Subscribe
Share
Share
Embed
Each week Adam & Ste focus on a specific feature, idea or prototype in Hardcover and iterate on it together or with guests.
Adam (00:01.007)
Hey, hey, stay, how's it going?
Ste (00:02.988)
I had a pretty good, pretty good coming back from my vacation and yeah, feeling pretty energized.
Adam (00:13.159)
Yeah. So how was, South, South France, South of France.
Ste (00:16.754)
Yeah, yeah, I was in Cannes and it was like really nice. But the trip there was pretty eventful. I don't know what's going on, but yeah, my life is like, stuff is happening because the train we took there basically broke down. I think it was because vandalism and they tried to like put it back, make it work three times, but eventually we just waited.
for another train to pick us off the track. And my five -hour long journey turned into a 12 -hour long journey with all of us waiting in the field besides the train because it literally stopped in the middle of the field and there was no water, no food in the train. It didn't even have a restaurant card or anything. So yeah, getting there was quite an adventure. But once we got there, you know.
Adam (01:03.621)
Wow.
Ste (01:12.18)
nice sunny beach, yeah, took my kids swimming, so he was very happy, we were all very happy, so turned out fine. But yeah, pretty eventful.
Adam (01:23.733)
Man, 12 hour, 12 hour travel trip with a small child. And yeah, that, that, that sounds, that sounds like a very stressful trip, at least the beginning. And then hopefully relaxing after you got there.
Ste (01:38.666)
Yeah, yeah, no, it was, mean, the food is good. know, everything is predictable once you get there. And even the trip, it was like, I got some nice Polaroids of the whole thing. His mom holding him next to the train and that kind of stuff. yeah, it was pretty fun in the end. And we were all stuck with water and everything because she really liked it. Yeah.
packed everything up. So yeah, we were very prepared. And on the upside, we can take a trip anywhere on the globe by plane right now, and we know it's going to be OK since 12 hours by train. It was pretty hot. yeah, so we kind of made it. Yeah.
Adam (02:33.231)
That's true. You accidentally learned some new information about that.
Ste (02:39.114)
Yeah, exactly. What about you? How has your week been?
Adam (02:43.81)
pretty quiet. Like we had a barbecue with some friends for labor day on Sunday. And then this week's been just pretty quiet. Mostly just been working on hardcover stuff and, yeah, just, doing some, like I've been reading through this like machine learning book for fun.
And I tried something different instead of like, have the physical copy of the book. but I, like earlier this year, I got an Apple vision pro that I haven't been using quite as much, but, it's really good for when you want to like really focus on something because you can't really not focus on it. So, so I started, the book I have has a, it's a O 'Reilly book. So it has like an online, version of the book available from O 'Reilly and it uses a lot of, Google.
Ste (03:27.604)
Yeah.
Ste (03:37.413)
Okay.
Adam (03:40.401)
CoLab, which is like a Jupiter notebook kind of thing where it's like an interactive way of exploring the code. So I, I've been like, just, pulling it up with Apple vision pro on and then dialing out my environment. So I can't even see my walls in my room. So all it is, is like the book here and the Google CoLab here. And I'm just like using my hands and just reading and going back between these two things. Cause it seems easier than going between like a book and a computer.
Ste (03:44.989)
Mm, okay.
Ste (04:01.462)
Wow.
Ha
Adam (04:10.555)
to have just between two screens.
Ste (04:10.752)
Yeah, I bet. I mean, that sounds amazing. I always wanted to ask if you use the Apple Vision Pro because I saw a lot of people not using it, and this seems like the best use case, tapping out of everything and just focusing on one thing. Nice.
Adam (04:27.611)
Yeah.
Adam (04:32.569)
Yeah, I haven't found a good use for it when it comes to like creating any kind of thing. It's only been so much for consumption. Like for creating something, I feel like I need a keyboard and using a keyboard and like when you're in a virtual reality and you're looking down with goggles on, it doesn't work too well. It might be better if you're like a hundred percent touch typer or have like a
Ste (04:39.284)
Yeah.
Ste (04:53.375)
Yeah.
Ste (04:58.878)
Mm.
Adam (04:59.075)
moon lander keyboard or something that you like really know the ins and outs of. But yeah, for me, it's been much more for consumption and gaming or something like that than any kind of creation.
Ste (05:10.91)
Yeah, well, it seems about right. mean, those are like really two use cases I like see it being used for. Nice. Yeah, yeah, that's really nice. Do you have like the custom backgrounds that you put in behind the like what you're focusing on?
Adam (05:19.471)
Yeah. A lot of video.
Adam (05:30.915)
Yeah. And there are some neat ones. They're like, like ones that it's like a mountain you're like by a lake, by a mountain. And then you can like, it'll, it'll sometimes start like raining a little bit. So you hear like little bits of rain in the, around you. And so those are, those are pretty relaxing.
Ste (05:44.899)
wow.
Ste (05:49.972)
Yeah, that seems very neat. Yeah, nice. I think it's out in Europe. mean, maybe I'll try one out at the Apple Store. Nice. Yeah. Yeah, very nice. Yeah.
Adam (05:58.909)
It's fun, yeah. But yeah, yeah. Other than that, yeah, this week we deployed a new update to where images are at for hardcover, which was a long time coming, but it was really great to finally get that in.
Ste (06:18.846)
Yeah, I read all the avatars for when you first create the hardcover account, get to a peak between six avatars, and now I color -coded them so they reach with their own color. I'll show them in the Discord so everyone can see them because you probably have an account, maybe with a profile picture. Maybe we could put that in settings for whoever wants to try one out. I see lots of people using the jewels avatars instead of uploading their...
Adam (06:23.654)
Hmm.
