 Thank you. Good morning and welcome to this week's edition of Encompass Live. I am your host, Krista Porter, here at the Nebraska Library Commission. Encompass Live is the commission's weekly webinar series where we cover a variety of topics that may be of interest to libraries. We broadcast the show live every Wednesday morning at 10 a.m. Central Time. But if you're unable to join us on Wednesdays, that's fine. We do record the show every week and then post it to our website for everyone to watch. And I'll show you the end of today's show, how you can get to all of our archives and navigate them. We, both the live show and the archives are recorded archives are free and open to anyone to watch. So please do share everywhere and anywhere, friends, family, neighbors, colleagues, anyone you think they'd be interested in any of the topics we have on our show. But those of you not from Nebraska, the Nebraska Library Commission is the state agency for libraries in Nebraska, similar to your state library and other states. So we provide services to all types of libraries in Nebraska. So you will find things on our show for all types of libraries. Public K-12 academic universities, colleges, community colleges, corrections, archives, museums, it's all across the board. So really our only criteria is that it is something libraries are doing, something library related. We bring in guest speakers to talk about cool things you're doing in their libraries across the state or across the country. And sometimes we have a Nebraska Library Commission staff that do presentations about things we're doing here at the commission or other services and things we offer. Before we get into today's show, I just want to do a brief mention about on the Nebraska Library Commission homepage here about the resources that we are providing to our libraries. Since we are currently still in the height of the COVID-19 pandemic. So here at the Library Commission, we are trying to provide as much resources as we can to our libraries in the state. If you're in another state, most of these resources you can use as well, they're all out there. But check with your state library or your state library association, they may be doing the same thing as we are. We have a post here, it's pinned to the top of our homepage. We'll always be there on top of any of our new posts that come up about our pandemic resources. Then we do have a list. We're attempting to keep track of our Nebraska libraries, our public libraries, who's open, who's closed, who's making special accommodations, a Wi-Fi in the parking lot, curbside pickup. Now, of course, who's re-closed. We've got libraries now that are having to step back because new outbreaks happen. So we try to do our best to update that. There's a form on our page where libraries can go to report to us and let us know what their situation is. But our library commission staff do proactively reach out and see what's happening in the news, see what libraries are posting on their Facebook pages or their other social media or their own home pages about what they're doing. Our specific resources for libraries we have here, lots of different topics depending on what you might be doing. But I just want to highlight what about my library section. This is where we've collected anything we've found that has to do with the pandemic, with opening, with closing policies, meetings, special resources for school libraries, examples of what other libraries are doing, all sorts of different information. It's updated regularly. We're always adding new resources to it as we hear about new things happening, new reports coming out, new updates happening. So keep an eye on this if you aren't a Brass school library to see what we're sharing out there. Something we do like to highlight also is the realm project. This is a scientific study specifically being done on library materials. Things specifically circulated at libraries and museums. So they're regularly doing tests in a laboratory of how COVID-19 lasts or does not last on certain materials. So definitely keep an eye on that to see what's going on there. But all the other resources on here are very good too. Like I said, if you're not from Nebraska, you're welcome to look at our page, but check out your own state library and your state library association and see what they're doing. So on to today's show. Today it is the last Wednesday of the month, which means Amanda is here. Hi, Amanda. I am going to actually hand over presenter control to you right now so you can get your slides up. Amanda is our technology innovation librarian here at the Nebraska library commission and every the last Wednesday of every month. Pretty much. We have scheduled her to come on to do her pretty sweet tech session. We are a tech related person. This is definitely the session for you to always keep an eye on and sign up for every month. Even if you're not taking some good things to look at here. And today she is going to talk about teaching kids machine learning, which sounds fun and terrifying. But it's using scratch, which I've used before and I've seen people do before I think it's a lot of fun. So I'm just going to hand over to you Amanda to tell us all about that. Can you see my machine learning? I do. Yep. Yep. Your slides are up there. Full screen looks perfect. So I decided to do machine to try to tackle machine learning today because it's one of the fastest growing fields out there. And it's going to braid in and involve a lot of the growing stem fields because the jobs that are in stem. And the jobs that are growing are actually in computer science, data analytics and starting to be machine learning. So it's something that is really important and we're just going to start looking at it. So we're going to go over just a quick brief level overview. I'll pull up some videos that'll show you demonstrations of what machine learning actually is. I'm not going to use Facebook or Google or market segmentation because those examples have been done to death pretty much everywhere. And then I'll do a quick demonstration of what machine