Building tools for creative communities – with Hugh McGuire の全文書き起こし

faster. So that's how LibriVox came about. Arthur Attwell: 06:15 And am I right that it now has over 19,000 books in its catalogue? Over 13,000 different people have read books or read chapters for books in forty-seven different languages, and other projects building on its API, reusing that content. I mean, that sounds extraordinary. Hugh McGuire: 06:32 That sounds about right. I'm very rarely engaged in the day to day. It's something that I feel a little bit of guilt about and have for many, many years. But I mentioned it to someone and they're like, 'Well, that's great. You set up a framework process, a vision, and a community that has just kind of continued on with that. That's excellent. Well done.' And that's kind of how I feel about it. I think those numbers, as far as I know, they're correct. My last project with LibriVox was building a system that helped manage the workflow for the volunteers to participate. So that data, in theory, is probably pretty good. So I think, closing in on 20,000. And it's been about a 1,000 books a year. Basically, not quite the first year, but close to the first year. It's been about a 1,000 books a year published. And so 20 sounds about right. We'll have our 20th anniversary next August. Arthur Attwell: 07:27 A really amazing thing. Arthur Attwell: 07:28 You said in an interview once of the recordings themselves, and I'm going to read a little quote here, 'Our recordings have mistakes, fumbles, coughs, and beautiful human voices being human, reading wonderful works of literature. It's not just transmission of audio information. It's something much more and something no computer, robot, etcetera, could ever replicate.' Arthur Attwell: 07:53 Personally, I'm very much interested in what the readers get out of the process as much as the listeners. What do you see the readers getting out of creating those audiobooks? Hugh McGuire: 08:03 Well, I don't think there'd be one answer, but I guess my driving force of the things that I've done in my professional life since leaving climate is just the fundamental belief I have is that creating something and sharing it is the most powerful and meaningful thing we can do in our lives, or maybe not *the* most, but *among* the most. And so I think that it's just that expression of wanting to do something and make something, and not just make something, but share it to the world. That's what it's about. And we always - to the detriment sometimes of the quality of the recordings, etcetera - our focus was always 'let's make it easier and better and and just a process for the people who want to make these books.' And that's sort of our target market, and so we never cared how many downloads we had or what was the most popular book or ratings or anything. We never did any of that. It was always just, 'what do we do to help foster this small community who cares about doing this?' That has always been the focus. Hugh McGuire: 09:03 And it's interesting since then, you know,

in those early days, there were these awful machine read versions of audiobooks - Arthur Attwell: 09:10 Mhmm. Hugh McGuire: 09:10 - and I guess it was a year and a half ago, I got contacted by some people at Microsoft who were trying to figure out what to do. They were building some voices that were going to read public domain books as part of their AI voice program. They ended up doing a huge collection of stuff and putting it on the Internet Archive. And it's fantastic. It's really great. Hugh McGuire: 09:29 AI in some sense is going to replace the functional kind of process of recording audiobooks in meaningful ways. And they did a great job, but to me, great, that's good. What I care about is do people wanna keep recording them and sharing them with the world? And if so, then LibriVox is a place to do that. I think about a lot of these things that are most interesting are things that people do *because* they love doing it. And the Internet is an amazing place to share that, to the benefit of others, if they wish. But that the most interesting thing is the doing it. And no one ever wonders why it is that people play recreational soccer on Thursday afternoons without pay and without a audience or whatever. They do it because they love it. And so just finding ways to help people do the things they love and share them with the world for whoever wants them. I find that just a powerful driver, I guess. Arthur Attwell: 10:22 I think it's absolutely wonderful to actually consciously recognize that the creating of the audiobook and of the art or the soccer game is kind of obviously as important or more important than ensuring that you have listeners or, 'an audience' or a market. You did build a business alongside LibriVox called Iambik, which produces 'professional' audiobooks. Did that grow out of your experience with LibriVox? Hugh McGuire: 10:50 It did. That project is, almost fully wound down now. So I don't know if you know much about the audiobook business, but I don't know if it's any nicer than it was then. But at the time, digital audio was dominated by Audible - Arthur Attwell: 11:03 Right. Hugh McGuire: 11:03 - the Amazon company, and their terms were not very favorable. And I had the, which is not untypical of me, the interesting idea, with not necessarily the keenest business insight, I thought it would be great for independent literary novels, that have audiobook versions of them. It turns out that there's not a huge lucrative market, especially when the percentage you get from Audible is, shockingly low. Hugh McGuire: 11:29 So our idea was to match audiobook recorders with rights holders and kind of build things that way. In the end, Audible actually launched something called 'ACX', which modeled that. One of my claims to fame is I can say that Amazon stole one of my business ideas. But anyway, the literary fiction side of things never made any money, and we stopped doing that. And then it continued for several years, probably

