← Hillsdale College K-12 Classical Education Podcast
Hillsdale College K-12 Classical Education Podcast
Craft Over Convenience: Building A Writing Tool That Fosters Independent Thought
00:39:16
Loading summary
Transcript74 lines
- [00:03]
A
Hey there, it's Scott Bertram, host of the Radio Free Hillsdale Hour. On this week's episode, a special edition of the program, we honor the life and legacy of Charlie Kirk, founder and President of Turning Point USA and friend to Hillsdale College. We hear reflections from Hillsdale College President Larry Arne, also from Hillsdale students, professors, and we hear from Charlie himself from a lecture he gave on behalf of Hillsdale College earlier in 2025. Together, we remember a voice that shaped a generation of conservatives and was silenced far too soon. This week, our special episode honoring the life and legacy of Charlie Kirk. Find it at Podcast Hillsdale Edu or wherever you get your audio.
- [01:05]
B
Welcome to The Hillsdale College K12 classical education podcast, bringing you insight into classical.
- [01:12]
C
Education and its unique emphasis on human.
- [01:14]
B
Virtue and moral character, responsible citizenship, content, rich curricula and teacher led classrooms. Now your host, Scott Bertram.
- [01:24]
A
Thanks for listening. The Hillsdale College K12 Classical Education Podcast is part of the Hillsdale College Podcast Network. More episodes at Podcast Hillsdale Edu or wherever you get your audio. You also can find more information on topics and ideas discussed on this show at our website, K12 Hillsdale. Eduardo Today we'll hear a conversation between Dr. Kathleen O', Toole, Associate Vice President for K12 Education at Hillsdale College, and Julian Peterson, son of best selling author and clinical psychologist Jordan Peterson. Julian also is the founder of Essay App. We'll hear more about that throughout this conversation. Here's Dr. Kathleen O'. Toole.
- [02:09]
B
Well, I'm delighted to be joined by Julian Peterson, the founder of Essay App, a fascinating and exciting new tool for improving not just the writing, but the writer. We're here at Hillsdale's K12 Summer Conference. Julian is one of our featured speakers and he's agreed to join us today to talk a little bit about Essay App and the process of writing itself. Julian, tell me this. Why is writing so hard?
- [02:42]
C
That is a great question. Why is writing so hard? Well, writing something good is hard. Writing isn't necessarily hard more generally. But in order to write something good, there's just so much you have to get right. And I'll kind of, I guess walk you through briefly what At Essay App we consider getting it right throughout the process. And that would be that it has to be something that you care about. It doesn't necessarily have to be something that anyone else cares about. It has to be something that you as the writer are taking seriously and something that you think matters. And so that already makes it a very hard thing because you have to understand what it is that matters to you. And you have to be pretty far along in your intellectual development to understand what that is and to be willing to take the time to investigate it. So that's partly what makes it hard. And then of course, there's just everything that goes into being an effective communicator. And writing is maybe the most difficult kind because it has to be succinct and readable, whereas in other forms of communication that's much less relevant.
- [03:51]
B
Yeah, that's right. In our work, we think a lot about how to teach teachers how to teach writing. And it's the most difficult part of our guidance. We do grammar, we do orthography for spelling, we do handwriting, you know, all of these things. And then verbal communication like delivering a speech, logic and rhetoric to understand the way that an argument is structured. But you know, we, we found ourselves searching for a formula for how to write. And there are lots of approaches out there how to teach middle school students how to write. But we found all of those unsatisfactory because they oversimplified the process. And the students would be churning out a kind of a boilerplate five paragraph essay that didn't really convey what they thought. It was more about the format of the assignment than the student's actual argument.
