navarre.aiClarisTalk AI · the FileMaker AI podcast
How AI is fundamentally changing education and learning FileMaker
"Sage on the stage" versus "guide on the side" teaching Cris and Matt look back on how we taught FileMaker in the past, and what changes happened with ChatGPT 4, how the Sal Khan book 'Brave New Words' https://www.amazon.com/Brave-New-Words-Revolutionize-Education/dp/0593656954 Discussed the "2 Sigma problem" - how one-on-one tutoring dramatically improves learning outcomes, and looking for a larg
Apple PodcastsSpotifyYouTubeAll episodes
Transcript
Transcribed by machine (whisper, large model), 15,031 words. Names and product terms are sometimes misheard; the audio is the record.
Claris Talk AI. Chris Ippolite, my friend. Hello, Matt. How are you, buddy? I like the little... Yeah, for the YouTube folks, I suppose, right? YouTube folks, for the podcast listeners, it was a little Victorian signal with the hand. Yeah, Victorian, I think. Yeah, it was a tip and a hat to my... Tip and a hat to you, sir. Fellow trainer, fellow podcast host, fellow AI expert, fellow friend. And Alexi Folger. Hey, Alexi. In the beginning as well as the end. So FileMaker 2024 slash 22. 25. 22 slash 25. What am I even confused? Okay, sure. 25 slash 22 is what I meant to say. It's not quite like OpenAI's model naming convention or anything like that. But I think it's got two names.
I think it's 20.5 and 22. 0.3 and 4.0 and yeah, yeah, yeah. Yeah, exactly. So I think the point is, though, we're going to not talk about it because everybody else is. And I think we're going to talk about implications of it and like what, when the dust settles, what are the things that's actually really useful. There's so many new find features that it's confusing. There's four commands in FileMaker that have the perform find function. Not counting perform find and replace, which is the weird layout one. Yeah, interesting. Like regular find, semantic find, and the two new ones. And those have a lot of features. Anyway, so, and I think connecting them all together, you know, it's like which one, which will do what the customer expects?
And the answer, none. So that's... Well, I think you're onto something though, too. I think part of when we deploy these features is, God, I think you just nailed on something really profound. And here's what I'm thinking. I mean, we can just end the podcast right here, actually, with you said something profound, and then we just sign off. No, here's my thoughts is, the customer, I don't think in the FileMaker community, and this is not a pejorative statement, this is just my observation. I don't think that there's going to be a groundswell of customers that are asking for things like semantic search, right? So I think what, not only do we have new AI, well, not that they don't need it.
Don't get me wrong. I'm saying there's a two-tier education of what the capabilities of these new features are. And it starts with, well, first of all, they exist. Then the developers have to understand how they work and what the use cases are. And by the way, that's what I think is really missing in the dialogue in our community is use cases on AI. I think that needs to be front and center. Just like, there's a couple of them on the Claris website that I was actually pretty impressed with. I tried to find them again, but they're somewhere in there. If you scroll down to the bottom of one of the pages, I think there's actually like a semantic search page or something, something like that.
I'm going to link to it. We'll put it in the show notes. Yeah. Okay, cool. I'm with you completely. And I've got, I don't know, five, six, seven design patterns, use cases that I think are pretty easy to implement and meaningful. And like, if you have a need for this particular thing, this is a drop-in thing that you should use, configure it to what you need. Well, but like, so you had to figure that out, right? Like you had to go, oh, what is the semantic search thing? Okay, what are the script steps? What are the options? I've been thinking a lot about it. How simple can it be? And then you have to figure out use cases because you have to communicate to your constituents, whether they be trainees or whether they're your customers about, well, this is why this is different.
And these are use cases where this would come in handy. And only then will customers start to be able to think, wow, I need this, right? Like this is something that I need because here's a problem that I think it can solve for me. But right now I think there's, I think where all the attention should be going is in anyone who's out there that's doing it. And I know there's people out there doing cool things with, I'm not just going to call it semantic search. It actually kind of bothers me quite a bit because the feature isn't semantic search. The feature is support for embeddings. And one of the things you can do is semantic search. You can also do matching and various other things.
And so that's important, but we should be elevating stories about who's actually using them and what problems they're solving. I mean, I'll put my, I try to put my money where my mouth is. My last two sessions at Engage have, literally just been, these are use cases of AI, like real stuff that people are using AI for. I even brought two customers with me at my last one and said, let's talk about what we're using AI for. And I think then the conversation kind of happens like, oh, that's interesting. Yeah, you know what? I have that same problem. Oh, this is a tool that can solve it. Rather than just like force feeding a bunch of features and just being like, here you go.
