 So Chris Hoffman is coming home, he's started here, oh how sweet, so Chris actually did his dissertation without any computing or any worldwide web probably, was it going then? Oh there was some new thing out there called Mosea Air? Yeah that's right but it really wasn't going so his dissertation was on prehistoric metalworking in the glorious islands called the Mallorcan Islands which belong to Spain and they're very lovely and he would go off every summer to this resort and tell us that he was doing archaeological research which he was, he created very lovely dissertation on that and I'm Ruth Tringham by the way for those of you who might not know and I was one of his advisors and at the same time however Chris had this other life which many of the other archaeology students at that time had in what was then called the quantitative anthropology laboratory that was run by Gene Hamill and some other social anthropologists and demographers and that was actually teaching them skills which are data analysis, computational data analysis and other aspects of what he then went on to develop now and so actually a lot of his cohort went on to work in IT, surprisingly perhaps not since there weren't very many jobs in archaeology and so maybe you'll kind of feel sympathetic towards that those of you who may be at that point in your careers so anyway now Chris has been working for a number of years with IT at UC Berkeley at one point he was overseeing the IT services for the graduate division and another point he was working with the what was that media that data thing media yeah the media vault of IT and he now has this illustrious position actually it's not up there now the associate director of research IT and so as with that little hat big hat sorry with that big hat he's going to be talking to us today so that's if I left out anything he's also a good friend and you can know each other for years but I won't say any more details yes oh Chris was the lead singer and guitarist perhaps still is of a fantastic band and when we remodeled my we remodeled my house he came with all his band and did the whole kind of live music and then used to also perform at the aft any parties we would have in the atrium here so this is your future perhaps so any of you who are organ you've got the job of helping to organise parties at the atrium just remember Chris Hoffman very good guitarist and very good singer okay I have to say everything I know about presenting was because we're doing a force meeting presentation we did videos shows all sorts of like experimental kind of media presentation so anything that is either incredibly embarrassing or just wrong like I will credit you in advance so um we're here talking with Nico and Aaron Poolech my colleague in research IT about some of the technology services that are available to you and then to also ask you some questions about where technology should go on campus because I do work now for a group called research IT and we're really focused on research computing in the most diverse kind of definition of research and data and computing that that you could imagine we're also now part of a larger organization that includes educational technology services the Center for Teaching and Learning I'm happy to see my colleague Owen here because we're really trying to look at the campus community from this total perspective of how your your researchers teachers and learners simultaneous what kinds of computing services and information technology services would be helpful to you but while we're doing that we're also going to show you some of the cool things that we're working on that we're collaborating on with the ARF and we'll just kind of take a tour through a bunch of these things so research I see very broad mission here to help the campus research community faculty and students help provide the best IT services possible we divide what we do up into a number of service areas kind of consulting just kind of the heart and soul of what we do Berkeley research computing which has a number of different things it does research data management and then museum informatic is a program that we have for a number of years also so um before I go into that I just want to say that one of the things that we try to do is to we try to be the liaison between you faculty and students postdoc staff between you and the kind of more nerdy technology service providers in central IT so we can be kind of a translator in the bridge so if you have questions if you're having problems you can come find us and we can try to really translate what you're asking for to the people in those offices unfortunately we don't always have like the perfect answer or the perfect solution but we're also doing is trying to find out well what's what is the need what are the demands so that we can make kind of a better kind of case to campus leadership that there should be investment in certain kinds of technologies so consulting again I mentioned this is really the heart and soul it's really what helps us be successful we really try to be hands on with people and understand not just the specific thing that they're asking work and I store my data but more kind of even the broader kind of workflow of your research so we can find out other opportunities to help you as well as what the real news are on campus yeah so that really is how we think of ourselves as consultants and facilitators of research so one of our biggest programs in research IT is called Berkeley Research Computing and you can see this kind of our logo here for Berkeley Research Computing which points to kind of the four different aspects of the program or consulting at the center and consulting here in addition to being something we do in research IT within Berkeley Research Computing there's just some special things that they do there fact that they have an NSF funded cyber infrastructure engineer who can really help even with some of kind of nuts and bolts engineering and a little bit of coding you know software help to really with a real focus on kind of data movement and how can we enable data to do more things um we've been hiring a number of domain consultants Erin do you know how many domain consultants we now have in