 Welcome to another edition of RCE. Again, this is Brock Palin. You can find old episodes and subscribe in your favorite podcatcher at RCE-cast.com You can also find links to all of our Twitter's and blogs and stuff on there. Also on my blog specifically, my group at the University of Michigan is looking for a Linux person. So you can find our job posting off of my blog there. Otherwise, you can use the contact information RCE website to get ahold of me. If you may have someone who might be interested or you want to introduce to us or if you're interested yourself. Also, Jeff, I think you've got some stuff coming up. Yeah, we do speaking of hiring. My group is hiring as well. I'm in the server group at Cisco and we need some Linux kernel hackers for some network work. So either ping me or go look at my blog. I put a couple of details out there and let me know. We're always looking for good Linux kernel hackers. Additionally, EuroMPI is coming up. The deadline for that I think is in. Well, it depends on when this goes. This podcast goes live. But the deadline I think is May 17th for paper. So get your papers in. We're going to Vienna this year. Be a good time. Vienna. Nice. So let's get into what we're talking about today. So today we're going to be talking about a group in an effort more so than there's actually no software involved with this. So we're going to be talking with two different people about data without borders and we'll let them explain it. But I have Nikki Rota, who's another graduate student here at the University of Michigan, as well as Jake Porway, who's one of the founders of the data without borders movement. So guys, why don't you take a moment to introduce yourselves? Hi, I'm Nikki and I'm a master student at the school of information here at the University of Michigan, studying human computer interaction and information analysis and retrieval. And I'm Jake Porway. I'm the founder and executive director of data without borders and was previously at the data scientists in the New York Times Research and Development Group. So what is this data without borders project and what are your goals? So data without borders is a nonprofit that connects expert data scientists with high impact social organizations so that they can better collect, analyze or visualize data in the service of humanity. Our goals are to connect data scientists who have extra capacity and want social outputs for that with social organizations who are doing good things on the ground and may have a lot of data or be influenced by data but don't have the capacity to look at it themselves. Now when you say capacity, what do you mean by that? Do you mean equipment? Do you mean expertise? All of the above? That's a great question. It's really all of the above and a lot of times even just the mindset of thinking about data. I think our long-term goals are actually to have a world where every social organization has enough data capacity in terms of thinking about data and planning out their projects as someone like Google or Amazon. We'd love it if it were as crazy for a mission-driven organization to plan a new project and not think about how they were going to use data as much as it would be for them to plan a new project and not think about a budget. So we're here thinking really about the mindset of realizing that data is important and then in addition to that the capacity that comes from needing people like resources, people who know what to do with that data, as well as a lot of the technology infrastructure. And we know that people in Silicon Valley and Wall Street have a lot of these things but the nonprofit and the social sector tends not to have the resources for that all the time. So is this like a commercial venture that's going to be funding itself or is this a nonprofit effort and how do you have this structured? That's a really interesting question. It is a nonprofit structure but we don't think that being a nonprofit means that you have to forever rely on grants and donations. So we have plans to make this financially sustainable and we're working through a number of avenues on that both from charging for services to licensing tools that we'll be creating as well as providing educational resources so that even groups that we can't work with directly could still learn from our webinars or from books that we put out on how they can think about data differently. So we are a nonprofit but we want to be sustainable as well. So let's talk about you specifically a little bit here. How did you originally get the motivation to do this and what was kind of your starting efforts to create data without borders? Well, as I mentioned, I was working at the time as a data scientist and I couldn't help but feel that we are at this really exciting time where there's so much data and so many tools to use data that you realize that very soon a lot of the world's decisions are going to be driven by data. And I was excited not just by that potential but all the excitement in the community around this. I'm sure a lot of your listeners and you guys have heard of hackathons. They sort of weekend events where people come and work on problems in their spare time. And I was so inspired by those. I went to my first hackathon. I was really kind of pumped up because I felt, wow, this is an amazing time where all these people are sitting in this room with the most amazing skills the world has ever seen. They're all great coders. They've got great machine