This is a bunch of stuff. Like, let's talk about problems that need to be solved. I really think that's where this begins and ends, honestly. Yeah, it dovetails into education, which is what we're really talking about. Yeah. So like, you know, like for example, if you have multilingual data, names in different languages, perform semantic find is magic. It's fricking magic. You can type a name in any language. I liked your demo on that. Find it in any language. Yeah, yep. And there's never been anything even close to that for FileMaker before. I don't know, maybe some, you know, difficult integrations with other engines. Basically, it has to be some sort of an embedding thing.
Yeah. I would say matching to me, you know, I keep getting this message of matching. Like, one of the key tactics, by the way, like duplicate checking is a good one too. One of the big things in like the AI development community, like I'm sitting here, like I, well, you know what? I do know actually. Yes. In one of the real tools that you can use, something called semantic scoring. And the semantic scoring, if I was to, you know, let's see, speaking of coming up in the example, like let's say you're saying I have, you know, a model with its latent intelligence. It's, you know, like some open source model that I got, and it's got no training. And I don't know how much it knows about, let's say FileMaker.
So I'll go and I'll come up with some prompt and completion pairs, like let's say 200 questions about FileMaker where it's a question and then the correct answer. And then what I do is I feed all those into this model and it comes back with its answer, right? So then what you do is you run embeddings on what the answer is supposed to be. And you run embeddings on what answer the model, gave you, and then what do you do? You match those together to see what semantic score you get. Like it's the cosine similarity of the answer you expected and the answer you got. And somewhere in there, well, you're gonna get a number, right? And somewhere in there is an acceptable threshold of accuracy, right?
So you'll say, well, out of these 200, really it should have been like 0.85, you know, accuracy for this. And so then what you do is you either go, this model is acceptable for moving forward, or you do other training, training tasks like you do, like, I don't know, introduce RAG into that workflow and then score your models or just do fine tuning, which it's such a rare, no one really does fine tuning, but let's say you do do fine tuning, then you'd run the same 20 questions again and then get your answers back and see if your score went up, right? That is all based on embeddings and that's not searching at all. Like that's like a matching type thing. So there's a ton of things we can do with this stuff.
So it's a matter of understanding use cases and getting it out there and really just educating people on what this is. And we absolutely are talking about educating people on AI, but one thing that you and I have in common is that we're both educators, we're both trainers. Well, maybe you. And so, you know, I've been thinking a lot about the impact of AI on education. And I say that in a couple of frames, like first, you know, my wife is a teacher, she's a elementary school special ed teacher. And in about two weeks, this is a big household victory for me, Matt. I finally got her to read a book on AI. So if you're out there and you're hearing me talk and you're thinking, gosh, there's a blight droning on about AI again, and you're not gonna listen.
Well, it happens to me in my home on a daily basis. My wife is a slow adopter, but that's what makes our relationship work, frankly. But she just read the "Brave New Words." You remember the sale con book we talked about last year. She read that because in two weeks, she's doing a AI for teachers, like a multi-week course that her school district is putting out and I'm really impressed. - "Sale Con Academy" book? Is that the one? - Yes, "Sale Con's Brave New Words" book in there. And there's some interesting points in that book as well too. And as it turns out, I'm also doing a speaking gig on Friday about AI in education. So I got a lot of this stuff top of mind.
And it's really, so we're not talking about like teaching people about AI. We were just talking about that. Now we're talking about like, let's look at the impact that AI can have on education itself. 'Cause I think there's some profound shifts happening and I'm very fascinated to see how this all plays out. - Absolutely. - And everything from school to university to just like how people can teach themselves new things. And these tools are really, they're here. They're definitely here. Some of those, I wanna get your opinion on something actually. So, I used to do a lot of training for Linda Weinman. Linda.com, do you remember Linda.com? And then it got bought by LinkedIn Learning.
So before Linda became a huge billionaire, because she sold to Microsoft is what I'm saying. Like that turned out really well for her. She, you know what, actually, let me share something with you, buddy. Here, let's do. There we go. So she had this philosophy that, can you see what I'm looking at here? Yeah, okay, cool. So she had this philosophy-- - We have every word for the podcast people. - Yes, just, we'll speak to it. It's just some visual. Don't worry, you're not missing anything cool. - The scientific paper. - A scientific paper on adapting to the AI disruption, the IT landscape and educational paradigms. Actually, let me move to this next one. This is the one that actually, so what we're looking at right now is a paper from 1993, from a psychologist named Alison King, and it's called "From Sage on the Stage to Guide on the Side."