research IT through the program yeah so we have three grad students or four grad students who are working in 25% time with right now we have two open positions that we're hiring for now you're interested or you know folks who'd be interested please go on yeah so this is so we can have people who not only know the technology but know a domain and that it's that bridging and translating thing that's been like the key to what I've done here another thing that Berkeley Research Computing does and I'll ask Erin to talk about this is we run a high-performance computing cluster called SOVIO and so Erin do you want to just say a few things about that please yeah so SOVIO is a high-point computing system which means they take a bunch of computers and put them together with a fast network um you can log into these nodes and submit what are called the FATS job so if you have sort of a long-running job these computers can handle that where your laptop you know is that the follower of the computation is too heavy or you can't do your laptop for other things um if you guys want to close it so we have a system that you can also buy into sort of any dedicated resources and this is kind of our large-stale computing system yeah and one of the things that we've been really good at is finding creative uses of this right it's like a super computer so it's like astronomy and physics and and molecular things but we've been really creative at finding ways that when people are trying to they're really you all you all are pushing the borders of what computing and data have done in this field um analytics environments on demands or or aion is basically a virtual research environment that we've been developing within research IT where we're taking advantage of some existing hardware that exists in the campus data center and building virtual desktops on top of that and we'll actually come back to that and talk about some work we've been doing with Niko to build out something that would really help archaeologists so that the notion here is is that right now to to get um arc gis and your favorite statistical package and everything else running on your laptop you're basically becoming your own system administrator on your computer right well what if instead you could log into a virtual machine environment you just go to a website and log in and you have it there that's kind of the notion there with the virtual research desktops and we'll say more about that in a little bit okay so i i mentioned that we're doing some interesting things a lot of these research IT groups around the united states and even internationally are really focused on those high end scientific uses right but because our focus has always been intentionally broad and we we've for a long time worked with archaeologists with museums with digital humanities we we've looked at from the very early estate at some other applications of these technologies what could we do with a supercomputer for a humanities project for an archaeology project and just a couple examples here we've done some really great things with ocr some of that have been spent focused just on licensing right abbey is great software but it's expensive so we invested a little bit of money in some abbey licenses and it made those available like i think the where where can people access those is it at the D lab or do they send an email to research IT to get access to the to the abbey service yeah so in the D lab and Barrow's home on the third floor there's physical computers there where abbey is licensed there's four licenses there but we also have in this aeol environment some place where you don't have to go to people that just from your laptop you can connect remotely for the shared calendar schedule to resource and we can talk a little bit about that more when we release them to aeol at this point well CR is probably on optical character recognition so this is if you have text and you want up to the if you have images of text and you want to be turned into text that you could do some machine reading over or or or data mining yes it's like some ancient script yeah right yeah yeah it's not it's not after that yeah absolutely now there's another ocr software called tesseract i think which is open source that we've actually made run on our saviol high performance computing cluster and we have some scholars who have massive amounts of text that they want to ocr and they don't need to find you know abbey is really incredibly good at ocr but with some depending on your research question you might be able to run something over our high performance computing cluster and then you're running against this massive documents so that's one of the other things we've done photogrammetry so the process of creating 3d models based on photographs that also has a licensing issue and we've been working we have a license for photoscan that we're using in a project that i'll tell you about in a little bit and we've also pushed the the photogrammetry processing off of laptops and into our more higher powered computing environments and Rita Luccarelli professor in Near Eastern Studies and her graduate student Gea Johnston have done some of that in partnership with the Boston Research IT and this has taken some of those photogrammetry processes can take days if you have large objects and lots of photographs it could literally take days to do one step and you've been able to push those two high performance computing cluster where it runs in like three hours instead so that has allowed them to be much more kind of iterative about the processing of this all right so that was Berkeley research computing one of the other things that we do that i'm more involved in myself is research data management so this is a partnership with the library to look at the broad set of challenges that researchers have with their data whether it's the data management plan or working with your data during the active phase of your research trying to move it from one place to the other trying to understand whether you can store it in one of the campus services and then what you do with your data at the end of your project so that's research data management and this was really because you know funders are requiring that you