learning skills and they're spending their spare time. This isn't work time. This is their weekends and their nights coming together to build something. This is really amazing. This is incredible. And I couldn't wait to see what we're going to come up with. And by the end of the day, the results that we came up with were like a little bit unfulfilling. There were a lot of mobile apps to help you find a restaurant. There were some programs people had come up with that were kind of like Twitter clones. I just thought, man, it's kind of tragic that we've got all this excitement in the community. And yet what we're coming up with are these sort of iPhone apps. And so I just put out an open call to my friends in the city and the data science community to say, hey, does anybody else feel this way? Does anyone else want to do something more socially conscious, find ways we can use our skills for good? And so it actually started as just a blog post amongst my friends saying, hey, what, you know, sign up on the email list. If people are interested, maybe we'll find some good data, maybe city data, maybe social data. And from that, we'll see if we can do something that's a little more socially positive. And I learned a lesson that day that you should never put anything online unless you want everyone to read it. Because somebody tweeted it and it sort of went viral. And before I knew it, the UN and the White House were calling and saying, hey, how do we sign up for this? And that was when I realized, oh, this is a potential to be something much bigger. Clearly people are interested. So we've been taking it forward ever since. So let me take a step back and ask you kind of a dumb question. What is a data scientist? What does that mean? It's a fantastic question and one that is hotly contested within the nerd community. It's become this new invoke term that kind of encompasses statisticians, data miners, data analysts of your, and tries to bring them under this umbrella now that there is so much data. So it really refers to people who are dealing with large amounts of data and trying to find insights from it and whether that's the guy who's programming the infrastructure to hold terabytes of data to the statistician on the other end who's actually mining that data to find patterns. And so the, while the definition is still up for debate, people tend to see it as kind of a mix of computer science. So being able to write code to get data and manipulate data and statistics. So being able to understand the, you know, the possible sources of bias and how you should deal with data and how to actually do some machine learning around it. As well as sort of a critical mind where the, I think, where the scientist term comes in, someone who can ask the right questions and drill down to that. So it's very much in flux, but basically kind of a coding stats, you know, numbers minded scientist looking for patterns in data is the way I see it. So Nicky, how did you get involved with this? Great, great question. When Jake sent out that blog post, I heard about it through some of the channels that I subscribed to. And believe it or not, I was actually in Michigan at the time when I heard about it. I was visiting the school for the first time and getting to know Ann Arbor. And immediately thought that this was something that I thought the school of information should get involved in. Somehow, somewhere another, there were people at my school who would want to join. So I met with some of our administrators here and some of the students who were on campus. And there seemed to be a ground swell of people who thought that this idea was awesome. And I said, great, I'm based in New York City. I'm going to contact Jake. I'm going to tell him that this is awesome. And that is exactly what I did. So the rest is history in a sense. We met up, I told him, and this was just when it was starting. Jake, we must have met in July or August of his past summer. I don't think we even held a single event yet, or we were just still building up and on. So yeah, Nicky was awesome and showed up and said, hey, this is great. I've got a ton of ideas. And we met up and it instantly clicked. And I basically said, you know, I'm at a school where this is what I'm intending to think about. I actually came to the school of information hoping to look at how nonprofits deal with information. That was specifically why I was coming to the school of information. I had, my background was working with various nonprofits in India in particular. And I saw what challenges they faced, not only meeting the reporting requirements that their donors had said, but also in terms of managing the finances of a nonprofit. When you've got funding from five different grants and these grants are all requiring that you spend the money in different ways, you can only buy one third of this employee using this grant and you can use the other two thirds of the employee using that grant. It seemed really confusing and I felt like this was a space where I could do a lot of good. So when Jake had this call for data scientists who were interested in helping nonprofits and I was starting at the school of information, the timing just seemed too perfect for me. So I wanted to get involved as soon as possible. Now, like Jake said, they hadn't actually held a data dive at that point. You guys were planning the data dive for October, the first one that you were holding in New York City. And I said, well, if you guys can't