learning for kids looks like. And then we'll talk about automation a little bit. I put together a quick lesson plan that you can use in your own library to show people what automation is using the example of travel plans, booking a flight and figuring out where you want to go and what to pack when you go there. And then I'll give you a little slide of tips that you can use if you want to start implementing this yourself or if you want to start digging deeper into what machine learning is going to mean in your community or just to everyone. All right, so this I picked this slide because it is intentionally disorientating when you first look at it. This is a slide that this is an image that they use in a website called towards data science. This is for developers. And I chose this because machine learning sounds disorientating, but it's not. So you see at the heart of this we have machine learning machine learning is just machines learning by example. They recognize patterns and are able to take action based on the patterns they recognize. Now this unsupervised learning supervised learning and reinforcement learning are three different types of machine learning. The one that we're going to dig into today is supervised learning specifically image classification. So image classification, and I'll just show you a quick video of this here. So what that video was showing is that there was a farm that wanted to classify images, classify cucumbers based on length, whether it was curved and whether it had scratches on the outside. And they use machine learning to they loaded the cucumber into the machine, the machine learning algorithm looked at it said, yep, that's a cucumber. And then it put it into the rest of the machine. It did a second little analysis that said, well, this is curved. So I'm going to shoot it down the conveyor belt and we'll toss it into this bin. And before that farm was using machine learning, it took his parents just forever to manually sort through those cucumbers and they were losing money, they were losing time. And they use this image classification to sort cucumbers. It's like, that was so interesting. Watch that video. I was like, wow. Right. But it's interesting to see how in the end it didn't work as successfully overall as would have been hoped, but we're going to go back. Oh, sorry. And that's what they're finding out about machine learning is we all think Terminator. But it's actually just really narrow tasks that AI is tackling. So Terminator state level yet. Yeah. So to get an understanding of how machine learning is going to work in the real world. You basically have to really understand the full process of what's going on. And then you would find out which mini task machine learning is best suited to tackle. So for example, this is reinforcement learning. And I'm going to play this and then I'll explain it. So in that video, we were worked, we were watching a robot learn. And the fun thing is that robots can now learn similarly to the way people learn. So you saw in the beginning there, the guy actually grabbed hold of the robot arm and did the movement himself. So whenever you saw the caption down there that said the number of trials that we're going by between each one of those trials. The guy would go back over, show it how to flip pancake again. And then the robot would learn better and better and better until you saw at the end it was flipping it perfectly. And you also saw you might have caught in the background. There was the little screen that popped up there that was mirroring the robots movements in data form. So that was tracking all the different sensors that are inside this robot. So there are little gyro sensors and different sensors that are measuring the acceleration and the angle of this arm. And what the robots actually doing is starting to recognize patterns better and better and better each time. It said, well, I just I tried to replicate this movement. We all saw it and go to well. But then it started learning better, but it was actually learning the angles, the acceleration and the movement that was gathered by those sensors. So then once this one robot got really good at doing that, now factories and in manufacturing, they can start connecting the brain of this robot to other robots. And now those other robots can learn more quickly. So this one took 50 tries to work to trials to learn it right. But now the rest of his robot friends can do it in probably about two or three rounds and even quicker as they get better at it. So this is why we're starting to look more machine learning because it's kind of a thing now. And it's starting to take over more tasks more effectively. And that's why if you ever look up the future of work, AI is going to be on the list. So I'm going to close out of this. And now I'm going to go over to how we can learn this ourselves and how we can teach kids about it. So before it used to be really hard to learn, it still is hard to learn deep learning from scratch. And I'm going to clarify something because from scratch means like from the bottom up, but in this case we're using scratch the programming language to learn it. So maybe change your phrasing when you try to describe this to people in the library. It's both. So machine learning for kids was made by Dale Lane, he's an IBM developer out of the UK. And he made it super easy to be able to use it. So I'm just going to close out of my slide to minimize my slides here we're going to open machine learning for kids. And we're going to teach your computer how to recognize images. And in this case, so machine learning for kids put together a whole ton of different machine learning projects that have step by step instructions. The one we're going to use is car or cup, which we're going to train a computer to be able to tell the difference between a car or a cup. For us, it seems obvious and easy, but for a machine to learn that same pattern and be able to quickly and easily recognize and differentiate between the between the two. It's a little different. So if you're using this in your own library start by going to