Ste (06:46.818)
Like own avatars, so that's like nice. I guess because I guess people like them
Adam (06:53.263)
At least they're, yeah. And well, for these, I'm a bit worried about changing them, because then everyone who's already set the existing avatars might not want to change to the new ones.
Ste (07:09.098)
really? OK, yeah, that's a thing I haven't thought about. Yeah, maybe we'll maybe announce the change and see if people want to go for it. I mean, they're similar. Yeah.
Adam (07:18.127)
Yeah, but looking at these, yeah, they're the same. They're pretty much the same. It's like the difference between the crown one, it's like has a bit of a yellow tint in the background, but it's.
Ste (07:28.234)
Yeah, I kept the same characters instead of the cap one. I put the cap one. There was one with a kind of like Spanish hairdo which no one seems to have been using, and I changed that. put the cap on that one, if anyone's interested about Jules talk. Yeah.
Adam (07:45.869)
yeah. I was doing some analysis on the database and yeah, profile image six was the one that the fewest people have chosen.
Ste (07:55.986)
Okay, there we go. So I put the cap on that one and I put a little like rounded bunny hat on the second, which was the cap one. Kind of look like a bro, so I wanted to like de -bro -ify it. So yeah, now it's de -bro -ified.
Adam (08:12.83)
And I think like, yeah, one of the things that I was figuring out with these images is like when, someone is going through onboarding and they pick out an avatar and we add it to the database. Like over time, there have been different ways that I've handled that interaction. Sometimes it creates, it uploads a new avatar for that user and associates.
it with that. we're like duplicating the avatar in our, backend every time. And I don't mind that much, the duplication part, but, then on the front end, if you're looking at like a, a book page and you see like 50 avatars and there, those could be 50 different avatars or that those could be 50 of the same avatar. And we're like, Fetching them from different URLs. So currently the way it happens right now is a
Ste (09:04.534)
Yeah.
Adam (09:11.002)
if someone uploads, if someone chooses one of those avatars, it's locked in to like our version of that avatar. So if we change it, it changes for everyone going forward. But.
Ste (09:24.786)
Okay, that makes more sense. Yeah, so basically, yeah, it's like we're not refetching it for every Avatar. Most of them have like the first jewels, like the official jewels, jewels, Avatars. So I've seen those on the book page. Sometimes they're like a lot, a lot of jewels, Avatars in there. Yeah.
Adam (09:41.509)
Yeah. Yeah, these will be neat with the different colors. I think it'll make it more colorful.
Ste (09:48.436)
Yeah, yeah, plus, I mean, one of the feedback we had was long -term people want to customize their colors, and we've added new colors to the book buttons. We're changing it to use color variables, so this is a step towards at some point being able to choose the colors that you want. Still have a lot of stuff in between, which is way more functional, but yeah.
Adam (10:17.519)
You
Ste (10:19.09)
It's like we're setting up for that just so you know.
Adam (10:24.739)
Yeah, it's kind of like signs of a good code base are that it's easy to make site -wide color changes, and we are not there at the moment.
Ste (10:31.712)
Yeah. Yeah, we will be. We will be. Yeah, there's a lot of code. you know, once we open source, maybe that could be like a very cool, predictable community project because all you'd have to do is change everything to a color variable. So, yeah, I'm hopeful that one of the contributors maybe will pick up on that.
Adam (10:54.863)
Yeah, that could be a one.
Ste (10:56.546)
Yeah, yeah, and today we're basically like you send the September report and in the report you sent out the survey that people had to fill in about stats and we got 25 submissions so far to what are three pretty like challenging I guess open questions because you know they really make you think about what you want from your reading and that's like for
I think everyone who replied, and thank you for the replies, everyone, a very...
Ste (11:36.018)
a question that requires a lot of effort, I guess, to figure out what you really want from your reading because there could be so much stuff. We got a lot of really interesting replies. And yeah, we have two different approaches. And maybe today we can talk about that feedback and how we're thinking of prioritizing things.
Adam (11:59.471)
Yeah. Yeah. So, stay and I both did like an analysis on, I only did the first question and I did like a manual analysis of it, kind of the way I've always done my analysis with like a spreadsheet, finding themes manually, like using my brain to pull it out. And you, went the chat GPT route to analyze the
Ste (12:18.167)
Yeah.
Ste (12:22.242)
Yeah, I figured, mean, you're doing that route, I might as well do the other route. So I downloaded all the results as a CSV, and I uploaded them onto a GPT and asked it to basically source insights from that. And I think it's going to be cool to compare your insights to what...
Chai GPT told us and see what are the common points and what are the points where we can develop on those insights. I think that would be pretty cool.
We had the mind map, so to give people a little bit of context, because this is that episode number five, this is like, since this feature is central to hardcover, we're really taking the time to make sure we're shipping something that's really meaningful. And each time, guess, we jump to answer the question, what stats do you want? We get answers that are more and more
distilled, guess, and more to the point of the question we're trying to solve here is what's the most useful thing we can give you, not just throw a lot of stats, which is what everyone is doing. Of course, we already have a lot of stats going.
They will be there, but we're interested in prioritizing those which are most helpful and maybe doing stuff a bit differently than everyone and seeing how we can make this work.
Adam (14:09.913)
Yeah. One of, one of the things I think we, did with our existing stats page is that we kind of designed it based on the data we had and what the libraries we were using were capable of. And so those were kind of our, two limitations. And now we're kind of like just blue sky and like, what do readers want the most? Let's build that. And then we'll figure out how to get the data and the tools to build it after the fact. But we're, letting the, what readers want.
kind of lead the charge on what we build. So yeah.
Ste (14:42.454)
Yeah, yeah. Plus this is like, yeah, it's an important supporter feature as well. So it's basically one of the central things that you create a supporter plan for, like more insightful stats. And I guess our promise is that we are trying to offer you the most insight into your reading. Hardcover is made for people who like have a reading habit that they track. So we're trying to...
not only show the usual stuff, but maybe stuff which is a bit deeper. So, yeah, maybe how do you want to start? Do you want to first talk about what's the first question you reviewed? And then maybe we can compare with JGPD.