- [04:44]
C
That tends to be the way that writing instruction ends up being kind of delivered in terms of providing a structure and then essentially pushing students through a workbook or a formula to push out a five paragraph essay. And, and that is just not what it's about. Right. Like that's not the, the point of the five paragraph essay is that it's supposed to be maybe the most straightforward way of delivering an argument, maybe the shortest way that you can deliver a complex argument. Well, I guess you could do that in one paragraph, but. But something that's longer than that. So the structure is obviously important, but only insofar as it has a, as has the purpose of being a great way to deliver an argument. And so I think that usually when people are trying to teach writing, they focus way too heavily on structure and way, not enough on the content and the ideas, because that is a way harder thing to actually teach. Because how do you motivate people to actually investigate their ideas through writing? It's a much more challenging thing. So people are generally much more inclined to teach grammar to teach structure. And unfortunately, by teaching grammar and structure and over teaching those elements, you kind of teach students what the purpose of writing is wrong because you're overemphasizing structure and overemphasizing the rules and underemphasizing the creative aspect and the developmental aspect of writing. And that's, I think, partly why so many students disengage with the process.
- [06:17]
B
Yeah, I think that's just right. We have a professor here at Hillstone College called Justin Jackson. His name's Justin Jackson. He's in the English department, and he's famous for being a great teacher of writing and a really tough grader. And he would come to our summer conference, he still does it, and he'll just say to us, writing is thinking. Writing is thinking. And we say, okay, Justin, but how do we teach the students to do it? Teach them to think. And that's hard facts from Justin Jackson. But it's true, right?
- [06:48]
C
Unfortunately, that is just the case.
- [06:51]
B
Writing is thinking. So tell me about the genesis of essay app. I understand it came because your father's in his teaching career. He found himself wanting to give students better advice about writing. And so he created this document about how to do a good essay.
- [07:11]
C
That's right, yeah. Yeah. So he had graduate students, and I think maybe it was his undergrad, his like, relatively high year undergrads, fourth year undergrad students when he was at Harvard and then later when he was at the University of Toronto. And he found that although they were very smart and very capable students, usually the one thing they lacked was being able to coherently write something that had some sort of substance. Right. They tended to do the reading and kind of participate in class, but when it came to an actual writing assignment, they were very, very low quality, way lower quality than he would anticipate from people who in theory were that far along in their education. And so he was trying to figure out how to help people who maybe had never been asked to write an essay properly, you know, kind of following the structure that I outlined earlier in the. In the conversation. And he tried to kind of make that process explicit in a document and provide really structured steps for a student to get there so that they could not really have to think about it, just go step, step, step, step, step, and then start to understand the process through the actual practice of writing. And so he, yeah, made a document, and then that's the way that he would structure his writing assignments in, I think as the final project in his fourth year classes is he would give them the document and tell them that that's how they had to write their term paper was they had to follow that structure. And he got excellent results from it. And students did like it. And he used that for probably 15 over 15 years. And then, well, he got extremely famous and we ended up putting it on his website just as a writing guide. And we didn't advertise it or market at all, but it kind of gained some general popularity. And it was downloaded 150,000 times by various people. And we, we were getting feedback through email and just people talking to us on the street and be like, hey, I downloaded that writing document. And that was really useful for me in whatever context they were writing in. And at the time, I was building a completely different online education platform. I had been a software engineer for probably four or five years at that time, and I was becoming interested in building education specific software. But the plan that we had at that time was way, way too ambitious for, for the moment. It was actually now that these language models like ChatGPT and AI systems are on the scene, it was somewhat analogous to what they were trying to do in terms of kind of creating a library of information and resources that you could search for using prompts. That was what we were trying to build in about 2018. And partly we couldn't do it because it was too complicated, and partly we were just a little bit too early to that idea. But so we pivoted and we focused it focused in on writing because from my perspective and my dad agreed and we kind of had great consensus that this was one of the most significant problems in education that hadn't really been approached effectively with a technological solution. Khan Academy had done a really, really excellent job of starting to teach math really effectively through software. And other subjects had also been kind of approached. But writing, there'd never been something that was created that actually helped teach writing in a proper way. And so we thought that would be an unbelievably great problem to at least attempt to solve. And so that was kind of why we started it. Extremely meaningful problem, extremely difficult problem. One that maybe could have the most widespread impact on education.