worksheets, find a car or cup project and download the guide for teachers and the guide for students. So grab this guide for students here. And this is what your what kids and possibly adults who are interested in it would be looking at when they try to do this project. And it will give like a little screenshot of what the whole system is going to look like. And then it'll just give them step by step directions for what exactly to click on. They would open up this website. Click on get started login and go to their projects. So this projects is something that we as the teacher or the admin in the site would set up for them. And then we would let what they would need to be provided access to this before they can get into it. Which is something that I've already done ahead of time because it took a minute, but there are also step by step instructions for how to do that. So when I click on the projects and the machine learning for kids tab, we'll go to projects. And you'll see that I've already pre trained a model called car or cup. And I also played around with an emotion interpreter, which is trying to teach a machine how to recognize facial expressions and spoiler. Didn't work so well, but cars or cups. Computers got down. So I will show you how I did this. I'm going to add a new project. And this is exactly what you do the same steps you would do in your own library. I will add a new project. It's going to be a. So this whole class project is if you have either, if you have a group of kids that are going to be doing the project, or if you only have one student or two students that are sitting in front of one computer. We're going to assume that there's only one or two kids that are going to be learning this at a time. So I'm not going to click whole class project. And I'm going to title it car or cup. Trial. And then we'll go to this drop down to tell it what it's recognizing. And it is recognizing images. But it does have the capability to do other stuff. So we'll go to create. Now it's going to create a new project here. And now we have to train it. Click open this project. And it's going to walk you through the different steps that you have to do to be able to train your computer to tell the difference between car or cup. So the first we go left to right, we're going to train the module, the model will click on train. And now I'm going to minimize this screen. We'll put this on the right hand side here. And now I'm going to open up a Google just regular Google image search. And I'm going to open this on the left hand side. So I'm going to search for car. Go to images. And now on our right hand side, we need to tell our machine learning project, which buckets we're working with. So we're going to go to add new label car. Add. Add new label add. So we're telling we're labeling our information to tell the machine learning model what to look for. And then we go in here and we're just going to drag and drop different examples. Okay, four, six, seven, eight, nine. And if you already, if you accidentally drag and drop one that's already in there, it'll just pop up a little error, but just grab a different picture and keep going. And we'll grab this one. And you want about an equal number in each side. So we've got 12345678910. So we'll grab 10 cups now. We're going to Google search for cup. And now we're going to start pulling in our cups. Two, and three. And I'm pulling in different kinds of cups and different kinds of cars. So that model will be able to better recognize different types and different settings. And so I'll grab this one, four, five, six, seven, eight, and missed one. So we'll grab a different one. We will grab this one, 10. All right, so now once we've dragged all over sample images into our machine learning model, we're going to go back to project. So we want the, so now this is going to run in the background here. It's going on in the server all the way over probably in the UK. And it's going to send us back the results. So we're going to go to learn and test. And I must have grabbed one extra example of a cup, but we'll live. And so I've got 10 examples of a car, 11 examples of a cup. And now we're going to train our new machine learning model. And this can take anywhere from a few minutes to get going or to. I've had it take up to about 20 minutes. It really just depends on how slow the server is running and how many other people are trying to access it. So recently, I've noticed that there have been more, it's been going slower. So there are probably more people accessing it since they just put out a news release recently. So it's just more popular now. So we're waiting for this model last check. We're waiting for this screen to shift over to let us know that it's completed training. And there's also a quick and handy machine learning quiz down at the bottom. So this is kind of a fun thing to do while you are waiting for the machine to quit to finish training itself. But, and in the chat, let me know if you want me to go through this quiz right now if you're curious about it. You can just give me a yes or no. Yeah, definitely. If you do have any questions or anything you want to see that closer have repeated type in the question section here. Everybody understands what we're doing today. And right now I just want to know if you want me to go through these quiz questions so that you get a better feel for it or if you just want to look at it later. We'll carry on for now. Yep. I'm going to minimize this. And I will, while we're waiting for this to finish loading to finish training itself here. I'm going to go on to the next step in training our project because I already have one of these models pre trained. Cool. Yeah. Which I figured I do since I know it takes a minute. Yeah. So this is the one that I've already done. And we have trained it. We've gone to learn and test. And this one I did with way more examples because since I did it ahead of time I was able to take longer to do it. And it will just ask you if you want to test it out by putting in like a link to an image but I've already done that I know it works. And we're