So let's talk about this. I'm really interested in your opinion on this. - You've talked about this before, yeah. - Yeah, and so Linda was really big on Guide on the Side. I didn't really, Linda Weinman, that was from lynda.com. I didn't really know what she was talking about back in the day. And believe it or not, we used to have access to her as authors. Like when they were still a little mom and pop shop, every Wednesday they'd take all the authors out to dinner up in Ohio, California, and it was really cool. I got to know her pretty well and her husband, Bruce, and they went to dinner. And it wouldn't take my calls now, of course, but because of me, not them.
But she talked about Guide on the Side. And now to kind of punctuate this, the technology at the time, the thing that actually made her a billionaire, believe it or not, was that she was like first to market on VHS training. I know it's insane to think about that. First of all, for the kids, well, there's no kids listening, so nevermind. - There's some. - Okay, so VHS used to be these tapes that we would put into this big, huge, bulky machine and then feed it into our TV. - What's a tape? - Exactly. And then we'd sit, I'm even trying to think about how this was effective. Like you'd sit in your living room, I guess. This was actually before laptops were really prevalent.
So like, I don't even know how that was working. I did it. I don't really, I can't remember how I did it. But the point was she was saying, "Look, this is the closest thing we've got "to this Guide on the Side philosophy." And so what's your take on Sage on the Stage? What do you think this difference is between Sage on the Stage and Guide on the Side, besides their fun words? - It's funny 'cause I think about these things a lot, like looking at these terms in the classes I teach. So Sage on the Stage is kind of what a lot of trainings do. Honestly, the way kind of we used to even teach FileMaker, like standing in front of the room with a podium, talking about the features and showing them in kind of a slick way, but not having the people in the class engage and check what they're doing, not adjusting the speed of what you're doing to what they're doing.
- Literally just kind of like spewing knowledge, right? - Yeah, spewing knowledge, yeah, yeah. - In this paper that we're looking at-- - Yeah, knowledge, knowledge, not skills. - Yeah, well, 'cause you're really not checking in to see if skills are, whether it's sticking, right? And so in this paper that we're looking at, this, Alison King talked about the transmittal model. So what you just described is transmittal. It's literally like, I'm the lecturer, I have all the knowledge, I'm just like transferring, I'm transmitting the knowledge into this student who is an empty vessel on this topic. - Right. - And all I'm doing is just filling up their brain with knowledge, right?
And there's really no, like, you really nailed it, there's no checking in with them, adaptability, adjusting, nothing dynamic, like literally every single person in the classroom gets the same information, they all walk out of there with the same information. But what do we, I mean, the users aren't all the same, right? So the users are all at different places, there are different types of learners, they have different curiosities, whatever it is, but they're all getting the same thing. And for years, that was really the move. Now, the opposite, so you're taking, you're taking on guide on the side or what they call in this paper, the constructivist model. Oh, whoops.
- Constructivist, okay. - Yeah, the constructivist is the guide on the side model and the, there we go. Well, nice job here, Chris, there we go. So the constructivist model, so the people who are listening can see what I just did there. Constructivist is guide on the side. That means that it's one-on-one, they're focused on the learner, they're focused on adaptability and making the learner be more of a critical thinker than just somebody who's memorizing things, right? So that's really what Linda was talking about. She imagined that VHS, believe it or not, was that VHS was actually the thing that was the guide on the side, because the person could sit there and they could sort of like hit pause or rewind.
Like that was a big thing for her, the pause. Pause and rewind stuff was a big deal. Now, of course that, you know, moved into video, right? And that's really what the paradigm is right now, right? Is that whole video thing. So your take on that, the differences there. - Well, for one, it's not a guide per se, because I think that, I think all of those things really totally still meet the definition of sage on the stage, because they don't, they don't involve the, what am I trying to say? - Well, there's no, there's no feedback loop, right? - Yeah, there's no feedback loop. There's no accountability. - That's correct. Yeah, that's another great point. Yep. - So I just, I was thinking about something else and I found this thing that I'll share here.
This is called the Learning Pyramid. - Ooh. - So, and this is, this I used very strongly in the past, but this is how I teach, right? Here's your lecture. This is your sage on the stage, standing in front of the room. And then, yeah, so if you, so you, this is retention. I think this is really about retention, because there's really two aspects of training that are important, engagement and retention. Engagement doesn't guarantee retention, but it's a big part of it. And so that's the worst one, right? - That's a huge, huge part. - Audiovisual, this is your videos. Demonstration, when you actually show stuff, this is your DEVCON video. Hey, look at your dog. - Yeah, it's Paxton.