- [10:59]
B
Yeah. And let's get into who's using SA app and how it's being used in a minute. But maybe you could just tell me about the experience of someone who first signs on.
- [11:09]
C
Sure.
- [11:10]
B
There's a person, a student maybe with a lot of ideas, maybe an assignment to complete. What are they going to see when they sign onto the app for the first time?
- [11:17]
C
Yeah. So we do a couple things. When you first sign up, we have this new feature that asks you a few questions and creates a profile for you. And we use that for a new tool that we've developed that is we're calling the topic Tool. And the whole idea behind this tool is it's supposed to enforce the idea or reinforce the idea, I suppose. Not enforce, but reinforce the idea that to start a writing project that's going to end up being a good writing project, you need to start from a place of intrinsic motivation, or at least. At least that has to be an aspect of the project. So it will. We create this profile, we ask them a couple questions, and then when they kind of enter Essay, they have a bunch of different routes they can take. So essay is kind of designed to provide students with a flexible structure. In most writing platforms, you go on, you click create new document, and it's an empty page. It essentially is the same thing as a typewriter. Right? You just have a page and.
- [12:16]
B
And the blinking cursor.
- [12:17]
C
Yeah, the blinking cursor staring you in the face. Exactly, exactly. You know, there are systems that have templates and things like that, but that's, I think, a completely different sort of thing. So on Essay, you go on. And we have a set of structured tools that's available. You can either start by collecting research, and we have a whole notes ecosystem for collecting research, organizing it, tagging it, and associating it with writing projects, or you can actually create a document and we can walk you through outlining your arguments. Perhaps you've chosen a topic using our topic tool, which kind of helps you refine your topic and figure out why it's interesting for you to pursue. And then you can break down your topic into sub questions, and then you can jump into the produce tool. And the produce tool is not a very technologically sophisticated tool, but from a philosophic, philosophical perspective, it's interesting because what it basically tells writers is that they shouldn't be editing while they're writing. That those two processes, editing and kind of the creative production that's a part of writing shouldn't really happen at the same time. And so we've broken down the writing process to make that more explicit. So you outline, then you allow your creative brain to kind of generate ideas. You put them on the page. Sort of like we're trying to emulate what people do when they hand write. Right? People when they hand write notes, or when you're reading and you're taking handwritten notes and you're writing down ideas, they tend to overproduce. Right? You just. It's more of a brain hand connection. Your ideas just flow out, and you can't edit that much when you're handwriting. So we're trying to enforce that idea in the Context of a word processor, which is a bit trickier, but. But it's an extremely useful thing if you can get someone to overproduce their ideas. Because then on the next stage when they're editing and refining, they have a whole lot more that. That they can work with a whole lot more raw material. And. And so we try to allow them and reinforce the idea of creating more raw material with their ideas. And then we have some cool, structured editing tools that are available. And that's kind of the basic process that we walk people through when they. When they enter essay.
- [14:24]
B
Yeah, that whole approach seems so effective and humane to me. I mean, I've. One thing I learned in graduate school was you can't start writing until you've decided what you want to say or you've chosen something that you were interested in writing about. And, you know, when you're in a position of having to produce a lot of writing as you are in graduate school, it's tempting to just think of it like, I gotta get this thing done.
- [14:48]
C
Yeah.
- [14:48]
B
But it won't really start rolling until you care and have become interested. And so the first step, you know, refining the topic and getting it to be something that you actually have thoughts about and want to have more thoughts about seems exactly the right place to start.
- [15:01]
C
It just makes the whole process way easier. Right. If you, if you don't. If you don't do that first, then you're just going to get stuck so many times when you're writing and it's. It's not going to end up very good and it'll just be way more arduous for the writer. But if you start from a place of intrinsic motivation or like you actually have some real questions that you want to ask about this aspect of whatever it is that you're supposed to be writing about, then it will. It comes just so much more naturally. So, so, yeah, building. That's. We're really excited about this, this topic tool because it starts people off on the right, right foot. And that's. And I don't think any other writing platform is attempting to do that.