going to go back to project and go to the make section. So this is where we happen to sketch or into scratch. So scratch three is the most recent version. We'll click open scratch three. I'm going to make this bigger so that you can see it. And we'll go to open in scratch three. So if you've been in scratch before, you'll see that this car or cup is new. This is a library that the machine learning for kids has pushed into the project. And we are going to go up to project templates. And we're going to grab car or cup, which is down here. So what this does is it will bring in 23 different images of cars or cups. So if we were to play this right now using this original code will go to the green flag and all it's doing is pulling up all the different images that are that we're going to sort in a second. And so now we're going to open up our if you remember that student guide that I had brought up. There's a little code snippet in those directions that will tell you copy this exact code and replicate it in your system to be able to sort this. And I've also already done that ahead of time because all I'm doing is making sure that this looks like what's in the picture on the screen. And so I'm making it look like we've already trained it. So I'm just making it look like this, and I'm going to open load from your computer. Download cup solution. So this is what it'll look like. And we'll play it. And you can see it's sorting and it has gotten them almost there. So that is basically high assort stuff. And I'll walk you through what this code actually means. So on the left hand side here, it starts out by hiding anything and everything on the screen. And then this set why axes is telling the computer where to place the image. Once. So I'll go here. So this just told the computer to put the image right here before it slides it. So this is telling it to that image that was just placed. It's going to move up the number. So that we have 23 images in our little set. And this is telling the computer, go to the next image, pop it up on the screen. And now we're going to check it. Okay. If our image is a car, slide it over to the left. Otherwise, slide it to the right. So this is basically, if you've heard about an if then statement, or an if else statement, this is it. All this whole code is doing is saying, I'm going to pop an image on the screen right here. And there it goes. If it's a car, glide it on over here. Otherwise, glide it over here. So if we had tossed in images of faces or anything else that wasn't a car, it would automatically classify it as a cup. And that is kind of one of the problems that they run with machine learning models now is if they program it like this. You can get some problems. But if you're only sorting between two different things. It's way easier. But of course, I mean, machine learning as it goes into, as it gets better and better. They also get better and better at the way that works. So this is an option if you actually want to be able to run people through how to do the full thing, but a lot of people aren't comfortable with that. So there are other alternatives. So I'm going to go back into machine learning for kids will go to about. And I'm going to scroll down to further learning. So these are other alternatives where you don't necessarily have to code anything yourself. And there are other options where you can just put students in front of the actual tutorial without you having to do anything. So and those are probably more popular options. And there's also places like apps for good and Google experiments and AI in a box and clips over Cosmo that has additional support that you can send people to an AI for all is another one that you can just send kids to and set them up with a teacher that will show them how to do it. If you happen to be with a high school, you can go to AI for all. And you would be able to get connected with trained instructors that would be able to help you set up a summer program with curriculum, and they will provide all the support you need to get high schoolers up and running. So if you don't want to mess around with the code and you don't want to mess around with anything that we just walk through with scratch. And you happen to be from a high school. This is a good way to go. And Calypso for Cosmo is kind of a fun robot thing. And it uses it introduces kids to the basics of machine learning. And if you've ever used like Q robots, it has like that really fun curriculum that you just walk through. And it's really fun and easy to do. And it actually got cheaper. That's cool. Must be on sale. And these are additional videos to that are geared toward kids if you want to just pump them down and they want to learn more about what machine learning is. So if you start talking about coding a website and they say, well, I just learned a whole ton about how Google works and I really want to know how I can do that myself or how I can learn more about it. Is this an option for me. And then you can send them over to these videos and they'll just be able to watch it through. And then you can connect them in with other people who can teach them how to do it. Right. So let's scroll up here. And we'll go back into our projects here. I'm sorry. So we're in our worksheets that just give the other options of what you can do with machine learning. And so in machine learning is used a lot in the internet of things and then smart devices. If you've ever heard of a smart house or a smart building or a smart room, it's basically being able to control your device, your some kind of device with an app on your phone. So if you want your lights to turn on at a certain time, you'd be able to go into an app, say turn all the lights on in the kitchen and the living room on at seven o'clock at night. And the computer would just do it. And so the way machine learning works is that instead of you programming your lights to go on at a specific time, you would just live the way that you normally do and you would start turning on and off your lights, just as you would on a regular schedule. And then the machine learning model