necessarily come to Ann Arbor, how would you feel about me putting or hosting one at our school? And Jake said, run with it, go, have fun. This is entirely the philosophy of our organization is to help people help others. And I got a lot of support from Data Without Borders when we put on the A2 data dive in February. And our constant conversations and talking about how to organize such an event really helped me and helped our organization develop on the ground here. Okay, so let's hold off on the data dives for a second. Let me jump back into something that you said a second ago that I didn't quite understand even when I was researching about this before the podcast. Little secret there, not supposed to research at a time, but I do. So I didn't realize I had always been thinking about this in terms of big data types of things that diving into publicly available databases and finding where groundwater is or whatever type of thing that nonprofits are geared towards for the betterment of humanity. But you were talking about even more mundane yet equally important types of things, reporting requirements. Is that a big problem for nonprofit organizations? Are they three to five people trying to do something and get bogged down with administrative? That's something that you can help with? Yes. So a great example that I have from my own experience is I was working with a nonprofit in rural India who was, they were trying to collect data on farmers that they were trying to help. And in these instances, their donors are wanting to know every aspect of these farmers live. So you have a community of maybe 20,000 families. And the donors are interested in knowing the demographics of the family. They want to know how much these families own. What are their various assets? How much do they spend on farming equipment? The level, the minutia of the detail that was required for this particular NGO to collect is mind-boggling to me because I hadn't had prior experience seeing the depth to which these nonprofits are required to create sort of transparency around the work that they do. So they have consultants to help them collect this information, but then processing it, analyzing it, knowing how to show the statistical significance of the impact that they're having. There's a lot of work currently on randomized control trials, particularly in the developing world. Statistical methods are being applied to how nonprofits have to think, how they show their impact, how they demonstrate whether or not they're actually doing the things that they say that they claim to be doing. And as we move to this more scientific approach to how nonprofits operate, the shifting skill set of what's necessary for a nonprofit to know how to implement, to know what data to collect even in order to demonstrate what work they're doing. Jake really hit a nerve when he thought of this idea is that actually there's a shifting need and data without borders is really trying to address these needs that are holistic in approach, not just analyzing the data, but even thinking about how and what types of data to collect. Okay, so that's fascinating to me. So this is not just your standard 501-3C mountain of paperwork that needs to be returned every year, but an additional level of reporting for nonprofits to be proving that they are doing what they're doing. I hadn't even considered or thought of that before in my own little narrow mind view. Let me ask you a trivial follow-up here. What's an NGO? You said that a couple of times in there. So a non-governmental organization. It's a word that kind of encompasses a lot of different types of institutions, but your traditional nonprofit generally. So back earlier in one of your answers, you mentioned something called a data dive. What is a data dive? Well, so a data dive is a weekend event similar to a hackathon where we bring data scientists together with three social organizations of community groups, nonprofits, who have data problems that are targeted but don't necessarily have the capacity to look at them. And we want by the end of this weekend for the outcome to be that these organizations have worked alongside data scientists, hopefully gotten some answers to the problems, whether that be a visualization, an analysis, even a consultation to say, hey, we couldn't quite answer this question, but next time if you collected this data, we might be able to, but alongside those results, have them think differently about data. So it's sort of a souped up hackathon that really focuses on collaboration and a wide range of outputs. So now do you have some kind of process for NGOs or nonprofits to go through to become eligible to attend these data dives? Like do they have to meet with you first and say, yes, we have data problems and you talk about it and determine that they actually do have data problems, or how does this all work? That's a great question. That's actually one of the things I think is one of the most powerful of the data dives. And that is that we have an open application process, but when we go to select the three organizations, we spend some time before the event talking with them about what their problems are, because a lot of times they don't know what they don't know. And so we get a data person alongside a social organization leader to iterate through with them where they say, hey, here's the problem that we want to solve and here's some data we've got. And the data scientists will say, okay, well, that's an