- [15:40]
B
Yeah. And then the approach is get your thoughts out there. Create a messy draft.
- [15:45]
C
Yeah, exactly.
- [15:46]
B
And tell me about that. Tell me about the process of drafting. Tell me why that is effective for becoming a good writer.
- [15:53]
C
Yeah, well, there's a bunch of reasons, but it certainly reduces perfectionistic tendencies, which is a big deal. That was something I always struggled with when I was writing in university and really throughout my life. Even now, it's a little Bit difficult for me to write something that's messy. But basically getting comfortable with having rough ideas is the way that you actually learn to think. Because you have to play with ideas to become a good thinker. You have to do. There's a couple different ways, I guess you can do it as a human being attempting to think. You can have an internal dialogue. That's very difficult. Most people can't handle that. They can't actually throw an argument at themselves and then come up with a bunch of counterarguments in their head and then have those arguments behind the scenes and then become a coherent thinker about a subject. Or you can talk to someone else and throw ideas off them and see their responses and talk to a few people, develop your ideas that way. But the way that you can do it on your own is through this process of reading ideas, you know, getting ideas from the world and then kind of just putting them on the page and looking at them and kind of using a different part of your mind. Not the part of your mind that collects ideas and finds things interesting and, you know, notices things. That's. That's completely different. Right. That's not exactly the thinking part of your mind. That's the experience part of your mind. Or. And. But then once you have them on the page, then you can switch on, you know, your left brain and you can be like, okay, what exactly have I been collecting? What is this? And how do I turn it into something that's productive, usable? How do I turn it into a story that I can apply in my own life? How do I turn it into something that's valuable for. For, for me and for other people to read?
- [17:38]
B
Yeah. And so much of the, so much of the, so many of the challenges that we face when we're writing have to do with the system that we're writing in. So handwriting, you know, that's great for getting your messy ideas down, but you're always nervous, what if I lose this notepad? You know, and it also takes a while. And then you've, you've filled up the middle of the page with something and then you realize, oh, I meant to put something above that and there's no space. And so now what am I going to do? So then you go to a computer and my own experience with a typical word processor like Google Docs or Microsoft Word is because of the way that I've used that in the past, I kind of feel like I need to be doing extra polished writing because I'm in that program. Especially Microsoft Word.
- [18:20]
C
Yeah. Well, that's what it is. That's the tool. Right. It's a formatting tool, basically. Right. It's a typesetting tool. In some ways it's not exactly a creative tool.
- [18:31]
B
Right.
- [18:32]
C
And so, yeah, definitely. People in Google Docs, anything that has that same interface that's been around for 30 years now. People are, are conditioned to format their writing while they're writing.
- [18:44]
B
Yes.
- [18:44]
C
You know, you want to get the font right, right at the beginning and you want to like, do like, you know, silly things like that where it's like, no, like it's so irrelevant, but you get focused on it because that's what 80% of the tools that are available to you are for.
- [18:55]
B
Yeah. And then you sit down to write your dissertation and you spend 30 minutes, you know, working on the footnotes and the page number. Page number B. I did my time and it's like. But you did not think about the next step of the argument.
- [19:08]
C
We'll get to that argument. Let's get those page numbers in the right squad.
- [19:10]
B
Yeah, but essay seems to be a great solution to that because you can throw your messy thoughts on a page and not be concerned that they're going to remain messy, that the task of moving them around and cleaning them up is not arduous.
- [19:26]
C
Yeah, yeah, exactly.
- [19:27]
B
So tell us about the features that make that happen, the revising, the polishing process possible.