He's our little guest here. So this is one of the bonuses for watching on YouTube. You get to see Pax, but-- - Yeah, so then you get this discussion where you start bringing in kinesthetic stuff in and practice doing. Practice doing is as far as I can get in easily in the classes I teach. This is the level I'm on when I teach all of the classes I do. Because everyone's sharing their screen with Zoom. Everyone's actually doing it while we're in class. I don't share my screen. They share their screen, the students. And so every single student during class has to demonstrate that they know how to do what we're doing. So that was something that I could actually adapt to.
It would take, I love to have the people in the class teach others, but I don't think that's realistic to do in a class where you have an hour or two hours. - Yeah, that requires a very specific structure, right? - It would be great to have that be homework, right? Like have students pair up and-- - Well, actually, here's a good transition. So you'd love to have people teach others, but what-- Why is that not realistic? I'm actually trying to-- - Time is really the only thing. - Well, and frankly, resources too, right? Like you'd have to have one person. I'm actually deliberately trying to hold on. Pax wants to share something, so hold on. So let me show you a screen for those that are watching.
Hey, buddy, I'm gonna have you sit down right there. There we go. So for those that are watching this, I've got something on screen called the-- The Two Sigma, there we go, the Two Sigma Problem. So let me actually dovetail this in for what you said. It would be great if we could have a classroom full of people teaching other people, right? But that would actually be basically tutors. You have 10 people in a class and 10 tutors, right? And that really is the ideal. And the reason that's the ideal is because the tutor themselves can adjust to whatever learning level the user's at. The tutor themselves can stop and-- - Wow. - All that stuff. - It's not tutors, it's the students teaching each other.
- I know, but I'm trying to, Matt, I'm trying to-- - Yeah, I'm sorry, sorry, sorry, you know, go ahead. - Yeah, so there's this paper, 'cause we're talking about AI in education, not just education here. So let me transition into something. What I'm talking about here is this really interesting study from back in 1984, and that's meaningful because of where tech wasn't in 1984. This guy named Benjamin Bloom did this study that he called the Two Sigma Problem. What he discovered there was that students, that had one-on-one tutoring, right? Which is why I'm bringing that up, and I, you know, so apologies for shifting gears there, but performed in the 90th percentile as opposed to conventional learning.
So like students who would be in the 50th percentile by just having tutoring would move to the 98th percentile, which he called the Two Sigma Improvement in learning, which of course, sigma is the, you know, the symbol for standard deviation. So he's basically saying-- - Yeah, which is funny, that's actually the Greek letter S there, but probably a lot of people don't know that. - Oh, interesting. - And there's two different symbols for the Greek letter S, depending upon if it's the middle of the sentence or the end of the, or middle of a word or the end of the word. Anyway, continue. - Oh, I appreciate, I like the way the Greek is thinking that through. So, so the reason I bring this up is because, because what he actually, you know, essentially was, was talking about back then was like, boy, it sure would be great.
Like he had, he was a proponent of this thing called mastery learning, which is actually what you're talking about, which is more, you know, specific to the user, there's more of a feedback loop there. They're actually asked to do some critical thinking, but what he really opined was, hey, ideally one-on-one tutoring for every student in the world would be incredible because they could actually do a two-sigma improvement on their education. That would be great. But it's 1984, how the heck are we gonna do that? I mean, this was like, wasn't 1984 like even pre-VHS? Maybe not, maybe it was the heart of the VHS era. - Yeah, I don't know. I remember like recording "Star Trek Next Generation" episodes on VHS when they were on TV.
I have that memory. - I was holding on, I was holding on tight to Betamax, but I guess the- - That's what they were filmed on actually, "Next Generation." They didn't shoot on film, they shot on video. - Interesting. - And that's why "Next Generation" looks like crap now. And the original series looks amazing because the original series was shot on film and they can transfer that to high resolution anyway. - Another one for our "Star Trek" fans out there. And if there are any, go watch the, is "AI Conscious" episode from last year. We dove deep into that. But the idea here is that in 1984, this guy had this study and he wasn't trying to be negative by saying problem.
He's like, look, if we can figure out how to get a one-on-one tutor for everybody out there, they'll get a two-sigma improvement in education, right? - Right. - Well, that's where, and specifically what he said was, and for those that are listening, he said, "I believe an important task of research and instruction "is to seek ways of accomplishing this under more practical "realistic conditions than one-on-one tutoring "for every student." He's just saying that would be great, but it's just not practical. If the research on the two-sigma problem yields practical methods, applicable at a large scale, he's just basically like, we don't have one teacher for every student, it would change the face of education.