Adam (15:29.903)
Yeah, well, yeah, I could probably do mine first. That's a good idea. Cause mine is gonna be a lot shorter, I imagine. It's mostly just like the individual top themes that were in it. So let me share my screen here for a sec.
Ste (15:42.795)
Nice.
Yeah, let's go.
Adam (15:53.502)
sure. Probably gonna have to make this a little bigger here.
Adam (16:02.645)
Okay, so, see how we had our initial survey results and I kind of just looked at this first question, what stats are you most interested in when it comes to your reading? And I kind of just pulled out individual themes with like a little more info for each one. And based on that, it ended up with,
a couple different top categories that had the most votes. So the three that had the most votes were like pages read. So understanding that, and when you think about pages read, there's pages read per month, per day, week. So a lot of people wanted it by like lots of different timeframes, not just by
Ste (16:57.991)
Mmm, yeah.
Adam (17:00.207)
that so I'm imagining like, you know, you're looking at your stats of all time and you might see a page is read over time over like the lifespan of your account, but you might also want to know like how many pages you have read on average every year and how that's changed. So it's like, it's almost like two different or multiple charts or multiple stats just from that.
Ste (17:21.078)
Yeah.
Ste (17:24.95)
I'm very happy to see pages read by day in there. I think some people mentioned it and yeah, that's gonna be pretty interesting.
Adam (17:34.063)
Yeah, and second one was tied between genres read and books read. I feel like a lot of people might assume books read is just like a given, so they didn't even mention it. And then books read could be.
Ste (17:43.787)
Yeah.
Adam (17:51.663)
like per year, per month, year to date, per quarter. But yeah, a lot of by genre, by month. yeah, books read in different timeframes and the genres read by like your top genres you've read. Some people wanted more sub genres like historical romance, historical mystery.
Ste (18:01.505)
Mm -hmm.
Ste (18:22.338)
Okay.
Adam (18:23.003)
Yeah, those, those are kind of like, I feel like those three were, were, and then, top authors were the top four and top authors, including,
Adam (18:34.915)
most read and reread authors.
Ste (18:38.786)
Okay, this is gonna be interesting, yeah.
Adam (18:40.931)
And there were a number of author demographic ones which are kind of adjacent to that, like wanting ethnicity, gender identity, race, kind of all of that information. That was probably the fourth highest and it ties in with the top authors as well.
Ste (19:03.874)
That's very interesting. Plus it opens up, mean, just the fact that people mentioned books read, even though, you know, it was clear that we were gonna like have books read because we already do. But I'm thinking if this is so important, there are like a couple of different ways we can show books read from like the most number oriented ways. Like you've throughout your like reading life, you've read like 100.
200, 1000 books. In March you've read five, in April you've read 10. And there are also like other ways, maybe we show a timeline, we show covers, we show like what type of covers do you actually go for, what type of...
know, genres associated with like the books, the actual books you've read are on that timeline. there are like some different interesting things we can do here, not just like show the actual books, which we have in the stats. We have like a beautiful visualization of the covers and the number.
Adam (20:13.349)
Yeah.
Adam (20:21.145)
One of the things I was curious or interested in is no one mentioned moods at all. Like, don't think anyone, like, I don't think that even came up. Yeah, it didn't come up in any of the questions. So it makes, yeah. So it's like, people really are curious about genres, but they're much less interested in moods.
Ste (20:28.662)
Yeah.
Ste (20:35.49)
That's such an insight, yeah.
Ste (20:42.176)
Yeah, I'm wondering why that is. It's also a bit striking because so many online reading platforms are focused on moods basically and how people feel throughout what they're reading, but we're not really seeing that. I'm wondering if it's like, yeah, what is it exactly? What's the reason why people didn't mention moods?
Adam (21:07.206)
Yeah.
Adam (21:10.917)
Yeah. One, one other interesting takeaway was a number of people mentioned wanting more like comparisons with other readers and stats. So that one came up three times with some, some good, good recommendations on like, want to see either people with like reading tastes similar to me, people with stats pages, similar to me. And this actually like just came up on the letterbox subreddit yesterday because, letterbox
Ste (21:19.44)
Yeah.
Adam (21:40.217)
released a new feature where you can search for, a person who has, so there's a concept on your profile and letterbox of your top movies. So you have four top movies and those are kind of like hard coded. Like those are, those are your top movies. And so you can, you can search for users who have specific books in their top. So everyone was on the, on the subreddit was like, no one has all four of the same top books as me.
Ste (21:57.601)
Yeah.
Adam (22:09.745)
But maybe like two people have three of the same and maybe like, you know, 20 people have two of the same. And so people were using that to find like new people to follow, which was, which was an interesting use of that.
Ste (22:21.194)
Yeah, yeah, that's so interesting. And even with top four books, that's very rare, I mean, my instinct would have been to say that, of course, you're going to find a lot of people with the same four. But those combos, think mathematically, it's actually very unlikely.
Adam (22:48.305)
Yeah.
Yeah. Unless you have like four super popular movies or four that are like maybe by the same director or four that are like just really, really similar. But yeah, I like, there was no one on letterbox who has three of the same ones as me. And there were only people that had two and, but looking at those people, I was looking at the other two books and their top fours to get like recommendations on things that I should watch, which was kind of fun.
Ste (22:57.772)
Yeah.
Ste (23:05.612)
Wow.
Ste (23:15.382)
Yeah, nice. How did that turn out? mean, did you find anything interesting just by doing that or something that you didn't know about?