- [19:33]
C
Yeah, so we have kind of a set of features that make that easier. And the first one I'll talk about is called Rewrite. So Apple also just released a feature called Rewrite, and what it does is it'll completely rewrite your sentence for you in whatever tone you want, which is annoying. But our feature basically allows users, it lays out anything you've written, your entire draft, and it breaks it down into sentences on the left hand side of our interface. And first of all, that clarifies what you've written for you just on a glance, so you can kind of evaluate your, your essay at a different level of analysis, let's say. But. And then the other thing it allows you to do is break the task of refining writing into micro tasks so you no longer have to worry if the whole draft is working. All you have to do is worry about that first sentence and so you can click on the first sentence. And we have a few things that we're attempting to do with the way we've built the tool. The first one is to teach people that they don't necessarily need to improve the sentence when they're editing one of the things it's difficult to do is to create ideas and refine ideas at the same time. And so what we've tried to do with the rewrite tool is make it possible for people to iterate on ideas, learn to present them in different ways, and then after they've done that process of iteration, producing variation, not necessarily better. It's a bit Darwinian. You know, just produce variation and then see what fits. So what you can do is you can produce as many iterations as you can think of in the moment. It doesn't have to be better, just has to be different. And then you can click them and select the iteration that you've written, and you'll see it in the context of your paragraph. And we just kind of save that history of edits for you so you don't have to be worried. In a normal word processor, when you're doing something like this, if you're looking at a sentence and then doing micro edits on it, changing words out, changing the structure of the sentence, you end up losing what you had before. And so you have to be really thinking when you're trying to edit. Whereas in this case, it's a. It's a kind of a completely different way of. Of thinking about editing, where it's like, no, you're engaging your creative mind again and you're producing variation, and then you're discerning between options. And it teaches writers a couple things. One, that that's a useful thing to do, to iterate and then look at an idea in context. And then it also kind of teaches you to evaluate your own ideas and the different ways that you've expressed them. And so you can kind of go through your sentences one by one, or you can choose a particular sentence and you can create iterations. You can view the iterations in context, you can delete things or save them for later, which I'll go into perhaps a bit later, or you can mark them for review. And so that small feature is a really cool way of reducing friction in the editing process. So let's say you're going through your entire essay sentence by sentence, and trying to polish it up, making different versions of it, seeing what's better, seeing what doesn't fit. If you find a sentence that you can't really do that with, or you're stuck, then you can move on, or you can say, well, I'd actually like to come back to that a bit later. You can click this little button called Mark for review on our platform, and it just kind of puts a badge on it and puts a little data point on the sentence and then you can come back the next day or whenever you're doing your next round of edits. You can filter all your sentences in the rewrite tool and you can just select the ones you've marked for review. And so it kind of gives you this way of interacting with your essay where let's say you've written like a 30,000 word thesis. It gets very difficult to track what parts of the essay need work, still what parts are polished. This is a really cool way of organizing kind of your assessment of your own work as you're writing without really slowing you down, without cluttering up the interface. So we're, as far as I know, the only kind of company that does writing that has a data structure that breaks the writing down into sentences and allows you to connect things at that level. So I found that's an extremely useful way of kind of progressing through the editing process.
- [23:35]
B
Yeah, yeah. That's another huge obstacle to writing is you have this massive document in front of you and you have to, in addition to keeping live in your mind the argument of the thing you're writing, you also have to remember where in the document you need to do the next task. It's just an incredible amount of different information to hold in your mind at the same time. Yeah, no reason. No wonder people just resort to fixing the footnotes.
- [24:02]
C
Yes, that's for sure. That's for sure. Well, and most people who end up writing really long documents like theses or books or whatever it is, you know, I've talked to lots of writers who end up printing their entire manuscript out and like laying it on the floor.
- [24:18]
B
Yes, right.
- [24:19]
C
So that they have a visual representation of the development of the arguments. And you know, our app doesn't quite do that, but it gives you a way to look at your arguments at different, different levels of resolution and a way to add kind of data points and add flags to two areas that you're going to need to come back to. So it definitely, it heavily reduces the mental load when you're working on a complex document.