would start to learn what your schedule is. And then it would start automatically doing it. And as you can see, as you probably saw on like that reinforcement video from where the robot is learning how to move, that can go a little haywire in the beginning. But then once it gets that 50th trial down or however long it takes, it starts to get really good at it. So when you have smart buildings like manufacturing, that's kind of an effective way to do it because you're running on a schedule, and you're going to be turning the lights on and off at the exact same time every time. And you can override it if the lights go off when you didn't want them to. Am I explaining that relatively well. I read a lot about this. I think it makes sense to me but like as you said, I have heard of smart homes and things like that. Yeah. Yeah. And I feel like most people probably have by now, because it's everywhere. It's pretty, it's relatively common. Either people having them or being in like TV shows and movies now too. Yeah. And so when I brought this out into the wild and I've talked to actual kids who are curious about it and I talked to 18 to 20 year olds who were just starting school and they were interested in entrepreneurship and learning how to code and learning about stem. A lot of the feedback that I got what from these students was I'm really interested in this and I know that it's really popular right now. But I don't have anyone to teach me. And I never even knew this was a thing, but now I like it. And like kids were learning about it briefly, but then they lose interest in it because there was no support to keep going with it. And that's kind of what you can start doing now is just connect them with options like these or options like the additional ways to learn or even videos and sending them over to online learning communities so that they can just dig in and start learning more about it. Because these are kind of the jobs of the future now and automation is a thing. So I'm going to go back in here and I will pull up the lesson plan that shows what automation is. And I'm going to open up my Google Doc here. And I will zoom in so that you can see this better. So this is this can be done with adults that can be done with high school students or even six to eighth grade. And it's worked across the board just needs to be maybe shifted a little bit and sixth to eighth grade might need a little bit more help with figuring out the full different steps. But we're going to use the example of building travel plans. So if you think about when was the last time you decided to go somewhere and you had to either take a plane or train or a bus to get there. And what was it exactly that you had to do to make travel happen. And then we're going to run through those steps and we're going to see where automation plays in to see how technology and automation is going to be changing the way we do just little things and how that kind of adds up over time without us even realizing it. So the warm up here is just asking people to brainstorm the last trip they took. Where was the last place he went. And as we go through, I will just type this in. For me, it was Seattle. So to get into the. So when I decided to go to Seattle, I first had to have to decide who was traveling with me and where to buy the ticket. So I went to the plane. How long to stay. What to pack. When and how to get to the airport. And then I went through and had to check in for the flight. And I did the pre check online. And then I wanted to find out well what in the world is, and this is something that I did before. What do we want to do. How to get places. And find where we are. In Seattle hotels where are you staying. Yeah, hotels will put that up by hotels. Restaurants is always a big one. So what do we want to do. Where to eat. Okay, so now how is this automated now. There is scheduling software. I used doodle. Oh yes, doodle coordinate multiple people free easy. I love it. I do not know what I would do without doodle now. When we figured out a day, I use Calum Lee. To coordinate schedules and then we talked about it over zoom. And Calum Lee will let you send out a link to people. Then you'll be able to check off the times that you're available. And then it will automatically send over a zoom invitation that will get plunked onto your zoom calendar. And it will also get plunked onto your friends calendar. And so that was that's like a mini automation so that and it's a task that we used to do manually as humans. We have to we used to have to call or even email someone to figure it to manually figure out availability. And then we would have to pull up zoom. And Matt like make an appointment and then send over the link invitation, but Calum Lee automated it. And it's delightful. That's nice that's one I've not heard of before I'm going to have to, you know, keep track of that. Yeah, I just heard about it in a that entrepreneurship group and then I tried it out and I was like, yes, for the win. And so then I went to Google plates. And I searched across the different price comparisons to find out who was cheapest and the best time to travel. And then I also went to Expedia and the price line. I'm going to be doing it from Southwest. That's exactly what I do I use all those resources that do all the comparisons, and they all encourage you and now by through us and I go no no no no I'm going directly to the airline, because I just feel more comfortable going direct. And sometimes when I pop over to an airline depending on what they're trying to push. I've gotten cheaper deals. Yeah. I don't want to go to the airline Expedia don't like to hear that but And that's what that's the same exact reason I do it because it just from experience I've learned. Yeah. So the how long to stay I go through these different searches and there's often a calendar on there that will say when you leave on this time and fly back on this time it's going to be this certain price. I've actually adjusted the duration of my stay if I can save 200 or 300 bucks on airfare. And so that is by the automated price comparison. And I've