Adam (23:25.161)
I added like one or two books I never, never, or one or two movies I hadn't heard of. I think more interesting was just like how different the other two movies were compared to the other ones on my list. Like, you know, one of, like one of mine's an anime, one of them's a, my top four were like Nausicaa, The Valley of the Wind, RRR, Lost in Translation, and Contact. And,
Ste (23:39.852)
Yeah.
Adam (23:51.759)
Like I would see people with like contact and like hell raiser or something like that. It's like, that wouldn't have been anywhere near my top four, but good for you. Yeah. That bringing in that social aspect somehow on our stats, I think is going to be, is going to make it more, more exciting. I think as well.
Ste (23:55.274)
Wow, nice. That's very interesting. Yeah.
Ste (24:07.083)
Yeah.
Ste (24:14.174)
Mm -hmm. Yeah, that's definitely something we should look into. Plus, this is such a simple stat. I we should definitely begin, at least, with the simpler stats, and then maybe complicated. But my feeling was reading through the results that people really wanted the core simple things the most, and then the more complex ones. But yeah, the simple ones, like...
Totally like one.
Adam (24:47.087)
Yeah. And I think also it's like, this question is very much like having people figure out what they can envision. Well, the last question in the survey is like their question on to us. So I haven't looked at that third question, which I think is going to be the more of the place for us to figure out solutions for people that they don't know what the answer to it is. So yeah.
Ste (24:58.603)
Yeah.
Ste (25:03.618)
Yeah.
Ste (25:12.982)
Yeah.
Adam (25:15.141)
I'm curious to look at the chat GPT results of this too, and, and, and see how that compares. Here I'll stop sharing.
Ste (25:17.026)
Let's do that. Okay. Well, let me share my screen. Okay. Screen.
Ste (25:29.164)
feedback insights. Okay, here we go. Can you see this? Okay, so basically my first question to it was, here's a list of submissions for our stats, can you look through them in detail and give me takeaways from the feedback people left? it basically, like, I made it somehow show what it's doing in the background. I don't know how I did this, but it's basically using Python to basically process everything.
Adam (25:33.387)
Yes.
Ste (25:56.23)
So yeah, it told me what the whole survey was about. then basically, the key takeaways for the first question were that users are particularly interested in checking the number of books read per year and pages read per month, which is kind of like, yeah, OK, so it got that right.
Adam (26:19.441)
Those were the top two I found.
Ste (26:24.61)
Then the strong interest in general related stats like top genres and how they align with reading goals. This was interesting like genres and goals. Pretty, you know, surprising association and also says so.
Adam (26:43.503)
Yeah, I saw a lot of genres, but I didn't see any like, didn't see remember much in the way of like, goals. So that's that's curious.
Ste (26:52.168)
Yeah, like this one. Yeah, it sounded like pretty interesting to me as well. And then says some users want more insight into their reading speed and author related stats, which kind of like aligns with like the top themes, including the most read or reread authors. So this was for the first question.
Adam (27:13.967)
Yeah.
I'd say those, those three match up pretty evenly with what I also found. I think it. Like each one of those has so many like alternative things, like number of books read per year and pages read per month are like, that's like one answer to the question, but there are so many other answers on like pages read per day year. But overall, I think those three things of.
Ste (27:35.724)
Yeah.
Adam (27:44.023)
books, pages, genres, and authors, match what I found too.
Ste (27:50.388)
Okay, so it's kind of like getting it mostly right, so that's good. So we can trust it for the next two, I guess. Let's see if... So basically now we'll do it the other way around. We'll see if your analysis... I'll do mine as well. More like quality analysis, I guess. This is like...
more of a broad analysis, more like brute force what it's doing. So for the stats, you'd be most excited to share with friends. Many, according to GPT said, want a year -end summary, kind of like Spotify wrapped, showcasing their reading journey, which is kind of like what we knew already, because many people have been saying, I really want to Spotify wrapped for books.
Then they said they also would like to share book counts, totals by month and year. And a few mentioned that, yeah, they don't have any reading friends, but they still appreciate stats that highlight achievements or unique aspects in their reading habit. So these unique aspects, maybe looking at the second question, we can see if people actually mentioned what unique aspects.
Adam (29:01.435)
Mmm.
Ste (29:17.16)
they want it because this is a really interesting one, like achievements or unique aspects.
Adam (29:20.815)
Yeah.
Adam (29:24.677)
Yeah, the word unique is not mentioned in the survey. So I'm curious how it pulled that out. I'm looking forward to looking over the answers to that question with that in mind on what unique means.
Ste (29:24.883)
and
Ste (29:35.464)
Yeah, yeah, exactly. Yeah, how did that analyze that unique? Yeah, I'm really curious as well. And then for the understanding, reading habits one where we ask people like, what's like a thing that's, yeah.
Adam (29:49.349)
What aspect of your reading habits do you wish you understood better?
Ste (29:53.938)
Exactly, it said users are curious about trends in their reading habits, especially what might lead to reading slumps or how their genre preference is evolved. This is kind of like a, yeah.
Adam (30:05.851)
That's, yeah, that's a neat one. Like I can understand that because like, imagine looking at a book on hardcover and being able to say like, this book is going to put you in a reading slip.
Ste (30:18.146)
God. Yeah, exactly. Wow. Yeah. Yeah, that would be like a really like sure way to like trigger, not trigger warning or what is it? Like it's some kind of warning. Yeah, exactly. Yeah, this is going to be interesting. And how their genre preferences evolve. This is another like interesting one that...
Adam (30:29.329)
Like, yeah, yeah, it's like caution. Yeah.
Ste (30:47.072)
I'm guessing it gives us a lot of space to generate visualizations based on this, like the word evolve. There's lots of stuff I'm guessing we could do with showing that visually.
Adam (30:53.371)
Hmm.