- [24:43]
B
Yes. Yeah. This hearing you talk about the development of the, of the app is kind of fascinating to me because you're, you're a, you know, you're a software engineer, but you also understand what the human mind is going through while the human mind is writing. And you're taking what you understand about technology and using it to serve the human mind.
- [25:07]
C
That's the hope. That's the hope. Yeah. Well, I Think that in a way it's. I'm somewhat of a unique person in that way because I got a classical education, I went to a liberal arts college and did great books programs and really value that type of education. And then I didn't do that for quite a long time, and I was learning to code. And I've always been a pretty technical person and kind of the combination of those two things, there's just not that many people that are interested in both creative things and really technical things at the same time. And so I am fortunate enough to be interested in both those things and also really care about people's ability to express themselves and to communicate. And I think it's unbelievably valuable. And so, yeah, that's kind of why I'm doing what I'm doing. And we're also kind of in a situation where we're lucky to be able to do it because our family has been lucky enough to become connected in somewhat of the right areas so that we can market a tool like this that isn't the most marketable tool in the world. Right. This is not a tool that makes your life easier right away.
- [26:25]
B
Right.
- [26:25]
C
You know, and most. Most things that people build kind of have to do that. They have to make your life easier in some way immediately, in an obvious way that's marketable. And we were able to invest some time and money into a tool that doesn't necessarily make your life easier right away, but it's going to make your life easier in, like, in the long term. And it's hard. It's hard to build things like that. You know, that's why politicians struggle with building large infrastructure projects. Right. It makes people's lives harder at the beginning, but it's worthwhile in the future. And so we're trying to build something that is worthwhile in the future.
- [27:02]
B
Yeah. You might think, if you're listening to this, that this is a tool for teachers, for students. But Essay App is being promoted to everyone. And the idea behind it, I think, is that writing is good for human beings to do, regardless of whether they're in an academic profession or not. Why?
- [27:24]
C
Yeah, well, that's definitely how we've attempted to frame it. Certainly we're attempting to get more engaged on the educational side, just because that's where people tend to do more of their writing. But we certainly believe that everybody should be developing their ability to write. People are actually writing more than ever. Right, People? Well, maybe not as much in the last couple years with the development of ChatGPT now, when you go on LinkedIn, everything's written by ChatGPT. But up until a couple years ago, people were using writing in their careers, perhaps more than they were before, because they were posting on social media or they were writing blog posts, and they were using kind of social media and the Internet to promote themselves through their writing. So it had kind of become a more marketable skill than it had been in the past, and maybe a more broadly used skill than it had been in the past. So that was part of the reason why we were kind of marketing it to everyone, because we felt, well, if you're going to succeed in the modern world, especially considering the fogginess of our understanding what the future is going to look like right now, there's not really a more useful skill than you can develop than the ability to tell your own story and to develop your ideas and communicate them. That's. That's a foolproof skill. Right? You're going to. You're going to want to use that. And so we thought, yeah, we can market, we can market this to everyone. And I still feel that way. And in a way, the problem has become more significant because people are still using writing in that same way, but now they're taking huge shortcuts by using these language models to communicate for them, and that's just not useful for them at all. It's completely counterproductive. And although there are places that I think the language models are useful in terms of writing, when you're actually trying to tell a story that you want to have, like you want to have that story as a tool, then it's a shortcut that's not worthwhile.
- [29:16]
B
Yeah, I worry when I think about AI, I worry that people will lose the ability to connect what they are thinking to the words that they are writing. I think about this a lot in schools. When you write, there are many steps you have to go through. The first is you have to have a thought, and the second step is you have to put words to the thought, and then you have to put those words down. Either you speak them or you write them in a way that conveys the thought that you are having.
- [29:50]
C
Right. So other people can understand the thought.
- [29:51]
B
So other people can understand the thought. And choosing the exact words and the exact structure of the sentence is kind of an art. And it takes practice and kind of noticing what your thought is and then being versatile enough with the language, the English language, that you can choose exactly the right formulation to convey exactly what you're thinking. And that whole process is Very delicate and sensitive. And it seems like ChatGPT or any of these language models just come in and they just flatten that.