ten years after that, doing a small number of audio recordings of textbooks - Arthur Attwell: 12:02 Interesting. Hugh McGuire: 12:03 - for some educational publishers. I think the AI voice generated outputs has kind of ended up killing that business finally. So that was my first effort to try to take, actually maybe my second effort, to take some of the things that I learned in LibriVox and apply it to the world of something that might be a business. It lasted for ten or twelve years or something like that. It never was a huge moneymaker, but it kept a number of independent audiobook narrators employed and a manager and things like that. So it had some positive impact on the world. Arthur Attwell: 12:35 Yeah. I think it's fantastic, and I think that you probably underestimate the kind of example setting you were doing to others, not just who want to create audiobooks, but who were just looking at what does it take to build new businesses, doing new things in publishing? And so many of these ideas that arrive and then become profitable in the hands of others later on, like Audible, needed to happen in the studies of bold entrepreneurs first. You'd also gone to start several projects in bookmaking, including Bite Sized Edits and Book Oven before Pressbooks, which really established itself as a significant step in making HTML first or digital first books possible, and we'll talk a bit more about that in a moment. One thing I did wanna ask though, in all these cases, you're building, as we said, tools and you're building communities, and when you set out on a new venture, how much do you think you're thinking about building the community, and how much are you just focusing on creating the tool, and we'll figure out the community later? Because both things take work, right? Neither happens by accident. Hugh McGuire: 13:41 I think I'm always trying to do both at once. Actually, in fact, Pressbooks is probably the one project I've done where that wasn't built in from the start. And I wouldn't suggest that the reason it was more successful was because of that, but I do think that there's you know, to your point, there's a focus problem. Focusing on building a big community is difficult. Focusing on building a working tool is difficult. And so trying to do both on relatively non venture capital budgets is not so easy. Hugh McGuire: 14:11 I have some new projects that I'm working on right now, and it's really that community place, that is most interesting to me. And it's hard to say, there are so many things where that just is built in by default. Would you say that Uber and Airbnb are communities? I don't think you would anymore. But I think there's just a huge amount of power in that, that we still haven't tapped properly in many domains. I would guess that if I were to think about it, the tools to me are always a means to an end of some kind, and so what's the end? I guess if I were to dream of what I want most in the world, it's communities of people doing wonderful, creative things together somehow or other, which wasn't possible before this kind of connective tissue of the web. So I think the community is really the driving force, but you just need the tools to make all that happen. Arthur Attwell: 15:03 For those who don't know it, can you tell us what Pressbooks is? Hugh McGuire: 15:06 Sure. Pressbooks, just a quick backup, LibriVox, which is a global community of makers of audiobooks, was enabled by a bunch of tools that existed that we cobbled together to make this community and make a publishing workflow. And there was Audacity which was an open free recording platform, and we used PHPBB, which was a free forum software, and WordPress, I guess, as a website. Just about everything kind of ran on those things. And we did a little bit more afterwards. Hugh McGuire: 15:35 But anyway, the insight was you could use these tools and then build a community to do something. And coming out of that experience, it seemed to me that there should be a tool online that you could use to make a book, and that was the original company Book Oven and Bite Size Edits kinda came out of that, with this kind of idea of 'let's build a toolset to allow a community to come together to make books'. And we couldn't quite get that off the ground at the scale that we needed it to. And then Pressbooks kind of evolved out of that as, 'Okay. Let's just focus on the tools and not worry about the community aspect and make a tool that people can make books online. And the idea is a web-based interface. This is kind of old hat now, but a web-based interface that would have outputs that would include ebook output, PDF for print output, and there would be a web version for those who wanted it, which in the early days was not very many people. That was the concept behind Pressbooks and it has continued on doing just that. Arthur Attwell: 16:33 And it's interesting you said old hat by now. I think it's old hat for you. Maybe for me too, I've been building something similar for a good eight years. But just recently, one of the sharp tech companies of the world, 37signals, they've just released a new web-based interface for creating books, and I think it's great that they're doing it. But those of us who've been doing it already for ten or fifteen years, we raise our eyebrows a little bit at the claim that it's new and revolutionary because we've been talking about it for so long. Arthur Attwell: 16:58 And so yes, old hat to us. But the rest of the world is just beginning to realize that creating books online and especially collaboratively and then being able to output a PDF and an EPUB, maybe a web site version of the book as well, is exciting. And it's hugely gratifying to me that we were right, Hugh. We were right, all this time later. Hugh McGuire: 17:20 I think it's kind of to your point earlier that I think the time has been here for a long time. And there's a bunch of projects out there that do this. So the