Adam (31:04.111)
Yeah, that sounds a lot like our area bump chart we use for genres over time. But I wonder if that's the answer to it, and we just happened upon it by accident, or if there's another genre preferences over time kind of visualization we could do.
Ste (31:10.102)
Yeah.
Ste (31:23.094)
I think that's pretty accurate to, know, general preferences evolving, I guess it's like through a bigger like span of time. You know, you won't be interested in seeing how your general preferences evolved in the month of like August, because they probably like didn't evolve at all. But like through a year or maybe through several years, that's going to be like interesting. Like we do have like the bump chart now. So yeah, I think we kind of like nailed that one.
Adam (31:49.67)
Yeah.
Adam (31:53.809)
Yeah. And we could, we could also do it as a, like a bump chart by year. Like right now it's, we do an area bump chart over 12 months, but if we did like a, a regular bump chart, which just has positionals, but it doesn't have a height for the number of them. We could do something like that for your genres. So you could see like over your last 10 years, like where
Ste (32:04.169)
yeah.
Adam (32:21.017)
a single genre has ranked amongst all of your genres.
Ste (32:24.286)
No, so basically this would be like a line that you'd see. So you'd see like the ears on the bottom axis maybe and then you'd just see a line of, you know, this is your sci -fi.
Adam (32:36.837)
Yeah, would be like this. I can share this with you if you want an example.
Ste (32:42.752)
Yeah.
Ste (32:46.72)
Let's see. Okay. yeah, exactly. Yeah, that's what I was thinking. So Adam shared like a bump chart. That's basically exactly that. So it's how your genres evolved over time. Yeah, I'm guessing that would be a good one. And it also said that they want to understand the correlation between genres and themes and how these factors influence their overall enjoyment or book choices.
I'm curious to see how they worded that because that's a very specific and complex, I guess, stat.
Adam (33:27.823)
Yeah. Like that to me reminds me of the, the correlation matrix chart we have where it's like genres. And then it has like the rating and like moods on the other side with the rating, except they don't mention it by mood. say theme. So I, so I'm gonna have to, when I'm looking at that question, I'm going to try to dig more into what, what theme they mean. If that's like tropes from a book, like a
Ste (33:40.769)
Mmm, yeah.
Ste (33:45.236)
Yeah.
Ste (33:55.212)
Yeah, maybe tropes, yeah. It kind of sounds like tropes. I'm curious as well because they didn't mention moods, but if they mentioned themes, mean, themes would be like maybe even more powerful than moods. And we could get that from the tags, I'm guessing. So that's a really interesting one.
Adam (33:58.768)
Like, yeah.
Ste (34:16.156)
And then it said the fuses are seeking insights into how the reading habits compare to others. OK, so again, like the comparison to other readers and what triggers certain behaviors like slumps. So again, maybe like outlining some kind of book succession that's really impacted your reading or the other way around, seeing
Adam (34:30.511)
Hmm.
Ste (34:45.898)
maybe it's a busy period in your life. Maybe you're doing something that's very intensive and seeing what books you've read or what books have gotten you through that period. think, yeah, maybe.
Adam (34:58.021)
Yeah. It's, it, it makes me think how neat it would be to have almost like, like the pages read over time kind of chart, but then have almost like, I'm imagining some kind of indicator on it that says like, when you see that dip there, that's when you read this book. And then it, like, it, it, goes up when you read this other book. So we could almost like annotate it with saying like,
this book might have put you into a reading slump and this book might have pulled you out of it.
Ste (35:29.694)
Mm -hmm. Yeah, I love that. I love that. mean, this kind of narrative -oriented things, I think people would really be interested in. And it's an interpretation that we offer, which you normally have to find yourself if we wouldn't. Even if we show you some graphs.
Adam (35:36.077)
Thank
Ste (35:57.1)
You know, it would be neat to make it like really easy through annotations.
Adam (36:02.449)
Yeah.
Ste (36:04.69)
Okay. And then I asked what's an unexpected takeaway that you can extrapolate from all the entries. And it said the desire for understanding the psychological or emotional triggers behind reading habits, particularly your slump. So I think some people like really mentioned like slumps a lot. So I guess the insight here would be like, people want to use book tracking to maybe get out of the reading slump. And this is where we could have
Adam (36:10.417)
you
Ste (36:34.868)
a pretty significant impact. mean, if through hardcover they could come in, track their library, and we could get them out of the reading slump by showing them other readers that are similar and show them, let them find new books. I think, yeah, that could be like a way to solve this thing. And...
Adam (37:00.347)
Yeah.
Ste (37:03.062)
Then it said like several responses to just a deeper interest in why they read what they do and how external factors like mood or life events might influence their reading patterns. This one's really like, it goes deep.
Adam (37:18.905)
Yeah. It's like, why, why, why do you read what you do read? Yeah. That, that would be a, I wonder how much we could figure out that question and how much of that's going to be left to the reader.
Ste (37:24.96)
Yeah.
Ste (37:31.88)
Yeah, that seems like a thing that...
Adam (37:36.997)
I guess if we can't answer it, then that would be giving someone an insight into themselves that they might not even know.
Ste (37:43.508)
Yeah, yeah, exactly. mean, this is on a different level than just looking at the chart. So this is your reading pattern and how it influences or how your personality influences your reading pattern or how your life events influence your reading patterns. This seems that it has a lot of potentials, but I'm not sure.
what form it could take right now. So it seems like there's lots of stuff we can do here, but I wouldn't be sure how to show a stat that actually answers that.