- [30:22]
C
Well, they just take out most of those last steps that you outlined. Right. You still have to have a thought. Yeah, probably. In order to communicate with the language model effectively, you still have to be able to generate a relevant question. And so that part is going to be the skill. If you're going to want to interact with these language models, that is maybe a more important development than any other aspect of your communication. Right. The ability to generate the right question so that you get a response that's relevant to you. But then the other aspect, using language as an art to facilitate communication, that part is definitely in danger. And that's very sad in a lot of ways. Right. Because beautiful communication poetry and the way that people who are good communicators use language is unbelievably moving to the population. People who are good with language get extremely famous because people want to spend their time listening to them. And so if people stop practicing, practicing that, we're going to lose people like that because they're just not going to spend time doing it.
- [31:32]
B
Yeah, I've messed around with chatgpt and GROK a little bit, and you learn, you discover that they kind of have a tone for sure, and it's all the same. I heard you say the other day that when a piece of writing comes in, there are some things about it that tell you that it's AI generated.
- [31:53]
C
Yeah, yeah, there are. There are a couple. I mean, and these do change as new models are released, but there are certain characters that are put in the writing, for example, that most people wouldn't use or that it's very difficult to write with a keyboard. And also just the general tone of the writing tends to be something that just doesn't exactly feel human, although you can use them. Because interacting with language models is all about the input, right? It's all about what you prompt. And so if you include voice of some kind, Right. If or if you train the language model with a certain selection of training material, that's, you know, maybe you only give it classic literature. Right. You can train these things to have voices that are different than what they are, but it tends to be that people. And this, this will be the case for the foreseeable future, I would say it tends to be that people interact with these generalist language models. Right. The things that have as much training material pumped them as possible, and so they end up sounding like they're very robotic. In some ways, and they're very general. And they also have tendencies to move towards certain ideologies or categories of thought, partly, maybe just because of the prevalence of those thoughts in the modern era and how much training material they've received in those categories, just based on the fact that there's more things in those idea categories recently than there was in the past. So, but, but they can be built, right? You can easily build a language model that has a core context that allows it to be essentially the same sort of tool as a chatgpt. And then you can add training material that makes it have a certain voice or a certain set of information that you want to have. And it's partly what makes the tools very dangerous. And it part. It's partly what makes them very exciting that you can really shape the way that they think.
- [33:46]
B
Yeah, scary. Cool and scary.
- [33:48]
C
Yeah, absolutely. Yeah. Well, I mean, powerful tools, right? Powerful tools are cool and scary. It all depends who has them and what they're doing.
- [33:57]
B
Yeah, exactly. Right. Exactly. Right. So back to the app. If I'm a teacher and I have my students write their senior thesis using the app, there are some tools within it that will help me be sure that they really did write that thing. Yes, tell me about those.
- [34:15]
C
Yeah, so these are tools that are still, still in development. And we are kind of using the fact that we have this structured environment for editing to our advantage when it comes to detecting whether or not students have actually engaged with the material themselves or whether they've relied on a language model or copy and pasted it from somewhere else, or however they're shortcutting their way to the final product. So we will kind of. We'll be able to show the teacher kind of a history of their edits in a unique way. Because when you interact with our structured tools, if you're creating iterations in the rewrite tool, if you're using our outline tool in certain ways, then we have kind of an understanding of the way that the document came together and the way that the edits came together in relatively interesting way. And we can even show the teacher those edits. Right? We can show them all the iterations that the student did and when. And so that can be kind of a great way of sort of a non technical or a non fancy technical solution to showing a teacher that the student did the work. Right. So it'd be rewarding the student for the process that they. That they underwent in a way that most writing platforms can't really do. And then the other half of it would Be kind of what we discussed earlier, which was, you know, using the sorts of, like, identifying characters that AIs tend to produce. And they will actually tell you sometimes. Right. The tech companies don't really want people to be handing in essays that are generated by ChatGPT or Gemini. That's not why they're building them. That's just an unfortunate outcome. So they'll actually insert characters that make it obvious that it comes from an AI system, and they change that sometimes with different models. And so you kind of have to stay on top of it and stay on the cutting edge of the tech in order to detect it. And that's partly why we aren't going to rely entirely on actual detection of the content. We'll rely more on the structure and the process that the student underwent, as well as simple things like how much did they copy and paste into the article or into the essay. And then we'll be able to kind of give a summary of that information to the teacher.