I think both you and I agree that we do that. - I mean, just reading that whole thing and thinking about AI, it's like so obvious. So I think I see where you're going. - Yeah, yeah, there you go. Sorry, I'm just asking for a little, I did a little bit of an abrupt thing there, no question about it. And there were a lot of improvements, by the way, since 1984. There was, I mean, from the mid '80s all the way to the 2000s, there were mastery learning, peer tutoring, intelligent tutoring system, all sorts of stuff. - And a lot of them are crap, right? Like the Microsoft CD-ROMs or whatever. - Oh, yeah. - And then some things had a pretty big impact, like Wikipedia, but nowhere near as big as where we are now.
Anyway, continue. - Yeah. - As a matter of fact, to try to get to that one-on-one tutoring, which is that two-sigma improvement, like the closest thing that ever got there was a one-sigma improvement, which is cognitive tutors. This is like the most recent stuff. This was in the 2000s. Domain-specific stuff, limited adaptability. It was all feedback, you know, kind of related. - A computer technology thing, or? - Yeah, it was basically computer. It was basically just like-- - I don't even remember that. - They're like, here's how everybody's doing. So maybe I should adjust the next lesson or something like that. You know, it's like that kind of a concept. So we talked about Sal Khan.
One of the things, you know, that he mentioned in this book, the "Brave New Words," if anyone's interested in. Actually, it's a good primer. It wasn't a very sophisticated take, but it really touches on a lot of important stuff, including the-- - I thought it was actually a great book. - I mean, I'm saying if you're like worried that it's all AI stuff, it's not at all. It's actually just like easy, approachable, easily digestible. Read it. It's not a weekend type stuff. So I still strongly recommend it. He talks about this two-segment problem in there. And ultimately what the answer is, is AI, right? Like being able to have AI as your true guide on the side. Every single learner can have this.
And is it scalable? I mean, it's limited by people's access to the internet, certainly, right? And having machines to be able to get to the, you know, internet and be able to have access to AI, but it's scalable. - That's not that much of a limit. I mean, if you take a look at the numbers of smartphones around the world, it's a very large number. - Yeah, and I think there are big tech companies that are endeavoring to make that 100% saturation as quickly as they can. - Yeah, yeah, yeah. I don't know how far we are. Obviously there's a lot of places where you can't get it. There's no service or whatever, but it's a-- I mean, we're closer to that than ever, right?
I mean, like we're within reach of that, certainly. And actually, you know, when we were at Engage, I was like not sleeping and I'm just like flipping through my timelines on stuff. And I think because I was in Austin, something showed up on my newsfeed and it's this school that's based out of Austin called Alpha School. A little bit controversial, I'm not gonna necessarily say that everything about them is great, but there was just one fascinating part about them. And that is that they have this structure where they have, where they do two hour learning. So they have like the days broken up between morning and afternoon. And I might be butchering this to some degree, but you know, I've spent some time checking it out.
And in the morning, the students are all doing AI tutored, exactly what we're talking about. They have their own personalized system. It knows who Sally is, where she's at on her learning, what's her Lexile level and everything, what are her challenges, all of that. - Very customized. 100%, 100% customized. - Well, what I mean is it's, what am I trying? So very early on when we talked about it, you're telling the model what its job is, right? - Yeah. - So you can't just open up ChatGPT and do this. You have to go in and say, here's what you are. You are a, you know, you're helping me, Sally, a nine-year-old girl learn. These are the things that I really like.
This is the place that I live. You know, these are the resources that, you know, I have a bike and I have blah, blah, blah. You know, this is how I learned best. I have ADHD. These are the meds I'm on. You know, really, really detailed about that. So it can- - As detailed as possible. As detailed as possible. You're absolutely right. And if you think about where technology can go, what you just described is writing a big prompt that makes it so that they know Sally and they know where Sally is today at the moment she's about to engage with this lesson, right? But if you think about it, technology absolutely exists to be able to track a conversation that Sally's having or an interaction that she's having with course material or whatever it may be and see where she stumbles, see where she excels, see where she, you know, hits the back button, whatever it is, and track all that information.
There's actually a lot of, one of the emerging, one of like, definitely one of the, like where the puck is going fields within AI right now is on memory and statefulness within AI. 'Cause if you think about it, I mean, well, people aren't probably familiar with the personalization options in like ChatGPT and Grok and Claude and all that where you just go in there and be like, hey, I'm Chris, I don't like exclamation points or whatever, right? - Yeah, you can also like, when you're just using it for whatever, you can say, remember this. You can say like, I don't like exclamation points, remember that, or anything like that, that changes. And it will, and then there's a little area you can go in that says, these are all the things that you told me to remember, and you can delete 'em.