idea that they're new is quite surprising, but I think at the end, anyone who's building a software business of any kind, the fundamental thing for success is 'what's the business model that makes it work?' Pressbooks kind of backed into a business model by accident because I was interested in the thing itself, and we kinda built the thing itself. And, it's probably the most obvious thing in the universe, but it's something that I have underestimated in my career many times, is getting that business model *first* and then building the things that support that, rather than build a cool thing, and hopefully it can figure out what a business can look like. Arthur Attwell: 18:09 For sure. If we can allow ourselves to get into the weeds a little bit about how Pressbooks works behind the scenes. How do you make a book with Pressbooks as a user, and then what's happening behind the scenes that means you can make a book without needing layout software like Adobe InDesign or MS Word? Hugh McGuire: 18:27 So first of all, Pressbooks is built on top of WordPress. So when you use Pressbooks, it looks a lot like WordPress. And roughly speaking, each post, we consider that a chapter. We've kind of reorganized the whole internal administration of the system so that it's focused on bookmaking. But basically, you make a post which turns into a chapter in the book, and we take all the HTML that's kind of in WordPress-y HTML, but we convert that into a very structured format, and then that gets exported into EPUB, and then it also gets converted using cascading style sheets. Hugh McGuire: 19:08 We convert that using Prince, which I think you use - Arthur Attwell: 19:11 Uh-huh. Hugh McGuire: 19:12 - so a conversion engine, which will convert that into type layout ready for a book. Basically, you make a bunch of blog posts in Pressbooks, and then we output for you a book that is formatted with: title page and table of contents and page numbering and footers and headers and chapters and all the kinds of things that you expect in a book. Arthur Attwell: 19:33 It's absolutely fantastic to be able to create something that looks that good with no particular technical knowledge or design layout knowledge. Hugh McGuire: 19:41 Yeah. Arthur Attwell: 19:41 That's really, perhaps been at the heart of its growth. Specifically, it's grown in the open education resources space. Hugh McGuire: 19:49 Yeah. So in the end, I think the growth came from the place that I was always most interested in, but most people in the book world were not, which was the web-based book. And so, in fact, what's happened over time is that PDF output has become [less important] it's still important, but in the early days, our users were self-published authors mainly who wanted print-on-demand books published into CreateSpace or wherever. But in the end, with the open educational space, so I should probably explain that, this is an educational movement driven mainly in the US and Canada, where textbooks are very expensive, and there's been a movement to develop free openly licensed textbooks, digital textbooks, that students can access for free. That really has ended up being our overwhelmingly primary business - serving that community, and it's largely higher Ed institutions. Hugh McGuire: 20:44 So universities and colleges who are our clients, and they make Pressbooks available to their faculty to create these resources. And the nice thing about this ecosystem, just going back to the community discussion we were having earlier, is that if Arthur makes a great open free digital textbook on how to publish books in South Africa and I'm teaching in Montreal, I can just clone Arthur's book and adapt it to what I want it to look like and present that to my students. So that open kind of web-based delivery of books has been the driving force, I think. And we're relatively affordable compared to lots of other platforms. And so institutions like working with us because the affordability of putting a tool like this in front of their instructors who can then deliver resources to their students at no cost, is kind of the driving interest behind how people are using Pressbooks for the most part. Many other things are happening on it, but that's the main driver. Hugh McGuire: 21:43 And then the PDF becomes kind of an extra thing that people want, but that print-on-demand workflow is just not nearly as important within the framework of open education as far as we can see. Arthur Attwell: 21:53 Yeah. It's really interesting to see web books establish themselves as a format that institutions take seriously as a textbook. I know with some of our projects, we have to have a printed book in part to make the whole project seem more credible. There's some weird human psychology around if it's in print, it's a 'real' textbook. And sometimes that's really the only reason we're creating the print version, so that it could be around, but the students are all gonna be using the free online version. Arthur Attwell: 22:20 Do you see something similar? Hugh McGuire: 22:22 I also have that human quirk. I have always thought that the best advertisement is the print versions of these books, but our clients and users and the students, they occasionally say they want it. No one actually does anything. So I don't think that's true across the board. I think there's a lot of value in paper books. I think there's a lot of reasons why they are still important and valuable, but basically what the students actually want is a digital version of that book that they can access anywhere at any time. I think some of them like having access to a PDF, and they can print off little bits and mark them up if they wish to. But generally, the main use case is just on the web. Arthur Attwell: 23:03 And most of these are freely accessible as well. Hugh McGuire: 23:05 Yeah. So I don't know to what degree it's format preference versus price, but, you know, typically, in the open