interesting question, but you can't quite answer it with best data, but you could answer X, Y, or Z. Is that interesting? And the person on the other end would go, oh, that's it. I didn't know we could even do that. Yeah, Z is interesting. And what if we, and so in that process, they actually start to think through how you go from wanting to solve sort of a business question to formulating that into a data question and sort of exploring all the things they can answer. So we work with them ahead of time to go through that process, and that helps us narrow down a question, a set of data we think would be good. And of course, that's also necessary because we don't have a month, we just have a weekend. So we really look for targeted problems that we can hopefully chip away at a little bit in just about a day. Now, is this something that the organizations themselves, they bring their own data, so to speak, that ahead of time with these consultations and whatnot, you kind of counsel them and say, all right, for this weekend, we're going to need you to bring this database and that database and whatever other data source you've got or a BYOD event. Exactly. So a lot of organizations do have their own data and they know either they've collected it because of a project, but they haven't been able to look at it or they suspect that it's useful. So a lot of times organizations will bring their own data. However, we also realize there are huge numbers of open data sources that are becoming available. So governments and people like the World Bank are opening up this open data. So one of our jobs when we talk with people ahead of time is to help understand what other open data sources might answer their questions that they may not even be aware of. So it's a combination of the two. Some groups bring their own, some combine theirs with open data or some just use open data altogether. So it sounds like you're really describing in the commercial where we always talked about business intelligence and businesses have invested in this for a long time. You're basically bringing that to the nonprofit space. I think that's exactly right. And it's a good time for this. Not to say that nonprofits don't already have business intelligence built in. A lot of them have the resources for that. But I think what's changing is that the amount of data that's publicly available that could come to bear on these problems is vastly increasing. And because of that there are new opportunities for organizations that maybe didn't have business intelligence or weren't thinking about it quite the same way to bring in this new data to think differently. So for example you could imagine a data set from the census becoming available or about low income housing that a community organization that looks at bringing cheap affordable housing to people could take advantage of could learn more about their space. And so even if they have business intelligence about their practices or about their donations they may not have had this resource before. And it's a very new skill that a lot of them, a lot of organizations are still learning and that we hope to bring to them. Yeah so just like a business trying to be most efficient you're trying to make the efforts of the nonprofits be most efficient so they have the biggest impact with their resources. That's quite interesting. But some of these things I would think would take a lot more than just a weekend in these data dive things. Is there other things that Data Without Borders does to do long term? Yeah that's a really great question. So the data dives actually began as a proof of concept. When we originally started the idea we envisioned a global network of data scientists being able to respond to that that signal in the sky, any organization that needed data help. But we knew we couldn't really start there so the data dives actually were just meant to help us understand what the common problems were and whether people were interested. And they've been really great and actually have produced a lot of long term engagements beyond them themselves. But we also used them to transition to what we wanted to see which was these long term engagements. And so we have a group called the data core which is volunteers that work in their part time or weekends on volunteer and contract basis where they work for a couple of months with an organization and there they can really tackle bigger problems and they can iterate further. In addition to that we're also building out a fellowship model which would be a year long commitment from someone to actually embed with the organization full time. And in addition of course we have on staff data scientists who work on projects as well. So we actually have a range of engagements and we've been really excited to see that every time we try one like a data dive it sort of naturally leads to the next one already. So we've been really excited that people are really sort of naturally proving out this model for us. So Nikki could you give us a little detail about what you hosted at Michigan and how has that shaped both what you're doing there and what Jake is doing with Data Without Borders? So we tried to mimic the same sort of model of the data dive just with the scope being a little bit smaller. So Data Without Borders has been great. They've been able to pull in some really amazing and large organizations of ACLU I can think of the United Nations. We decided to focus basically just on Ann Arbor and Detroit