have a data management plan with your grant proposal that you're thinking about your data sharing plan how you're making your data available for the future and you know frankly we were just getting a lot of researchers asking us where where do i store my data i have more data than i've ever had before where can i put this stuff and and still work with it in real time so these are the kind of questions that we're helping people with and we're using a very broad definition of data so it's not just about large scientific numeric or tabular data it's about the digital inputs and outputs of your research so we work with people you know creating images of old books and things like that that's that's the data of your research right the great thing about archaeology is that we have so many different kinds of data so archaeology is so great at challenging you know what we're trying to do and you all know that because you don't work with just one kind of thing my definition we're looking at a whole bunch of different things so research data management i will mention so one of the things we do is we do consulting so you can send email to researchdata at berkeley.edu and we'll get back in touch with you and try to put you in touch with the right resources or answer the question ourselves we do we have a website researchdata.berkeley.edu which has some guidance some information we're always trying to improve that and add stories about how people have worked with data and just kind of making a case about this outreach and raising awareness is always a big effort how do we make clear really the campus leadership that this is researchers out there are struggling you know how many times i've heard from a faculty member that they've had to build a whole kind of storage environment in their lab and and they're taking the backups home in their back pocket taking some DVDs or something you know you know people who are becoming kind of technology system administrators because they don't have any other option so part of this is making the case that we need to take some of that burden off of the research community so that you all can really do the research part of that is looking at some specific areas so active research data storage at that phase like during the research how do you really store your data where can you store it how do you move it from the field to data storage environment how do you get it ready for analysis and later publication and then a lot of issues around research data that you have different kinds of security requirements where they can't just be put in any open place where the public can see them how many people have this issue where the data they have has some kind of restrictions you could just put it on a public website for instance i suspect many if not most people right some of it is just all these are valuable data they're in process i i'm not ready to publish these so i'm even concerned about that site location data site location data is a really big absolutely in fact i get to go to the saa this spring for the first time in like 25 years or something michael black from the first museum of anthropology and i will be talking about some work we've been doing with the collection management system they use and one of their big efforts spent around setting some rules for what information about field collection location can be published and to what audiences and similarly what what objects cannot be made available visible on the website at all really important questions and when you do something in an information system you actually need to make those decisions formally and then somehow code the system to to follow your desires data management plans we do a fair bit of work here actually not as much as as we might expect i think most how many people here have actually written a data management plan just curious so few hands okay so it's this kind of thing where five years ago nobody had written one they weren't sure what to do with these things they were just asking their friends and now many people have used one and have an example of copy we're happy to help with this kind of thing if you have another project that we recently got involved with in research it is around kind of the broad set of issues related to visualization right so within research it we knew that we can't just focus on scientific computing research data management research is changing so we have to be paying attention to that landscape getting out there talking to the graduate students the undergraduate students who are starting to do new things with new technologies talking to researchers in computer science about services that they're developing that might have an application in another area visualization is one of those things that we certainly wanted to track this year we had an opportunity to work with Hearst Museum of Anthropology on a project to do some photogrammetry so the Hearst Museum is in UC Berkeley are part of a four UC project called the UC catalyst at risk archaeological at risk cultural heritage and the digital humanities that's a word actually from the UC office of the president to four campuses UC Berkeley UC San Diego UCLA and Bersed to install a set of visualization environments this is basically a set of 3D TVs hooked up to a high end graphics basically a high end computer a workstation and then this is showing 3D views at risk cultural heritage sites and can actually how many people have actually seen this over at the Hearst great a bunch of people okay so some of what you see there okay so there's Luxoria in each other that's great but they have some things over there that are not really at risk cultural heritage sites there's a stone bear at UC San Diego and I don't think that's a cultural preservation kind of play there where they're really concerned about that thing being torn down but but the notion is that that you could use these shared platforms actually a network of these visualization environments to both record and share out information about cultural heritage sites archaeological sites they're at risk due to whether whether it's terrorism or erosion this is a way another way for a museum to help kind of store record and share and curate this kind of information