ecosystem, your printed textbook is gonna cost you $25 instead of the $250 that you would be paying for your chemistry textbook or whatever. So I guess the problem is once you're doing books this big, even in print-on-demand, the price tag goes up no matter what the margin is. Arthur Attwell: 23:28 Makes a big difference to all students, for sure. I wanted to speak a little bit about the fact that community and tool building still takes money of one sort or another, and as I mentioned earlier, a lot of these kinds of projects that try to empower communities with new tools face a collective action problem in that no one person or organization will benefit exclusively from the development because they're inherently shareable, both the code and the content being created. So it's hard to find people willing to invest in their development, I suppose apart from philanthropy. Have you managed to piece together the resources for building these projects like Pressbooks over the years? Hugh McGuire: 24:06 I would say I was naive about that collective action problem. I thought open software that meets an important need in higher education, that there would be a lot of enthusiasm for paying us to develop that. That was not the case. Pressbooks in the end just became a software service company where the underlying technology happens to be open-source. Hugh McGuire: 24:31 So we're just a private business. Some of our development projects have been done essentially through clients who might have gotten a grant of some kind. Sometimes we collaborate in that way where there's a specific toolset that we would like to accelerate, and there's a client who particularly wants that and has some funding available to do it. But Pressbooks itself has never gone out to get foundation funding or anything like that. So we've just run as a a little bootstrap software service business, and I'm happy about that in the sense that you're forced by necessity to just focus on whatever your clients think is the absolutely most important thing. Hugh McGuire: 25:09 LibriVox came out. I don't know what I was imagining the world to look like, but that project was really just on the enthusiasm of 'everyone's gonna volunteer for everything in the world, and it's gonna be awesome in the future.' And so that's kinda how that came out. I guess we did get *one* grant to build some software at that point, and occasionally, we did little fundraisers to pay for a few things. Hugh McGuire: 25:30 The Internet Archive houses that. One of the big costs would be server costs for all that. So that's by the Internet Archive. And then Rebus, which we haven't spoken about, but that's a nonprofit that I started, and that has had foundation support. That's more around the process of helping groups of people create open textbook and open resources, but basically building capacity within institutions to build these resources because publishing is hard and harder than people expect. Arthur Attwell: 26:04 What sort of capacity needs building? What are the skills you reckon people struggle most with when they first start creating books for themselves? Hugh McGuire: 26:11 I think people don't recognize the scale of all the things that goes into making a good book. The planning - what is the content going to be? In our case, what is effective educational content; copyediting; structural editing? Oh, you've found a great photo on the web, but it's copyrighted. You can't put that in even if it's a web-based interface. Hugh McGuire: 26:33 So there's all that. There's the producing even. Pressbooks is an easy tool to use, but there's twiddling that has to happen to make sure that it looks good on the output. Often, these are collaborative projects with multiple people, so there's kind of like a cat herding problem. Deadlines! What do you do when it's done? Who do you announce to? Publishing houses exist for a reason, and they do a lot of work that people are just completely unaware of. It seems to me many people think they just took the word file and turned it into a paper thing - Arthur Attwell: 27:04 Sent it to the printer. Hugh McGuire: 27:06 - yeah. There's just so much more that goes into the publishing process that people aren't aware of. And so when people take on an open textbook publishing project or a programme they learn, 'Okay. There's all these things, and we have to figure out which things we can do, which we're just not going to do, and so be it, which we're gonna do to the best of our abilities.' There are just many, many, many aspects that go into publishing. Hugh McGuire: 27:29 And at the end of the day, also, writing is hard, and it takes a lot of energy to really get something good and final and edited and clean and ideally peer-reviewed and all that kind of thing. There's just a lot that goes into publishing. And so Rebus was built as kind of a companion, or envisioned as kind of a companion, to Pressbooks where if you were starting an open publishing program to be a mechanism to go through with a handful of books, in a cohort, to learn how to do this, and then you could go from there. Although, originally, actually, it was also trying to build a platform with community, but the cost of doing that right didn't quite work out. So, we focused just on that kind of capacity building training and we call it very light project - not project management - but project structuring. Hugh McGuire: 28:18 So we help people understand what the project looks like, and there's kind of a set of interventions that happen along the way where the cohort gets together to talk about where they are in their publishing journey. So it's been very cool to watch that. Arthur Attwell: 28:30 Every book comes with its own surprises, and I suppose one has to be ready for them. If an organization was interested in building that capacity, can they reach out to the Rebus Foundation and become involved? Hugh McGuire: 28:41 The website is rebus.foundation.