also used the scheduling software for availability. And then what do we want to do. I've used. I mean there's a million websites that you can look up things to do. But the automation part of it was coordinating it. So we used a Google Doc to share travel ideas. And then we used a polling software to vote and then added it share calendar. And so what do we want to do and where to eat was basically the same thing. So I'm just going to get rid of that one. And one thing that I. And hotels. You can do it the same exact way that you searched for a plane ticket. There's those aggregate searchers that you can go to and then you just go to the actual hotel website to see if they have a better deal for you. Yeah. Price compare. Now the budgeting. There's an app for that. Of course. And so there are more automated travel bots right now. And so Amazon Alexa and Google Home. They both have travel software. So right now there's a travel. Place called KLM. That is using a travel bot that you can say hey KLM. What should I budget for this trip. And it'll pull up like the it'll automatically search a database that will pray that will do a price comparison for the average hotel costs. And it'll pull up like average flight costs and like recent data from the area and say if you want to have a leisurely vacation do the budget this if you want to have a luxury vacation budget this. And it'll give recommendations for where to stay and where to eat. And there's also a chat bot out there now that will. If so you can say hey KLM. What should I pack. And it will search the weather in the area and reminds you to pack an umbrella if it's supposed to have a high percentage of raining. And it'll look it can also there's an app that will be able to check a travel itinerary and say it looks like you're hiking. Remember to pack your boots. And that is the where the automation comes in. Is instead of you having to manually search through your entire activity list and itinerary and make sure that you have the coordinated items that will help you do. All those activities. The software automated automates it for you. It searches through seas hike and knows that it has enough data collected to know that if you're going hiking. You need your boots. And it has enough data to know if it's going to be raining. You need your umbrella. So what is what to pack. Chat bot. And KLM is a popular one. So now when and how do you get to the airport. I use Google maps. Well, but now Google maps is automated. I have the Google Pixel phone. So when I added the flight into my calendar. My phone will automatically prompt me to a Google Pixel phone. And then it will pull up instructions from my house to the airport destination. So I don't have to pull it up when I go. And so not every phone does that. But it's getting more popular. Automated drive instructions. And then it will pull up instructions from my house to the airport destination. So I don't have to pull it up when I go. And flight reminder. And it will, some of them will also do a check-in reminder. And so the check-in reminder, you can do it. You can check in online now. I do that all the time. Yeah. So much that they, when you get to the airport. Right. So when I, when we put in that shared. Itinerary, you can also put in a link to the direct restaurant. If you click on that link, your phone will now let you tap a button that says directions. So instead of manually having to go through and search for it in your, in your directions, you tap a link, hit a button, and it's there. So that is automation. Link yourself to restaurant. Or location. Tap button for directions. Go. And so for the pro, this is, so I typed this process all up here. But these are some additional examples in case you are walking people through. How this automation system works. And if you are curious about what this KLM travel assist does. We'll go to it. The better one is videos. Oh, they added another feature. That's helpful. There's always something new. Yeah. KLM, they had 1.4 million customers searching it. Wow. That is a lot. So they integrated it with their customer service support. And there is now a messenger option. So now instead of logging in for pre-flight check-in, you can just text KLM's messenger bot. And it'll do it. And then it'll send you that itinerary and then it'll automatically add into your calendar. And then you'll get the notifications and reminders. And it's all automated. So, okay. So now one thing I didn't add in the lesson plan, but I will add right here. What does this mean about jobs? Which jobs are created? Which jobs will no longer exist? Jobs, things will change. Yeah. So in terms of which jobs are created, we now have app designers, graphic designers for videos and marketing. And we have user experience designers. We have user experience interface designers. And so user experience designers would be the ones that go through and talk to different customers and interview different customers, find out exactly how they run through the process of booking travel appointments and how they like the little pain points along the way. So they would have been the ones talking to people who say, yeah, I went down to Seattle, but it was started raining. I really wish I'd remembered to bring my umbrella so I could have had it in the airport. And then they say, oh really, there's an app for that. And they just kind of make sure that app turns into a user experience instead of just book of light. That's what's helping companies get more competitive and start bringing more people in. And I think that KLM chatbot brought in millions of people. Well, almost millions. And user interface designers would be, so when you saw that app that was up there, there, they would be the ones that say, this is what this app's going to look like. This is how people are probably going to interact with it. And then you would have a quality control analysts spell check automation. Yay. And quality control analysts would look through the final design document. They would look at the request from the stakeholders and from KLM company and KLM executives and say, okay, this