so I got involved with this because I could see that there was a lot of work to do I thought it was just really interesting and opportunity for research IT to get actually some hands-on experience with visualization visualization technologies photogrammetry etc so Ben Porter the director in the Hearst has been such a great partner on this he's been so helpful actually getting hands on helping to support a team of students that we hired first in the spring but they really got started over the summer to go into the Hearst Museum collection and start doing some photogrammetry and I'll just go on to the next slide as again some of our partners here there's a little little movie of the first wave over there in action our goals are to get some of the museum content excuse me from the first action display here too so that you could have an archaeological object on the site or a face or something like that actually displaying in this and you could zoom way into it or look at the bottom of it which you can't usually see it's in an exhibition it's sitting on something and you won't see all angles of this object so you know we've actually created a bunch of models and now we need to figure out how to get these built in to the application that runs on this on what we call the Hearst K so you can see here we have some other partners here on campus through the Citrus and the Pacific Research Platform that's another NSF funded project that's looking at what could we do with high speed networks between the different campuses to really enable large data and data intensive research to happen more in real time how many people currently are using like thumb drives or portable hard drives to move their data from either the field to their computer or from a scanner to a computer everybody is using a thumb drive right well what if the network were just so fast that you didn't have to do that so that that's kind of the goal of the Pacific Research Platform is campuses and the states have actually in the federal government have invested a lot a lot of money in some of that underlying network technology but that doesn't mean that that the high speed network you know comes to Ego's office it doesn't mean that the software that you are using on your computer knows how to take advantage of that high speed network so there's still a lot of work due to to unlock that potential the reason the other reason I read you like this is this project is just right at the intersection of research teaching and learning and also public service right because that's that's those are the roles of the museums and I work a lot with museums on campus I love them and this is a I mentioned that we've done this reorganization where research IT has joined with educational technology center for teaching and learning this to me is is one of those perfect intersects I'll mention just a few things about kind of our approach to this we've we've done this current project here really on a shoe string got a little bit of investment from research IT we've got to have some funds from this UC catalyst project that actually built the visualization environment we applied for funds from the student technology fund students pay a technology fee and most of that goes to things like Adobe and Microsoft and some of those some of the big expensive things but they also have an open competition for innovative technology projects and so we applied for a small amount of funds to hire some undergraduate students to help really get into into the Hearst Museum collections do more photogrammetry and then build out the applications to display those things in the Hearst Museum and so we've just done a small small amount of funding and funding have really helped us there the fact that we have museum at this museum but this incredible content is just such a powerful thing when I talked to my colleagues at UC San Diego or even at ever said that they're working with some really good archaeologists too and I say well we've got 3.8 million objects to work from we've got a museum director who's actually doing the object handling himself and staff there are really into we've turned our work into a museum exhibit over the summer we were doing photogrammetry pridings right so that we would actually be there in the museum working at the Hearst Cave itself with the students and visitors would be coming by and maybe telling them what we were doing and it was it was just great it's such a great summer I mean this was like a job on top of my other jobs but just this fantastic so and also working with students thank you great it's been a lot of fun practice and learn from mistakes wow we've made some really good models we've made a lot of mistakes how many people have actually done some photogrammetry I see some familiar faces okay yeah so you probably know it's not it's it's amazing but it's not like a hundred percent perfect every time there are plenty of opportunities to make mistakes and so we learn from those mistakes and we move on we consult with experts we've had Michael Ashley from codify we had an undergraduate what's the what's the guy's name he came through and he's been doing a lot of this work as well and so we're we're learning a lot from the expert but there's still a lot of learning and again I mentioned we're also trying to push some of the processing the faster technology so you can see this like how much of this is a technology project there's technology in it but there's so much so much else really than enable technology be powerful and useful oh okay so museum informatics and collection space this is another program that we work on in research IT this has been a Andrew Mellon foundation funded project to develop an open source software package that museums can use to track information about their objects as well as all the transactions what's on loan what's on exhibit where did it come from and this is an open source community source project to develop something that could be used in art museums in anthropology museums in science museums in kind of popular material culture museums and it's in use in use kind of around the country and in a few places globally as well so this is not visible in any way but a museum collection management