- Yes, forgetfulness. So that's adding state to memories. Maybe some people don't even know, like, that that's been turned on, because it is by default on ChatGPT. So maybe you've noticed sometimes where it's referenced another conversation you had. But there's a mechanism behind the scene that's gathering all these. And that's personalization and memory. And that is absolutely where AI is going. I mean, you think about it from an education standpoint, where it knows everything about Sally, every course she's ever taken. That's what Alpha School did. I mean, loosely, I dug into it. I was a little more impressed on the surface before I went, than I was when I got down to what they were really doing.
But what they're really doing is piping in, actually, Sal Khan's Khan Academy material, some other material, like Magic School stuff, some other, like, just standard courseware. But what they're doing is they're tracking the engagements and the feedback part of it, right? So every student has their own tutor that is not just, like, spewing knowledge or reacting in an AI tutor type way, but it knows them, like, perfectly. - Mm-hmm. - Like, if Sally's stuck on this one thing, they'll spend all day with her before they get past it, right? - Yep. - So it's interesting. And then, just for whatever it's worth, then they spend the whole second half of the day because the kids are accelerating.
They're basically getting a full day's education in half a day because of this, like, concept that they have. So the second half of the day, they teach them, like, soft skills, and they let them do, like, lab work and stuff like that. It's actually kind of cool. And what that's led to is that they have multiple different schools, and all of the schools are now, their classes are in the, top 2% in testing in the country. So this is working, right? Like, it's actually, like, a thing. - So are the teachers, like, guiding the students and taking a look at their AI conversations and taking a look at their scores? You know, obviously, the AI agent is gonna be able to track, without making a mistake, it's gonna know how the student did on all the tests, how they're progressing, you know, the speed of their progression, stuff like that.
But then the teacher can look at it without necessarily looking at every single thing they type, so there's still some privacy. - The teacher's role changes, and the teacher's role changes. It's like a facilitator rather than that lecturer that we talked about before, right? - Yeah, and they can see where the student is and make a human judgment for, like, you know, this is the stuff that you should learn next now that you're here. - Bingo, yes. And, you know, and then there's, obviously, there's, you know, high-touch human elements and there's edge case stuff. - Yeah, I'm loving this. This is exactly how my wife is learning Greek, by the way. She is so into ChatGPT for that.
Because you can, you know, you can say, you know, "Give me a test." There's a thing called language transfer, which is amazing. And it has hundreds of lessons that help you learn. ChatGPT knows every single lesson. So you could say, "Quiz me about language transfer lesson, "you know, 126," and it'll know the whole thing word for word. It'll know all the words you're supposed to learn. Yeah, it's amazing. - I have a scoop for you. - It's amazing. - I have a scoop for you and for your wife. Now, check this out. So you might have remembered, for those that are watching. - I've heard this video before. - Yeah, just as sort of a reference point. You remember this video where we played the, it was back when like the Scarlett Johansson voice was the thing on ChatGPT 4.
And you could do side by side. You can still do this. And actually they never really, they never, did they ever release the screen broadcast part? So yeah, what we're looking at is this is actually Sal Khan with his kid and they were doing a demonstration on the day that the GPT 4.0 came out. And they're sitting on an iPad and on the right-hand side of the iPad is a Khan Academy. Like a geometry course. And on the left-hand side is what we all now know is ChatGPT's voice mode. And it had the broadcast feature on. I think, God, how do I not know this? But the broadcast feature lets ChatGPT see the screen. It's the same thing as if you're talking to your ChatGPT voice mode with the camera on.
Like it can see your environment, but in this time it's your screen. - Put a pin in that to talk about the brand new feature that dropped a couple of days ago with ChatGPT, by the way. - Oh, I, is it this? Oh, a little bit of a scoop is that ChatGPT dropped in limited release. I keep hitting refresh and keep trying to update my application to get the new study together feature. Is that what you're talking about? - Oh, that's not what I was talking about. - Oh, tell me what you're talking about before I- - It's like a virtual computer thing where it can run its own session. - Oh yeah, the agent mode. - And you can have it like log into a FileMaker database or whatever.
Yeah, yeah, yeah. And then actually operate things that it never has seen before. - Agent mode is insane. Agent mode is really- - A combination of a operator. Did you ever get access to operator? It's, or computer mode on Cloud. It's basically where it gives you like a, like almost a Citrix type environment, you know, your own, you know, instance. And then it, what it does is it takes a ton of different snapshots so it sees what's on screen. And then it actually gives direction to the computer based on the snapshots and it can move your cursor around and perform functions. And so the evolution of, of computer mode was operator and the evolution of operator is now agent mode.