Adam (38:25.329)
There's a, there's a neat tool that I've, I've used before called a exist .io. It's, it's this like a tracking system where you kind of like connect it to all your different services in your life. Like you can, you can connect it to your, like Apple health to get your number of steps and your amount you, you sleep every night. can, track.
like productivity by GitHub commits, like Todoist checked off items or things like that. Number of calendar events. can do other things like inputs, all that, as well as like heart rate, temperature of your location that you're in. And then it has mood tracking. So you can track like your mood on any given day. And it uses all this to find correlations between these data points. So it's like,
when you sleep X amount of time, then you rank your day happier. when you have fewer meetings, you're happier. So it's like, yeah, it's just like finding connections between that. I could almost see what we could do here as doing something similar, where it's doing that more for the number of pages you've read than for how much you've rated a book.
because you can always sort by your top rated books and just see the books you like the most. But if we understood how does the shape of fantasy books that you've read compare to your pages read, and that would indicate whether or not fantasy is something that causes a reading slump or not.
Ste (40:13.578)
Yeah, this is really good. huh.
Adam (40:15.428)
Yeah, it'll be an exciting thing to mess around with.
Ste (40:19.744)
Yeah, it's like, yeah, we could like find out all sorts of things. I'm always like, when it comes to book moods, I'm always like that question, you know, when you get to review a book and it asks you what moods does this book have? That kind of gets me so confused sometimes because like, what should I rate? It seems like I'm like categorizing the book, but in fact, I think I should be like...
categorizing my mood reading that book. So I don't know like where to approach it from and I'm guessing, know, maybe there's it's related to to this. I think people are interested in like their like feelings and moods not like the book moods themselves. So like I could read a very happy book and be angry or I could read a very like a word book and be really happy. So
Adam (40:52.635)
Hmm
Adam (41:09.528)
Mm, yeah, good point.
Ste (41:18.37)
How would I rate the book mood and who am I doing service to? Am I contributing to other people who... Or if some other people categorize a war book as being whatever, don't know, let's say intense. Maybe it could be not intense for me, but they categorized it because that was their mood.
So yeah, maybe something that's like personal, we could get out of this.
Adam (41:50.054)
Yeah.
Adam (41:55.875)
Yeah.
Ste (41:57.608)
Okay, so then I asked if you made a list of stats based on this feedback and prioritized the ones mentioned the most, how would that look like? So we analyzed these and basically took the...
Ste (42:13.026)
I guess keywords of all the feedback we got from readers. it's a books mentioned 59 times, genres mentioned 17 times, pages mentioned 16 times, authors, then time, averages, totals, like total reading stats, and then reading speed over time.
I thought this would be more higher up, but yeah, this was only mentioned one time. So I'm guessing this aligns with the priority from question one. It seems like books, genres, pages, authors is exactly or pretty much accurate.
Adam (42:58.671)
Yeah. Yeah. Yeah. Those the same top four that I got. Yeah.
Ste (43:05.374)
Yeah, okay, this is interesting. And then I asked it what other way we can approach stats based on the feedback we receive just to see like what other things are out there. And again, some pretty interesting answers. So it's first said contextual stats or comparison with others, offer stats that compare users' reading habits with the broader community trends.
which is what we did in some of our prototypes, like you're in the top six readers in this period for like fantasy or something like that.
Adam (43:46.224)
Mm
Ste (43:47.21)
And then it says personal benchmarks, checking whether they're on pace to meet the yearly reading target. Then I'm just going to go fast through this because there are a lot of insights. So the second one is behavioral insights, like reading slump analysis, and then genre exploration, which is kind of like what we talked about.
Adam (44:01.243)
You
Adam (44:10.329)
Mm -hmm. Yeah, got a genre exploration for suggesting new genres based on reading patterns or show how often users explore outside their typical genres. That sounds neat. Yeah, it's like.
Ste (44:23.658)
Yeah, and the outcome of those explorations, this also like very, very nice thing we could.
Adam (44:30.536)
Yeah. Yeah. Cause I'm imagining like that, that reading slump kind of thing where it's like, maybe the reason you entered a reading slump might've been because you read a book that's in a genre you're not used to and you, kind of threw you off in some way.
Ste (44:47.988)
Yeah, this would be pretty neat to see how we can show. And I think I like this the most, narrative -driven stats, like reading journey visualization shows the user's preference, reading pace, and how it evolves over time. This is kind of like a meta -stat, I guess, of all the things, like how many books, how many pages, what speed.
Adam (45:04.847)
Hmm
Ste (45:15.35)
They read the genres, how they correlate the reading pace and how it evolved. Pretty nice. Then impactful books, books that have significantly influenced your reading habits. So you read a book and then maybe you read something by an author and then you read everything by that author. And we highlight it as the period where you read that author because, you know,
Adam (45:22.619)
Yeah.
Yeah.
Ste (45:44.776)
one of their books like triggered something.
Adam (45:47.611)
feel like that's kind of like what books got you out of a reading slump in a way. It's like books that you've read and then after that you've read a lot more.
Ste (45:53.825)
Yeah.
Ste (45:57.482)
Yeah, yeah. And then they mentioned genre shift here. So we could also correlate that to, or if you actually read the genre you weren't really used to or read before. Yeah, pretty neat. And then it said social and sharing focus stats. So social influence metrics, like how many people have read books on
Adam (46:04.369)
Hmm.
Adam (46:14.309)
Mm
Ste (46:26.742)
based on your recommendation and year -end summaries. This is kind of like, it aligns with what we talked about and what we're planning, I guess, right?
Adam (46:33.883)
You
and
Ste (46:39.514)
and psychological and emotional metrics like mood and genre correlation. think this is like what we already did with the bump chart and the motivational triggers. So.
Ste (46:53.824)
Basically, this relates to reading habits as well. And then it's funny because it said customizable dashboard. So personal staff tracking and dynamic reports, which is like exactly what we were planning to do with the dashboard to basically like get some of the core stats and show on your dashboard.