system is a really complex piece of software you think of the metadata all the object all the data that you could use to describe an object that's already a pretty complicated model and museum people are anybody from the hearst here anybody is he a former director in the back there okay they're crazy about they're crazy in general i love these are crazy people the the level of kind of data that they want to report about every object is fascinating and so we had to come up with this extremely rich data model you're seeing the first two percent of the cataloging screen so it scrolls there to about the back of the room but then along the side there are all these transactions like a conservation what conservation is this object that is subjected to exhibition information as it is on loan there's all of those again recording everything about those transactions so this has been a fascinating kind of software and an information ecosystem project one of the things that we'll be presenting at the SAA is about when we we talked about this collection space system is that once the museums really became very comfortable with their information and the quality of their data then they said now now we have to unleash the strategic power of this information so how can we use this to address our most strategic issues things like um you know sharing information with the public but not not all information protecting some information how can we use this to really improve the quality of our data how can we support researchers who want to find out what's in our collection how can we make it easier for our staff to respond to requests for for research business those types of things so on top of collection space we use the ABI then to build a set of portals and other applications that are being used a lot right now and I just put the the link here to the first link web app c space Berkeley BDU this actually has links not only to the First Museum of Anthropology but also the Berkeley Art Museum Pacific Film Archive the University of Jepsen Herbarria and the UC Botanical Garden which is a living collection so that's one of the other museums on campus that we work with and then this is the link to the First Museum search portal and it's just fascinating again 3.8 million objects it's just an amazing collection right so actually so now we're going to actually talk a little bit more about the um what we've done with the archaeological research facility in this analytics environments on-demand service again this is the the virtual research environment where we're trying to build this out so that it can be useful so you don't have to get everything installed and licensed and paid for something yourself this could be something that that I think could be really really powerful Nico has been you know the person really helping drive what that should look like um so Nico do you want to kind of talk about I think you can talk about the experiments um so the idea here is that we can provide access to a windows desktop and a browser accessible on a nine-year-old Mac for example so should we do a live demo on your phone it's just a I want you guys to see how easy this is it's great here if I can find my mouse point all right so browser window citrix.broteam.in here it's in there you sign up with your calmet ID and that's my that's just all my calmet information active directory so it's syncing with the campus wide windows or directory system and then I have access to a number of these abby fine reader or something we we can provide you access to this way there's an issue with license with our you know with the number of licenses we have and scheduling so we can't like open the you know open the gate all the time and I have to coordinate with others but here's one PIS research so the idea here is it'll be available to you it'll store your data not we don't have a ton of space available at this point but if you really find it useful and your project invests in it there's a condo storage option that's part of the funding model for this program is that projects that become really involved in this can pay a little and the window opened on my other screen here they can pay something by the by the megabyte or the gigabyte or the terabyte yeah or I can just turn my night can turn my screen out it doesn't seem to okay so anyway you know I'll just show you on here so this this just became the windows desktop so they've got the start menu and everything and launch ArcGIS so it's quite easy and so the whole point is that this is running in Warren Hall I believe and you know I could start a job and I could like get it going on something GIS project and then close my laptop go to class you know open something else reboot the laptop and it's still running unless you actually shut down the windows machine. The other application here the back end slides so this is the research version that was the research version teaching is so you could you could also use it for teaching a class we have some specialized software it has to be windows or there's talk of a Linux possibly a Linux version of this also launching and so you can have uh uh this technical software available with identical experience for all the students in your workshop or your class and they would just log into this web page and now my mac is turning into a windows machine. I think that's there one one powerful aspect to this is that licensing for much of the software is on an institutional level so like Berkeley has access to for example I think four photo scan licenses and it's perfectly fine for us to share those licenses within Berkeley institutions like this but practically speaking who's machine does it go on well you can pool them and put them on here now four Berkeley people can sign in and simultaneously use photo scan so it's you know it jumps it solves that obstacle. So just a couple more because we want to make sure we have a few minutes of discussion. So one of the things that's that's driving some of this is that campus recently went through an IT strategic planning process called reimagining IT probably nobody here except for like these four people and including myself have even heard of it but this was you know kind of a nine month process to look