based nonprofits to do a very similar type of event. We've brought in the African Health OER Network which is based out of the University of Michigan and we've also brought in Focus Hope which is a nonprofit based in Detroit. And what kind of things did they need answered? Is this something that you can talk about? Yes, and actually you can go visit our website and we've got several stuff up. Even on the Data Without Borders Wiki you can actually see the output from our event. The Data Without Borders Wiki was a great way to kind of demonstrate the flow and the process that our participants went through to actually create output for these nonprofits. But yes, the two groups had two very, very different question sets. Focus Hope was very interested in learning more about the participants who take advantage of the programs that they offer. They don't know much about how much overlap there is in participants and or where the participants are based in Detroit. And so they gave us lists of participants and we actually were able to map using GIS software where their participants were based using some open data sources from the census. And we were actually able to pull in some information about income levels and the various census tracks in Detroit to sort of help them get a better sense of what's going on in their community. And with the African Health OER Network this is a nonprofit that is actually providing open educational resources to universities across Africa. And they were interested in two questions both related to their network. The first was who in their network of doctors, of educators are participating and contributing to the network. Who is contributing materials, who is going to their conferences. And then the second question was which materials that they provided are being taken advantage of. So the videos that they've put online, who's viewing the videos? Where in the world are they viewing the videos? How often are they viewing the videos? So tracking the Google analytic data on some of the various resources that they've been available. And in both cases for both nonprofits we had some very interesting findings which was a lot of fun. So we're all a bunch of high-performance computing geeks, Jeff and I. And I was invited down to the A2 Data Dive and I gave a little talk showing high-performance visualization for massive data sets. What sorts of things can data crunchers like ourselves who are really just high-performance computing guys help out somebody like Data Without Borders? Something that I took away from the first Data Dive that I went to in New York back in October and then I saw again at the A2 Data Dive in February is anyone can contribute both in terms of your skill set but also the sharing that goes on between the participants. So Jake talked earlier about kind of the watershed moment when he realized there were so many people that were excited to volunteer their time. And we saw something similar at our event with how many folks, not just in the Umich community but also in the Ann Arbor community were excited to share their skill sets. So we had folks who were really great at network analysis. We had other folks who were great at doing geo-processing. And in this space, with all of these different heads coming together to face a problem, there was a lot of sharing of information, of toolkits, of skill sets. So it doesn't much matter what exactly you bring to the table. It's being part of the conversation and the dialogue that can help yield and bubble up these really interesting solutions to the problems. Yeah, I think Nicky really hit the nail on the head there. You know, at her event, I think she saw something really incredible which was how excited people were to use their skills in this capacity and to realize that there really was a place for everyone's skills, whether you're a high-performance computing guy or statistician or just somebody who knows a little bit of Python. And in addition, even though this is a conversation primarily about data without borders, you know, we'd be remiss not to mention that there are similar efforts going on in different spaces where people could use their skills as well. There's sort of this new civic coding or kind of hacker activism growing up in groups like Code for America trying to bring together developers to help with local communities, especially through government. There's also Scraper Wiki. So if you're not familiar with Scraper Wiki, they basically have a platform where people can say, hey, I found this data source online, but I don't know how to read it. It's not machine-readable. Maybe it's just a bunch of PDFs. And then developers can sign up to solve that problem. Actually, you can write a scraper in PHP or Python that will deliver that data in machine-readable format that the other person can then download or visualize. So there's sort of this growing movement of ways that people can use their coding skills for the greater good. So let me ask you a question on a slightly different tack here. Being a data scientist yourself, you must also value data, obviously. And so therefore you must also be gathering data on yourself, on data without borders, and have some success metrics and things like that. What kind of things are you tracking about yourself? Do you track about the long-term viability of the answers that you provide to people and are they able to sustain that and how well they do with the answers and the knowledge that they gain? And what