Adam (47:14.827)
Yeah, that's pretty cool. Does anyone mention dashboard in the survey? Let's see. No, no one mentions dashboard in the survey. So it was able to generate dashboard based on other information it had, which is fascinating.
Ste (47:16.629)
Yeah.
Ste (47:32.426)
Yeah, I mean, it's good that we kind of like think at a higher level. We can, I guess it's time we can like, we've been like hearing so much like feedback from readers these past years that we kind of like have all that data stored. So I guess our personal like brain LLMs got to this conclusion independently.
Adam (47:53.551)
if
Ste (48:01.955)
Yeah, and then I asked it what was the one thing readers understood less based on their responses and it said again reading slumps. I'm really curious like if people mentioned like reading slumps that much. So...
Adam (48:17.877)
It was only mentioned twice in the survey. So it does, yeah, it does make me a bit cautious about that, but I'll do my own review of them and see. Maybe it uses other words than slump. Maybe it's just wanting to understand how to read more and read more or what makes me read less are all synonyms for reading slump.
Ste (48:23.616)
Yeah, I'm wondering why, yeah.
Ste (48:47.084)
Yeah, yeah, that'll be like the, I'm waiting for like the official atom analysis on this. It's kind of like seemed a bit odd to me, but yeah, I mean, when it like signals some stuff, I'm always curious, like why did they do that? And then I asked like what other thing because reading slumps seemed like a bit curious. So it said how their genre preferences influence their overall reading experience.
Adam (48:55.621)
You
Ste (49:16.494)
So they weren't sure about which genres they truly enjoy or how their genre choices impact their satisfaction with reading. genres again and how it relates to the reading habits that they have. yeah, that was like.
Adam (49:36.589)
Yeah, all of this genre exploration and it being so high up and it coming up so often on hardcover makes me wonder if we need to like switch to...
Adam (49:56.389)
I think it's called like, what's it called? It's like a different style of genre tracking.
Ste (50:04.322)
Okay, let me stop sharing this. So what would that involve? Because it's like, yeah, I'm sensing there's something here as well, just based off on what people are telling us. And it's pretty fascinating because I haven't thought of it that way before.
Adam (50:17.221)
Let me see.
Adam (50:29.209)
Yeah, it's a, like, you know, our genres, for instance, it's like, it's one level deep. It's a like fantasy, science fiction, science. there's another type of genre tracking that, publishers use that have like, fiction slash fantasy or fiction slash fantasy slash romance fiction slash fantasy slash dragons and mythical creatures.
Ste (50:29.398)
people.
Adam (50:58.915)
And so like the fit, like when you're in that genre of dragons and mythical creatures, that implies that you're also in the fantasy genre and you're also in the fiction genre. While maybe there's another one that's like, art drawing dragons and mythical creatures. And that one, it's a completely different cat. It's a completely different genre. Although the last step is the same in both.
Ste (51:15.795)
Yeah.
Ste (51:23.692)
That sounds like a lot of what you've been describing with book moods. Isn't that very close to how we'd make up the book moods that... Book moods are basically vibes. Sorry, book vibes. Yeah.
Adam (51:46.469)
It would be similar, but I would see book vibes as more like you add a bunch of books to it and then we calculate a score and those books that you add to it could be in multiple genres and we'll figure out the similarity between those books for you. But I feel like if we did have this data on like deeply nested categories or genres, it would make vibes more accurate because we would be able to know, these two books are both
dragons and mythical creatures and you're adding two of those same books. So we need to weigh other dragons and mythical creatures books really highly. While if we did that today, maybe we have a genre for fantasy and a genre for dragons. And maybe it would see that both of these books have those two, but it wouldn't be as accurate as if we had nested genres. that's, yeah, this is kind of.
Ste (52:17.172)
Yeah.
Ste (52:31.062)
Mm
Adam (52:43.747)
making me think of one more vote towards improving our genre system.
Ste (52:50.184)
Yeah, yeah, it's been like very revealing, you know, what what people have been telling us. I mean, we asked about stats, but I think we got like way more about what people want from from their reading. And I guess like the primary goal, I mean, we help people discover books, but the way we help them discover is by tracking like what they've read or what's
they like. think this is like particularly because like we do have like people with established libraries, but there's like a lot of people who don't know what they want to read and what genres they really like. I've heard and you know, this is also valid for me. I'm reluctant to go read a different genre than I...
I usually do just because I don't feel like there's anything that would interest me there. But then again, if we make it more granular. So for instance, let's say I don't like reading fantasy. I do like reading a bit of fantasy, but let's say no fantasy. What if I actually like a
sub -female fantasy that doesn't really strike me from the first go. And I don't think some books are, they could be something I like. I think for those people who don't have a lot of data, think taking it by genre, especially in the onboarding process, if they have the usual books, if they choose Lord of the Rings, they choose...
1984 and they choose like another very popular book, like whatever. Yeah, Greg Gatsby. Like there's not a lot we can like tell them. It's like, yeah, go read like any book. But if they tell us something about themselves, like what they like, maybe they like dragons. We could give them like a whole lot of books about dragons and a whole lot of like themes, tropes.
Adam (54:51.867)
GreatCatsP, I don't know, yeah.
Adam (54:59.419)
You
Ste (55:14.738)
So yeah, is really exciting. mean, it goes beyond stats, I think.
Adam (55:17.115)
Yeah.
Adam (55:23.247)
Yeah, it, it, it, it makes me want to improve our, book data. Like, and one of the other things is like a pages red, for instance, like right now calculating that on a daily level is not something we like store in our database. Like we just have, we have, like the times you started and finished books. And with the reading journal, we now have the times you
updated your progress from one page number to another page number. So using those reading journal entries, can, if you were to do that every day, we would know how many pages you read each day. Otherwise we average out the time between when you started and when you finished. And even a reading journal, if you update to 10 % on the first and then on the 10th, you update to being done. We know that you finished 90 % of that book within that,
date range, but we don't know how much you read every single day. So we average it out. So what I'm thinking that we'll probably need to do with stats is, probably create some kind of database table where we're storing like the number of pages you've read every day. And then we can recalculate that every time they do, every time they finish a book, every time they do a progress update.
and then we'll just be able to show that stat on the stats page rather than having to calculate it out, which will make the stats page load a lot faster.