at what should we be doing to be more to help the campus really respond to the situation we're in now I think that the main goal that's stated here or a single goal is that enable UC Berkeley to remain a great public university and to solve the campus financial crisis. Was that? So let's do it. Yeah let's do it. That's a great goal. So how do we get there? So so within that because that's actually the objective the first goal within that all the IT people are going to solve this problem. But the first goal is that all faculty and students have the tools they need to advance research teaching and student success. So that's so that's kind of an iterative process where you draw a drill down into more specific things. The first strategy supporting that goal is to support major campus teaching research and public service initiatives here reallocate IT funding so we're trying to make the case that the most the way the campus has evolved and the way that the budget evolved is that a lot of the IT spend all the finances go to administrative systems like your payroll and the financial system and your procurement all those things are important but do they need to be the best they are or should we actually have and that they could possibly be rather or should we actually have a little bit of more funding for some research teaching and learning services should you really make sure that those are improved or that more services and capabilities are made. If I was very involved in this process it was another job up top one of the jobs for about nine months about this plan and now we're actually saying like okay so this this is what we're saying we want to support this this goal and this strategy so so what do we do some of the things that I've told you about and having some licensed software having some more flexible compute environments having you know like we have more access to things like box and drive really finding out how finding out how to make those helpful to researchers teachers and learners how to make those the effective parts of what you use in your day-to-day work that that's kind of some of the next steps. For that is hearing from you about about what what you would like and also creative what you were doing that maybe other people could also take advantage of. We actually have a couple of other sets of information here about grants and things like that and I think that I know that Nico's going to have those available on our website so is it okay if we kind of skip over some of this? Yeah those are just links to grant programs and other labs on campus that you may be interested in pursuing and I'm going to link to this. I'm going to piece this into our website and send you guys the link but if you're not aware of D-Lab, BIDS and GIF then you definitely should know what it is. Oh yeah that's right in fact I wanted to mention so so partnerships with does everybody know about the D-Lab? Is that D-Lab? So it's in Barrow's Hall and also the Berkeley Institute for Data Science Bids which is in the library is also a great partner and a great environment to learn from other people. We have some really good and strong relationships with those groups so please feel free to drop us down and talk about those things. So I want to just at least spend seven minutes here asking you all some questions so maybe we could even get the lights up and have some uses aren't you agree about? So kind of being able to talk about the visualization and like environment, what are the things that you all are doing or that you would like to be able to do that you don't have access to equipment or something like that, open it up? Do you have questions or if you have ideas, I'd love to hear what's going on. Maybe I'm the only one in this situation but coming up this culture working at his background I am lacking some fairly basic computer skills and I find myself needing to construct and deal with databases. I don't know the first thing about databases by school, any of that. You guys have this sort of training program for stuff that we don't visit like advanced fancy stuff just like basic computer stuff for non-computer people. Absolutely I know you are not alone. So I think there are a number of good resources and I think we're going to add a slide with training resources. The DLAB, even though it's not the Social Sciences DLAB, it's a great place and they will work with anybody. So they have, have you checked out there? Actually I have a typical insurer programming I've forgotten about. I'm very sure that our are also forgotten about. So this is how I feel about it. Some of it is either not at the right level or at the right time. Exactly. So and that's where consulting can help. So for instance what I suggest is if you're at a point where you're actually going to start working on like how the data model would look or something like that, you can set up a research data at working.tv, you can schedule a consultation, you can sit down, kind of get a sense of where you are in your project right now and then you maybe do some direct help or say, well maybe it's, maybe look at the R class, maybe what you needed was this class or oh there's this resource on Lingedoc.com, which I think we all have access to now. Is that what they have access to, Lingedoc.com? Actually do you know who has access to Lingedoc.com? I think they were trying to add students. They were trying to add students. Lingedoc.com is considered by some sort of a premier online training outfit for largely for computer-based applications. They've been around for, it's a very competitive marketplace and some of it just really got on ahead. And what's different now is the university has, come up with a mind to start licensing it so that initially it was just for faculty and staff, but my understanding is that they either have this summer or are about to extend the license for the entire community. So faculty and staff can already get into that portal that you may have seen called Llew, Llew not perfect, of which is now the Learning Center, I think it's been rebranded, so that you should be able to find it there, and then I think there's going to be some announcements about students being in access. Yeah, so for example, you could probably