want, and this for me is the tremendous opportunity in this space. So in short, what you want to be able to do, to go beyond being there, is to think about the functions that we have from co-located collaboration that are so successful. The ambient awareness, the informal interaction, the trust building over time, the ability to know when people are busy, when they're free, ask quick questions without getting in the way. How can we support those functions distance collaboration? In a form that could be, boy, anything we want at all. We began this lecture by thinking about the potential for video conferencing to collapse distance. And we're seeing video conferencing becoming increasingly used today. What's next? What's beyond Skype and Google Hangouts and similar? Well, one great example was originally introduced by Eric Palos and colleagues at Berkeley in the 90's. And these are personal roving presences, or props. They're telerobots. I could have a physical avatar be somewhere else entirely. So if I'm the person who is the spoke, I could have a physical robot avatar at the hub, that robot could go to meetings. It could knock on colleague's doors. It can do all the things that a physical person can do. This definitely brings you to a different kind of, and much more exciting, collaboration, than just being a video camera on the wall. And now, the challenge is yours. Given all these opportunities we have in the digital world, what kind of collaborations are you gonna create?