are the kinds of things you look into? That is such a good question. And while we do have analytics around our day-to-day functioning, how many people are signed up, how many projects are active, what's volunteer retention, a lot of our impact is measured based on how we help others have impact. So like you said, a lot of the indicators that we're actually looking at, and what I said looking at, we're young, so we're actually just beginning to collect these, but are on exactly the questions you asked. How often are the organizations using the results that we gave them? How can we measure their impact? I mean really if a lot of them are trying to do is use data to show that they're doing their mission better, then we show that we're doing our mission well by showing that we can show that they did their mission better. So a lot of the metrics that we try to figure out early on are the kinds of metrics that would show that they had an impact. And I think Nikki gave some really great examples of organizations thinking through those problems and how that's a very tricky game to play. Jake, more than the last data dive you guys did in DC, weren't those some of the questions too that the nonprofits were asking That's a fabulous point. Thank you very much for bringing that up. So Nikki, ever the rock star, was at our DC event, so not only holding her own events in Michigan, but traveling across the country to come help data do-gooders in need. And she saw there, she was referring to a group called the National Environmental Education Foundation, and basically they try to increase environmental awareness around the country, and they basically do that by telling newscasters or healthcare providers information they should be sharing with their constituents. And the big question for them was, how do we know that our message is actually getting out? And more importantly, how do we know people are doing anything about it? And they didn't really know where to begin, and so we spent the weekend, the volunteers came up with a strategy for a number of different data sources that they may not have thought of before that could help them understand that problem better. So there it was not only, let's try to help them with the data problem that we can then measure the impact of, the problem itself was, how do we measure this very difficult impact question? I think we gave some really good answers to that. So Nikki, so we did this A2 data dive this last year, and are we going to be having any more of these going forward? Oh, yes. We're excitedly getting ready for next year. Obviously, you know, all of us are winding down in the semester over here, but we've already got a formula for how we want to structure fall semester and spring semester. And the idea is actually to start bringing in the nonprofits over the fall semester for what we're calling data jams, like small afternoon, maybe two or three hours information sessions so that the students and the UMish community can get introduced to them and their data, but then also provide that space for these nonprofits in the Ann Arbor area to actually think about the data they might want to collect for our event in February. The issue that we found when we held the event in this past February is that a lot of the nonprofits in our community are small enough that data might not be actually a primary goal of a lot of these institutions. And so helping them think through these problems of what information is important for me to collect in order to answer the questions that are pressing for me is actually part of the service that the A2 Data Dive can actually provide to these groups. So just curious, you said next year, you mean next academic year starting in the fall, right? Yeah, we already have the dates set. We're probably going to be doing it on the weekend of February 9th, 2013 is when we're hoping to have the next event. Okay, here's another kind of off the wall question. If you had to characterize what do nonprofits need? What do they ask you? Are there two or three kind of common questions that you get asked from these various NGOs? Yeah, definitely. Obviously impact measurement is a big one. But in particular, some of the smaller goals in answering that they ask for a lot are especially visualization tools. I don't say, oh my God, I'd love to visualize my data. Put it on a map, give me a dashboard. I just want to see what's going on. The big problem with data is it comes in this big opaque brick. People can't understand it. We're very visual creatures. So visualization is really one of the big ones. I think another one really is just understanding what data to collect and really knowing what's out there. So they have some sense of what data they have and they kind of know about what other data is out there. But we just want someone to verify that they're doing it right or open up their eyes to what other data could be useful and how it could be used. I think those are really, by far, the two biggest ones that we see and then from there it breaks down along specific business questions usually really to trying to find some kind of predictive analytics around a program that they're doing or just trying to understand the general statistics of what is actually going on in their programs. So what are some of your goals for data without borders going forward? We asked Nikki about what we're doing here at Michigan. What about the national organizations? That's a good question. So I think we're going to