take a 20-hour course and find a link with the, in its videos and exercises that you think I don't need about coming, or the final cut. Are you saying Linda, L-I-N-D-A? L-I-N-D-A. That's fantastic. I don't know why I didn't know about this, because I'm paying like $150 at the moment just to have a three months or five months license, so if I can get it through the university, I've never seen anything about that, so that's a well-guarded secret. Now it's going to be renewed. Yeah, it is a great resource, so thank you. It's going to be structured in small, manageable modules, so I mean you get on exactly the case. When you really want to build your data model, that's when you want to concentrate on that part, you don't need a whole course on my SQL, you just want to get in and be able to do, that tends to be the size of these. Yeah, they're really cute. I have a question about whether or not you also have a space for creating spaces. I don't have any one in the computer science department that's also a graduate student, and I would love to collaborate with another graduate student, because I know that they're hungry for data, and we have the data, and they have the hunger. Are you guys also thinking about creating collaborative spaces where you introduce graduates because they're from different backgrounds that can work with each other? So I think that's been a big question on this campus for a long time, and you may or may not have heard, there's a new division on campus, the data science division, and it's partly in spirit trying to address some of that question. It's been the Berkeley Institute for Data Science, which has been around on campus in the Doe Library on May 4 for several years, has been kind of the first experiment around that, that sort of predated the data science division. They just on-boarded their newest cohort of grad students and postdocs within bids, and I think that would be a great place to go to make those kinds of connections. They have formal talks and they also have informal events like teas, and they also have working groups, and I'm sorry, I don't know your name and questions earlier, or what are you planning. I think if you're trying to connect with people, it's a great physical space to go and do that. If you're trying to learn things, they're also doing formal and more informal trainings in the form of trying to build a community together. So for example, on Wednesdays, Wednesday afternoon is the plan of today. In fact, the Hacker Within is a community group that meets, that mixes people from lots of different domains together, from completely beginning levels to advanced levels in the same room around different topics. So you can go and propose a topic, or meet other people in the community that augment some of the official training stuff in ways that sometimes when you learn something, you don't know the kind of questions you might want to ask. Some things are un-gluble because it's like, oh, I don't even know what question to ask here. So that community is a great one to tap into. Is the book Leason 2 for Data Science? Yes, it's in the ground for the library across from what we learned from it. The Morrison? Yes, just following up on that, they also have an annual post-reception, it's a great place to chat with people, and remember, we have awesome content in our ideology. A lot of people are excited about what we have to talk about even if you're going on the technical skills. So just bringing that knowledge, if you're interested in these, is definitely an answer. The love of what you've been showing us is on where, you know, access on the web, but not all of it, and some of the meetings obviously are not, you know, there's an aspect of this which is physical on-campus. I'm just wondering, you know, how you see research IT, whether it's something that you really need to be on-campus to take advantage of, or can it be done like the library research now on the campus? Well, that's where the virtualized research environment is, and so that really, really can come in and become quite powerful. So I'd say it's, yes, it's all of it, and certainly we're seeing that the kind of community part of this is extremely important, and we've spent a lot of time investing in building a community with our other kind of service provider organisations, research IT, with educational technology, with chemistry services, with the library, because we realize we're all trying to help basically the same people, which are people like you. So, but for years, we all talked about the silos we've all created out of our individual silos, and we've been working very hard to break those down. But so the community part of this has been really important. And the last two years has been, I would say, a game changer in terms of how successful we have been building those kind of relationships. What I might add to that, thinking about the work that's done on our technology that you've done on Niko, is see how some of these tools can fit into your fuel work and be a tool when you're there. So I don't know if you might want to say what Susan was doing. Oh yeah, so Susan Paul, in front of a, from the Mt, an imagery library, was in Mongolia in some field project, and she was trying to use our GIS on this AI, the same local environment on demand, and she, so she was able to sign in from, I think she was going to go on a tour, but she had a pretty good web connection, but one thing we're working on is making sort of a slim down version that will run on a rather weak internet connection, so that you have access to this really powerful computer, and all the, you know, real licensing headaches, and you can set up a job, or you, the other thing you can do is transfer data really fast, because, you know, the Warren Hall is on a gigabit, you know, it's on a really fast web connection, see, things move back and forth from that AI machine to Google Drive and Vox really quickly, then you can do your work through that in the local environment. Maybe one last question. Great, thanks so much, everybody. Again, research.data.berkeley.edu, you can send email, rysnop.com, and at berkeley.edu, you can meet the egos, have a get in touch with us, and we'd love to talk. Thank you very much.