I think that will get you there. Arthur Attwell: 28:46 Fantastic. We'll have a look, and we'll put that in the show notes too. As we're running out of time, I wanted to ask a forward looking question. You wrote back in 2012, and I'm gonna read another long quote, you said, 'A real change that I see coming is not so much in how we read, but the context around the book we read. If you imagine every book with its own URL, every chapter with its own URL, then you can start to think about the information in books being truly connected in ways it can't be with print books or ebooks as we've conceived them so far. I think that will change where books live natively on the web first and then have different output formats. With this webification of books, we'll see connections being made from book to book, passage to passage, layers of commentary, links, the extraction and use of data, such as locations, times, and names. All these connections will sit atop a book and be turned on and off depending on the reader's preference. This layer will not necessarily be generated by the author or publisher, but rather by readers and other services that can access the book data and link to it and use it in different ways.' Arthur Attwell: 29:51 I couldn't leave any of that out because it's all so rich with thinking and imagining. How do you feel we're doing? Are we seeing that, or has it worked out in a different way? Hugh McGuire: 30:02 It's funny. If you would ask me two years ago, I would say we have done almost none of that. Very little of that has actually happened, and the reason is two things. The world of traditional publishing is largely not interested in this as far as I can tell. I know there are some siloed applications where some of this is happening in the academic space a little bit. Hugh McGuire: 30:27 And then on the open side, you know, in the Pressbooks ecosystem, just the cost of building these really cool things goes beyond what the business model, at least my version of the open publishing universe, has allowed. We haven't seen much of that. I think, however, that this is effectively what artificial intelligence is doing by having made an end run somehow. In my understanding its basically slurping up lots of copyrighted content that probably they weren't really allowed to do, but those kinds of connections now between works and books are enabled by AI, and that's really exciting and interesting. I don't know what it does for the business model of the universe. Maybe does McGraw Hill layer an AI on top of their catalogue? Hugh McGuire: 31:18 That's probably going to happen. So I think that AI might unlock a lot of this that was hard for people to imagine, whereas now it's a lot easier to build these kinds of things with AI than it was previously. And so I suspect we start to see more of this. But, again, there's kind of battle lines that's trying to get drawn between publishers of content and the AI companies that would like to be doing things like this. So I think I underestimated the entrenched business model of publishing that doesn't want all these cool things because they can't figure out how to monetize it. Arthur Attwell: 31:52 Right. Hugh McGuire: 31:52 And I'm not sure that people want to pay for the scale of what's needed. So it's hard to imagine that 200 years from now when you and I are woken up from our cryogenic pods to have the 200th anniversary of this conversation that all this won't be happening. But a lot of the things that I thought were going to happen to books since I started working in this space in 2009, just have been so much slower than I would have expected, but such is life. But I don't even know what piece of writing that came from, but it's a very compelling vision of what books could be. So I congratulate young me for thinking about all that. Arthur Attwell: 32:30 Yeah. Hugh McGuire: 32:31 Yeah. We're definitely not there yet. Arthur Attwell: 32:33 Well, it's an interesting future, and thanks for thinking of it. I think that's an exciting thing to look forward to, and thank you so much for all your work and pioneering with communities and tools. It certainly made a huge impression on me and the way that I've tried to innovate and build and, show for many others as well. Hugh McGuire: 32:51 I mentioned this before Arthur, but I'll just say it again, way back in the early days of Pressbooks, when I thought maybe I need to shut this down because I don't know how to fund it. You gave me a Shuttleworth 'lightning grant' or something. I can't remember what it was called - Arthur Attwell: 33:06 Flash grant. Hugh McGuire: 33:07 - yeah. Flash grant, and it was $5000, and that probably kept Pressbooks alive. So the fact that it managed to stay alive is at least, in part, due to your generosity there. So you're a big part of it. Arthur Attwell: 33:19 That's made my day. Oh, I'm so glad. That was a great privilege for me to be able to make that nomination. The Shuttleworth Foundation at the time had these little grants that its fellows could give to people that they thought were doing really important work, and you certainly were. Thanks, Hugh. It's been a pleasure. Hugh McGuire: 33:34 Bye, Arthur. Arthur Attwell: 33:37 This episode was edited by Helen le Roux and researched by Klara Skinner. How books are made is supported by Electric Book Works, where we develop and design books for organizations around the world. You can find us online at electricbookworks.com.