design meshes in with everything that the executives were looking for. We are a go. And they would also look at the completed develop product to say, okay, this works the way we wanted it to. So there are also app developers, machine learning engineers. And then there are data analysts. There are data collection specialists. And then there are customer care representatives. Because people aren't going to go away. People are going to still want to talk to people. Yeah. And, but the thing is the way they work is going to change. So customer care. If you think about the example of an ATM. Where put into banks, everyone was terrified that the teller was no longer going to exist. But they still do. Yeah. It's because instead of just handing people money or having people walk up and say, can I withdraw $20? Instead, they do more personalized care. They say you can come into a bank will help you figure out your retirement will help you set up a personal loan will help you do a variety of things. And that's why they say that the skill level for a lot of these jobs is changing. And it's because those wrote tasks are disappearing and they're being automated by ATMs and those apps that I just talked about. And it's shifting over so that you have to have a higher digital skill to get into any even entry level job. And there are even hotel people who work in hotels that now have to use the Internet of Things scanners and be able to know how to interact with the software app that says this is the next room that you have to go to skip this room but go to this room. I worked with a woman from Ethiopia who was trying to improve her digital skills here in Lincoln. And she was talking about this device that she had to learn for a hotel she's working in. And she was talking about how she didn't feel prepared about how to use to interact with this software. And she didn't know how to excel in that field and how to excel and move up in her job because she had to understand that software to go anywhere. And she had to be able to we talk like we talked about it and she said well how do you ever get anywhere and how do you move forward in this country. And we thought we looked at the software and said well you start off by using the technology in a job. And this one you're trained specially to do it. But as you move up you understand the technology well enough to make recommendations to be able to improve it. And then you start making recommendations for how you can improve other processes in the job to be able to make life easier. So maybe instead of a software app that just tells you where you're supposed to go and where your next room is to clean. You would build a software app that will monitor the number of clean towels that are available. And then where you would need to be able to route towels from so that you don't have to move as many carts or have to walk quite like all the way across the hotel to be able to get them. And that's supply chain monitoring. And no one ever thinks about the towels in a hotel as a supply chain, but it is. It's just supply chains are things start out in one place. They get routed through different people in places and get to their end destination towels will start out in one main washroom store room. An employee will fold them and put them into a main storage room. And then the towels get routed out through different rolling carts that are picked up by the employees and distributed throughout the building. But if you have different sensors and monitoring systems that tell you a better system or where those towels should go. You're saving the backs of a whole lot of hotel workers and you're also helping them save time so they don't have to run back and forth across the hotel to go back to that main supply room. You pretty much know exactly which quadrant of the hotel to put it. So it's just little things like that, like it can seem like the simplest thing of where towels are going to go. And they now have a delivery robot that's going through different hotels. And he's kind of cute. He's awesome. But like the. So if you order room service in a certain hotel chain now. The liver your stuff. There's like a little door you open up in the food to gives it a room number, and then the robot will roll on over. And then it's going to go through the elevator system and it'll not like knock on the notify the person that they're there. The person will pop up in the little door and turns out people like robots. Safe ones that are cute and helpful yes. And then here to let everyone know who's on line here we are a little after 11am we do the other shows usually an hour long will officially scheduled to be an hour long, but we don't get cut off, just because it hit 11 obviously. So, we'll stick around as long as it takes me to wrap things up and if anybody doesn't have any questions if you do have any questions. Nobody's put anything in yet, but type into the questions section if you want to more about something you have you're confused about something you want something. Or explain explain more or if you have something that hasn't mentioned get in there so we can get your questions answered. And so I'll just go over which jobs will no longer exist. And that is wrote tasks. Anything that has specific step by procedures. And this is not just manufacturing. This is also starting to look at paralegals and they're looking at machine learning can now study text documents. Look at different patterns in the text and make recommendations for which legal cases would be most applicable to the case you're looking at right now. So that is speeding up the time that paralegals have to do their work and it's also reducing the amount of back work that they would have to do in a legal library to be able to find the information. And it's changing the way that paralegals are working. And it's, it's still a toss up as to whether those jobs are going to reduce or if they're just going to shift. And it's going to be that agility to be able to recommend different tasks that you're able to do in your position to be able to adapt. And that's why when they say