Ste (56:55.966)
Yeah, that sounds good. That sounds good. yeah, pages like read per day. mean, we didn't have like tracking when we first did this. We didn't have like the reading journal. But I guess the like total experience, we're getting close to it now because you can update your reading every day. You can update, you know,
your progress and we'd be able to tell you. Plus it gets me excited because like lots of the things people mentioned like more, I guess, sensitive analysis, that will totally be opened up whenever we reach discussions because then we would be able, even now with the reading journal, know, if you offer your take on something you've read.
or submit a quote that really impressed you, I think would be able to tell how you felt or what's, even stuff about you that you, mean, if you highlighted things from a book, like your impression or a quote, right now, if I...
take that and put it into Chad GPT. If I take my quotes and my impressions reading a book and I put them into Chad GPT and just ask it, what do you think my experience was with this book? And maybe what can you tell me about myself through what's my tone, the words I've chosen? I think it's able to tell me pretty accurately how that...
how that went and it's not 100 % good, but I'm seeing it move in that direction as well. So yeah.
Adam (58:52.921)
Yeah, and the same thing's possible using like a sentiment analysis on it, which is basically what ChatGPD is effectively doing in that case. But we would be able to do it using sentiment analysis, which is a less costly way of answering the same question, where it's like understanding, like, yeah, you can, it's a much more low.
low cost way of doing it. So we could try something like that if we wanted as well.
Ste (59:26.346)
Yeah, it's great that stuff is evolving in our advantage, I guess. So yeah, I don't know how sentiment analysis is done, but if we can actually do that, that's great. We're starting to get more of this data about books. So we're able to tell readers these insights. yeah, pretty cool that.
For stats, I think we got a pretty good impression of what we should build. So I think next time we'll actually start drawing out the stats and how they're prioritized. How do you feel about where we're at with the feedback and with what the next steps are?
Adam (01:00:20.421)
Well, I was thinking about it like next week on Wednesday, we're having our like a hardcover, like general meeting for the team. Maybe we could do like a sketching exercise in there where we pick out like the top four stats, like books read, pages read authors and genres. And we spend some time as a team, like sketching what those could be like.
But I also think like that's like ideation while I feel like the two of us, could probably just like create some prototypes of this and go from there. So what would you think about you trying to create like a, prototype of each of those top four, for next week. And then we could build on that.
Ste (01:01:14.154)
Yeah, that sounds great. And it will be good to compare it with what we do in the team meeting and see how it lines up. So I'm not going to share anything on the team Discord. I'm just going to do that stuff. Maybe I'll tag you so we can both look at it. And then in the team meeting, we could do that. And then we jump on the live and see how those align. And people will be able to.
I finally see some prototypes. I know we've just been chatting mostly for them and brainstorming for the past episodes, but yeah, finally it gets visual and then it gets functional. I was even thinking we could basically start after this, after we sketch them, basically doing staging that people can access, and maybe we could get some live feedback on the staging domain.
Adam (01:01:46.085)
you
Adam (01:01:50.329)
Yeah.
Ste (01:02:10.626)
for the new stats. So basically, maybe just pull them out and get some, make some iterations with the community. But after, we actually sketch them and do everything.
Adam (01:02:26.821)
Yeah. Yeah. You mean like once they're in a state where you can look at your own stats for it? Yeah.
Ste (01:02:32.97)
Yeah, yeah, exactly. Like people seeing their own stats because they might like be very different and I'm curious how people like actually interact with them, which is going to be different from design.
Adam (01:02:40.485)
Yeah, that's true.
Maybe we can make this a, like a supporter prelaunch thing where like we could, we could ship it all the way to production, then only release the new stats to supporters and maybe librarians too, and get their feedback on it while we're building it. Because I have a feeling we'll be building it over like a month or two. So like, but it would be nice if it's not like unfinished until then, but we can like deploy it every day as we're working on it.
Ste (01:02:48.538)
yeah.
Ste (01:03:01.78)
Yeah.
Ste (01:03:10.786)
Yeah, that sounds great. And we can even try out some different approaches. I got some ideas on how we could do different kinds of tasks. So I'm curious. And we can release it to supporters. It's going to be interesting because it's going to be the first feature we actually 100 % release to support and elaborate as before. It gets to production. So if we can do that, yeah, let's do that.
Adam (01:03:38.875)
Yeah, I'm down for it.
Ste (01:03:41.792)
Yeah, that sounds great. Perfect. Well, I think, yeah, we're at time. Founder mode. Did you read the last essay from Paul Graham about founder mode? Brian Chesky from Airbnb gave a really like, well...
Adam (01:03:54.455)
No, you have to link me to it. I'm curious.
Ste (01:04:08.662)
they call it, well -reviewed talk at Y Combinator, and basically he talked about founder mode. yeah, I guess it struck me because it seemed like a lot, very close to what we were doing. So that's why I mentioned it, just to wrap everything.
Adam (01:04:30.233)
Nice. I will, I will go check this out and then, yeah, he was just on the Dax Shepard podcast. I'm chair expert.
Ste (01:04:37.684)
really? wow, I didn't know that. I'll have to check that out. Nice, perfect. Well, I guess thanks everybody for watching another episode of Hacker Vibe and see you next week.
Adam (01:04:47.227)
You
Adam (01:04:53.305)
See you next week.
Ste (01:04:54.796)
Bye bye.