continue working on this chain of long-term engagements that I was mentioning before. So continuing the data dives, building out our data core of our volunteer staff and working on fellowships and hiring more people on staff so that we can just get more people involved and doing good things. In addition, we think that a lot of this, we've been able to tackle a lot of problems from the data perspective with social organizations, but there are a lot of other voices that could really be a part of this. So we realize that we don't necessarily have development skills. Once you build that dashboard from the weekend, who builds the software that keeps it running or the content management system that runs behind it? And who designed it to be usable? So we realize that far beyond data, we need to start talking to developers, designers, getting other communities involved. So how do you get the word out about this? How do you recruit for data dives and things like that? You talked about the watershed blog entry at the beginning of time, so to speak, but how are you spreading the word and spreading the excitement? Like have you ever been flashed on it? Or how do you appeal to the mass geek audience to get people to help? So I'll let Nicky answer as well on this, how she found this experience. But we sort of got lucky early on that so many people are excited about this idea. It didn't really take much to get it to spread. So we've been very lucky in having a lot of support from big networks like the United Nations, PopTech, the O'Reilly publishing group, and groups like that who have really supported in getting the word out, and we've mostly just been fortunate in connecting to the right people. In addition, we know our own, so we often hit up meetup groups and other data hackathons to spread the word. But the geeks are a tight-knit community and word travels fast, and so far we've just been putting it out there and it's been spreading on its own. So let me interject actually. Have you actually ever interacted with any of the maker organizations? Because I'm in Louisville, Kentucky, and there's a maker space here and an open-source organization, and it seems like a lot of the ethos that you guys are talking about are very similar in values to all the people who attend the maker meetings and whatnot. That's a really great point, and we haven't actually directly reached out to the maker community yet. Certainly some hacker groups, but mostly on the development side, the software development side. And so it's a great point and another great community to mix in, but we've mostly been talking to sort of the data software side. Yeah, I guess I should clarify. The maker group here in Louisville, there's a strong software side to it, and that's why I was saying you don't necessarily need to build a better bread box to hold questions, but there are software components to these guys. That's a great call, and it's conversations like these that get us to spread out further on those networks, so where should someone check out to be able to set up their own data dive in their location? That's an awesome question, and I'd love to let Nicky speak to this after this answer, but we have sort of a protocol, a best practice that we're very much working through, and Nicky was so awesome in helping us draft and really work through our first remote engagement with somebody who was passionate about doing this, which is datawithoutborders.cc, which is our website. You can sign up to get involved through our Get Involved page, but in addition, if you are really interested in running one of these yourself, please reach out to us directly. We have an email contact at datawithoutborders.cc, which we use for people who are interested in that kind of engagement, so if it sounds fun and something you want to bring to your town and your local communities, you can contact us at datawithoutborders.cc at datawithoutborders.cc. So, again, this is about the documentation that datawithoutborders had created about best practices. That document was instrumental for us in terms of making sure that we had taken appropriate steps to prepare for what kind of outcome we would get, because, again, like Jake was saying, this data-dive concept of knowing the mechanics of what would be useful for us to implement what processes we should have in place, making sure that we had contacts on the A2 data-dive side with the nonprofits, working them through, making sure their questions and their data was prepared was instrumental in pulling off the event. I'd also like to follow up on the point about civic engagement and engaging the community, which is a great space for some of these hacker groups and for practitioners of data science in their work, but also as a hobby. And we contacted All Hands Active. There's the American Statistical Association of Ann Arbor. And what was amazing to us was we had assumed that because it was a University of Michigan event that we would get mostly students, members who stepped up and said, this is an amazing idea. I'm so excited to contribute something that I love to do good in this world. And so having Jake and having Data Without Borders to kind of walk us through making sure that we were prepared for this event, but at the same time reaching out to the community and seeing how many people were excited to participate was very heartening to us. Okay, Jake and Nicky, thank you very much for your time. We'll have this up soon. Great, thank you. Thanks guys.