And it combines that with reasoning and some of the, obviously a bunch of tool use and, and now, and it creates agent mode. As a matter of fact, I was, you know, when I saw Todd Geist at Engage, I stopped him and I was like, Hey, cause he, you know, he has that cool new tool. The, speaking of AI and education, the, the, the script, I don't, I don't know what it's called. We'll find it. We'll try to put it in. Show notes, but where you type in, Hey, I want a calculation or something. And then it, you know, it does the calc, it goes to like open AI and creates the calculation for you. Right. And he's got some instruction in there. And back when he was like first doing it, I ran up to him at Engage the day after.
And I was like, Hey, you should run that in operator mode. Like, I'd be curious if you fed your whole PRD, like your whole, your projects back to operator. So the agent knows the, all the different calculations and all the different scripts that you need, but what, but the tool that you have is this tool that you created that requires someone to actually interact with it. I think it would be fascinating to actually point operator or now agent mode at like a online based file maker, calculation and file maker, script generation tool, and have it actually do multiple things and actually create, you know, elements of your project. I mean, that's a glimpse into where we're going in the future, but the big scoop and you're, that was a scoop that you just dropped there.
That was a big one. This one I haven't seen yet, but it's in, like some testing mode, it's called study together. You'll eventually know it when you see it, because it'll be at the bottom of your tools dropdown. We're looking at some screenshots here in case anybody wants to see what it looks like. You can go on the YouTube, but study together does some very specific things. It's really some instruction that's on the backend. First of all, it asks you a bunch of questions. This is what we were just talking about before. The first question that I'll ask you is they'll say, Hey Sally, what's your current level? Have you already studied key sections like selection four, five and six, whatever.
Right? Whatever the topic is, what's your goal? Are you, you know, studying for a test or do you have a quiz? Are you just curious? Like, where are we at on this? Are you like trying to tutor somebody? And then what's your preferred approach? Do you want real life cases, hypothetical situations, problem solving, whatever. And then it does a Socratic approach and you're the Greek representation. Why don't you tell everybody what the Socratic approach means, live from Greece? - Well, I mean, it's the question, a question. I don't really know a ton about it. I guess he didn't live where I live. - Oh, okay. - Basically like asking a lot of questions, but you can, you can probably.
- Never gives the answer, nudges, nudges and encourages. And it's, it's literally meant to continue through a, a deliberate set of questioning to actually get to the user to actually discover the answer on their own. - Right, I think, yeah, yeah. - And so this is really what this study together piece is. It's like the, this is a deliberate approach to, to education to, to allow any learner anywhere with ChatGPT, to get access. And unlikely, I think there's probably gonna be some sort of, I mean, my guess for this is that we'll see this later in the year, we'll talk about it on some other podcasts where it's literally like a tool that they're for education.
You know, they have ChatGPT for business. They'll likely have ChatGPT for education. And then that will really start the sort of like the big shift in how everybody's learning. And so I don't know your take, what do you, like, is, I mean, I think this is the, this is the path to the two sigma, to solve the two sigma problem. And I mean, are we missing something here? I mean, is this not the, like what's wrong with this approach? - I mean, I don't have this to play with, but I mean, even what we have now is really kind of approaching it. It's just, you know, all this stuff we're seeing now is taking things that are already there and refining them and connecting them together and, and sort of like narrowing it and focusing it really.
Right. With those big sets of prompts and best sets of, of system, I don't know, I'm not using the right word term for it, but. - No, you are, it's system prompts and instruction. It's instruction, you know, it's like system prompts stuff and other, you know, instruction, the, the looking at it from an educator standpoint, you know, something just occurred to me, like from a FileMaker class standpoint, what I, what I used to do, and maybe you do the same thing. I would have people come to these classes. It was actually a multi, I was, I was very proud of a multi, modal, I don't know, different formats of training that I could offer to people. So classroom was one thing I'd love classroom.
Frankly, I miss it. You know, COVID kind of killed it for me, but, and now I just do like the big ones at like training day. And there's just too many people to really engage. I used to love having 10 or 12 people in a class, looking them all in the eye, adjusting the speed of the course using their actual, you know, I was doing as much as I could as a human to not necessarily be the sage on the stage and try to be more of a guide on the side, smaller groups. Like I tried to do that with, with 30 people last year at engage you. And I'm going to do that again with an AI class there like in a three hour class, but it, but you, I mean, the, you can control, you can do a lot less content, a lot more deeply because you're making sure like most of the class can actually do it.
Um, yeah, we had, yeah, that's hard when you get, when you get a larger class and zoom is yeah. Zoom is magic. The other thing that's weird, let's throw it in there, throw it out there. If you're teaching in a classroom, if you have everybody in the class. Log into zoom, um, and turn off audio off microphone and everything, then you could ask anyone in the class, if they ask a question about how to do something where they're stuck on something, they can share their screen and it shows on the projector or on everybody else's screen. Uh, and then they can work through it. And the other students, other people in the room can sort of help. And then we're seeing someone actually gets stuck.