- [36:24]
B
Yeah, I saw yesterday that you're also building in a way for the writer to track his or her own progress toward goals. You know, so, you know, maybe I set a goal for myself. I'm going to spend 30 minutes every day writing on that. Writing, working on this big essay that I'm working on. And I can look at the app and see every day this week I spent 30 minutes. Good for me.
- [36:48]
C
Yeah.
- [36:48]
B
You know, and that's a really valuable thing for teachers to see too, because of course, the curse of. The curse of the literature teacher or the composition teacher is you have to spend hours upon hours reading essays the students wrote the night before.
- [37:02]
C
Yeah, I know. How painful is that?
- [37:05]
B
And so you could make a requirement for your class that I will not read the essay unless the app tells me that you have made consistent progress on it for multiple days in a row.
- [37:17]
C
Yeah, yeah, that'd be a great requirement. Yeah, that'd be a great requirement, people. I think teachers tend to try to enforce that through multiple submission models or something like that. Right. Like, give me an outline, give me a draft. And then. But then, of course, you have to look at all those things.
- [37:31]
B
Yeah.
- [37:32]
C
And that becomes. Yeah. Even more arduous on the teacher. So hopefully. Hopefully. Yeah, With. With that kind of, I guess, dual. Dual value tool. Right. So we. We both create motivation through gamifying writing a little bit. And so you can have a daily writing goal. And then we have a project planning feature that I think you alluded to when you introduced that, which is that you can say, well, I Have this essay due. It's due on September 20th and it's a 2,000 word essay. And we'll even potentially have the topic of the essay and that sort of thing. And then we can help you create a writing schedule. We can say, well, if you write 200 words a day for or 250 for the next 10 days, then you'll have, you know, 25% more than you need. And then you can spend the next four or five days editing and then you're in. Then you've been in good shape. And we can lay that out on a calendar for the student. And we could, if they want, even send them fun emails that say, hey, good job. You've done three days in a row.
- [38:30]
B
Positive feedback.
- [38:31]
C
Yeah, exactly.
- [38:32]
B
Oh, that's great. Well, if people are interested, Julian, in learning more about Essay App, how do they find out about it?
- [38:39]
C
Well, our website is sa app, so that makes it pretty straightforward. And we have a page that has quite a bit of content. We also have a support center that's accessible through the landing page that has a bunch of resources, written resources about the tools, why we built them the way we did, how to use them, how to use them in the context of a classroom. And then you can also reach out to us. We have kind of a live chat on the website and that will connect you directly to our core team. One of the things we're doing as much as we can, especially early on in our development of this app, is really talking to the people that want to teach this way and want to write this way and making sure we're building what they need. And so we definitely want to hear from people.
- [39:20]
B
Yeah, a much needed tool, I think. Thanks for joining me, Julian.
- [39:24]
C
Oh, my pleasure. Thanks for having me.
- [39:26]
A
Thanks to Dr. Kathleen O', Toole, Associate Vice President for K12 Education at Hillsdale College, and Julian Peterson, founder of Essay App. You can find more at Essay App. I'm Scott Bertram. We invite you to like us on Facebook. Search for Hillsdale College K12 classical education. You also can follow us on Instagram hillsdalek12. That's hillsdalek12 on Instagram. Thank you for listening to The Hillsdale College K12 classical education podcast, part of the Hillsdale College Podcast Network. More at Podcast Hillsdale. Edu or wherever you get your audio.