adaptability is key. That's why. And if you want to learn more about it, the World Economic Forum and the Department of Labor is a good place to go. And MIT just put out a course on Coursera. Is it Coursera or edX about shaping the future of work. I'm in the middle of the course right now, but I don't remember if it's Coursera or edX. It's between both sites so much that it doesn't even, it's one of them. Should be able to find it easily. Yeah. Yeah. It's a flip the coin and if it's not on one, it's the other. But so it's 1111 now. And so I will kind of wrap up that machine learning basics. And I will share out this slide deck here, just in case you want to be able to see the links or access the links. And I'll put that into the chat. And of course, this is also being recorded. So it'll, when the recording goes up, this will have a link directly to the slides as well. Yeah, feel all of access to it then too. But if there are no questions, that is about the long and the short of it. It doesn't live in any right now. And that's fine. If you guys couldn't think of anything to ask right now, that's perfectly fine. This is a lot of information. I love the example, the little pre-made lessons. I think that's really important. So as I mentioned the beginning that doing, you know, teaching kids programming things for some people can be fun or terrifying. And if I terrifying, I mean, I don't know anything about programming how the heck do I even do this type of terrifying. So those pre-made lessons there, I think be very helpful and hopefully a lot of you will be able to use them to get this kind of programs happening at your libraries. And if you want help being able to set up any of these programs just give me an email. My email is, so if you have any questions or want help setting up a program. Let me know. Absolutely. Thank you, Amanda. Thank you, everybody. I think this is great. Like I said, lots of great resources, lots of great things you guys can grab and use without having a lot of programming knowledge yourself, which is awesome. But that is why we have Amanda here so she can help you get some of that knowledge maybe if you want to experiment more, give her a call or email. Yeah. All right, I am going to pull presenter control back to my screen. Not that one. There we go. All right, so I guess that will wrap it up for today's show. Thank you so much for being with us here again, Amanda. We'll see you here again. So what's the next September 30th is the next Wednesday, last Wednesday of the month. See what we'll have next week. Next time. As I said, the show has been recorded and will be available on our website here every by the end of the week, everyone who attended today or pre registered or was registered for today show get an email from me letting you know when the recording is ready. Also push it out to our social media. We have a Facebook page for encompass live if you are like use Facebook give us a like over there it's linked from all of our help pages. We post reminders here's reminder about today's show when recordings are available information about our speakers. So do give us a like over there if you want to. We also use on other social media like Instagram and Twitter and comp live as our hashtag. Our archives go right here underneath our upcoming shows most recent one at the top of the list so today's will be right at the top here. And we'll have a link to the recording in our YouTube and link to the slides the Google Doc slides that Amanda used. We have a search feature here for archives I want to mention, you can search our entire show archives or just the most recent 12 months. The reason we have that kind of limitation is this is the full archives of our show and compass live premiered in January 2009. So we have over 10 years worth of recordings here if I am not going to scroll all the way to the bottom because that would be crazy, but you can see it goes all the way back and we have everything here. So either if you want to look for a topic and you want something really recent and current limited to the most recent year, or some of our topics will stay on the test of time reading lists certain things are always being from good. But some of the information on here will become old and outdated being correct links might change services might no longer exist. But you can always just pay attention to the original broadcast date we always had the date there when it originally happened. And you can tell like this was for 2017. Possibly people may be doing different things with makerspace kits now so just pay attention to that whenever you are looking at any of our archives. But we are librarians we do archive things keep things for historical purposes we will always keep our full archives up here for everyone to go back and watch everything and anything if you want to just pay attention to the dates when you're watching something. So that will be for archives, so I'll be joining us next time I'm filling in dates here for September keep your eyes on our schedule I've got a few other things confirming for those middle dates, but next week we will talking about the toward gigabit libraries toolkit. This is a project we did a session about this about a year ago when it first came out and Carson block and Stephanie Senberg have been working on this it's kind of Carson's baby through the internet to and they have received a new grant from IMLS Institute Museum library services to expand this. So on this is a great resource for check to figuring out what's going on with your broadband what do you have what might you need how could you improve it. So definitely give us sign up for next week show if you want to find out more about this toolkit it's free available anybody to use our it out there. They're just going to be doing some great new improvements and updates to it. So please do sign up for that and ever other upcoming shows that we have coming up on the net thank you everyone for being here with us this morning. Thank you again Amanda for joining me today. And hopefully we'll see you on a future episode of encompass live. Thank you.