If one person's stuck, somebody else is stuck on a given thing. Um, so it's kind of weird. Yeah. Plus you can record the whole thing, including everyone's speaking. They can turn their microphone on while they're asking a question. It's just vastly, uh, better and you can send out sample files. So like as during the class you can send out, you know, data tables or another updated file or whatever, just in the zoom, cause it's perfect for distributing. Anyway. So that's one of the things I try to do in the zoom era. And, you know, Reluctantly, I, I forfeit. That classroom training is probably behind us. That was really, I felt like that was where I was at.
My best as a educator was just in front of people. Um, but I've seen you, I've seen you shine, uh, so many times when you had came up and taught classes for customers we had in Portland. It was great. Oh, thank you, man. I appreciate it. Those are, those are, I do it. It was probably like 20 something years ago that we're getting, but no, you know, even back then the approach that I, that I thought was the most effective was like, yeah. I would literally, I would say, look, even if you're coming to take my class, you should go watch my lynda.com course, which is the basics, like just that's your primer, do yourself a favor, do that. So now you're getting a, you're getting a video training.
Then you show up for in classroom. Then I would deliberately say to people, um, okay, look, you're probably going to have some, I would, I had this policy where anybody who took my class, they just had a lifetime subscription. They could come back to any public class I was teaching and take it again, just because like, you know, I'm not saying I'm, you know, assuming they're not learning something, but like, I know when I go to conferences or I take a class or whatever it is, like all of a sudden I like, I learned something and I start focusing on it. And then all of a sudden, you know, half a day has gone by and I've missed a bunch of stuff that the, you know, the instructor's talking about and I just want another shot at it.
So I would try. Yeah, yeah, yeah. So light bulbs come on and then that makes other things that didn't make sense the first time. Also the content keeps changing, right? And the content keeps changing too, like updated versions and whatnot. And then the other thing I did was we started offering mentoring to people in the class and meaning afterwards, which, which really need, not just coming back to take the class again, but you just probably have one or two questions. Let's do some zoom one-on-one mentoring stuff. And we ended up, that ended up being really popular. We actually have some other people here at iSolutions that really excel at the mentoring stuff.
Then towards the end, well, I don't know, we're not at the end or whatever, but actually, you know, then I started doing, um, remember 42U school and Quasar, those ring a bell. Those, those are these, uh, programs that Claris was doing where they would go to these coding universities and take coding kids and then teach them, like do fast track bootcamp file maker training. That's coming again with, uh, Apple, uh, the Apple developer Academy. Yeah, that's exactly what it is. So it's, I think maybe underway and so what happens is once the kid, once the learners come out the end of the tunnel, they are now available for hire or internships with, uh, file maker companies.
Right. So, um, I taught and I taught, um, I taught like, I mean, there's like a hundred or I don't know, 180 or something like that. Kids out there that went through that program that I got to teach. And I'm so proud of it because I go to DevCon, I see them speaking. I see where they end up. One's one's even ended up back, you know, uh, contracting with us at iSolutions. Like it's just so cool to see people journey. But one of the things I was experimenting with back there as the, okay, do the video training, show up for the class, do the mentoring afterwards. Then I created this thing. Um, you know, the first thing I ever did, we've probably talked about this on your, the Matt and Matt podcast, but the first thing I ever did with AI ever was I created the, and I just kept it to myself.
So whatever, you'll have to trust me that this actually happened, but the, the, the file maker learning companion, I just ended up calling it the Claris learning companion after a while, but it was just a file maker application that you could type in a question about file maker. And it would go to open AI and answer, you know, create a calculation or script or give you an answer. Sounds novel now, but dude, like in 2020, this was a big deal, right? Like, you know, like, oh my God, we can do this. So I started having the, the learners in the 42 is probably by the time I was doing quasar. Um, I had a couple of these cohorts where I was giving them that tool.
They were the only people I shared it with. And I would say, use this to embellish your training. So it was really kind of like the guide on the side or the AI, you know, tutor. I didn't, I'm not saying I did, I executed like the alpha school did or anything, but that was the intent of giving that other tool. I'll even say that one of the private onsite trainings that I did, it was a nursing school and, uh, you know, a bunch of adult learners and they had a student worker that worked with their department that they invited to the file maker training. And they said, look, she, um, you know, doesn't know anything about file maker. The rest of us, this is like the third time we've done this.
So we're like, okay, we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.
And then we're going to do this. And then we're going to do this. And then we're going to do this. And then we're going to do this.