[Older Man] Call me Ishmael.
[Older Woman] It was the best of times. It was the worst of time
[Young Woman] There was no possibility of taking a walk that day.
[Young Woman #2] Undeamed sprag. How can you?
[Young Man] Alice was beginning to get very tired of sitting by her sister on the bank and of having nothing to do.
[Ryan Weber] Those openings line form classic literature are brought to us courtesy of the LibriVox community. LibriVox is a group of volunteers who record and upload free public domain audiobooks and so far, they’ve posted over eleven thousand. It’s impressive work but as you can imagine it’s also difficult to get a large group of volunteers, with different goals, different missions, different amounts of technical expertise to work together on such a massive undertaking. So, to explain how this group has made it work, I’ve brought in today’s guest.
[Amelia Chesley] I’m Amelia Chesley, I’m entering my fifth year of Ph.D. at Purdue University and I’m interested in digital communities and online publishing.
[Weber] Amelia is currently working on a dissertation about how the LibriVox community organizes a large group of volunteers through policies and communication. Her findings are interesting for technical communicators of all kinds because technical communicators often have to play a role of helping to facilitate large-scale online collaboration and getting multiple groups of users to work together to a common goal. This is Ryan Weber at the University of Alabama in Huntsville. Welcome to 10-Minute Tech Comm and I hope you enjoy this interview.
[Intro Music]
[Begin Interview]
[Weber] Welcome to the podcast Amelia. Thanks for joining us to talk about the LibriVox community and the research you’re doing with it. And what you’re interested in is looking at how this community runs with volunteers to produce content but I guess to get started can you tell us just what is the LibriVox community?
[Chesley] Yeah, LibriVox is this online volunteer audiobook publishing endeavor and it’s gotten bigger and bigger since it started in 2005. A web developer from Montreal name Hugh McGuire came up with the idea back when that Web 2.0 was getting really big and he was inspired by the open source movement, and Wikipedia, and all these things that people were doing online and he thought, “We need more free audiobooks on the Internet. How can we make this happen?” And he invited a few friends to just get it started with him and their first project was The Secret Agent by Joseph Conrad, so yeah the project draws on everything in the public domain, so that’s out of copyright that can be freely remixed or worked with in new ways. And volunteers just come together, they read the books that they’re interested in reading. They get to pick and choose and make new collections if they want and it’s solely run by volunteers who none of them get paid for this, none of them get anything out of it. They donate all their recordings back into the public domain.
[Weber] And what public domains like before the 1920’s, am I remembering that right?
[Chesley] Yes. The 1923 is the cutoff date and LibriVox is very strict about that so if there’s any doubt at all about it being in the public domain, they’re like, “Nope. We don’t want to-we don’t want to worry about lawyers.” They have that strict restraint to work within but other than that they’re very open. Like they pride themselves on being a very open community and inviting of anyone and any language and any reading abilities, any level of technical expertise. They’re like, “Everybody come help us record audiobooks,” and that’s their mission.
[Weber] And so, well the reason that you’re interested in this from a technical communications standpoint is because you’ve got this group as what you described as kind of an initially messy collaborative process, and then overtime it becomes more stable, into organizing these volunteers and getting them work together. What is the initial messy process look like and then how did it become more stabilized over time?
[Chesley] Yeah, I’m still looking into this and it’s very fun to go back to 2005 and 2006 and look at what was going on, on the Internet back then. And part of what I’m discovering is that it’s still quite messy in some ways, but it has stabilized in other ways. So, it’s maybe differently messy now than it used to be but back when it first got started they had what they had to work with and it was a little blog that McGuire had set up and they emailed files back and forth and used file transfer websites and they did set up a little forum and that stabilized a lot of things really quickly. So instead of just emails and a blog, they had a central place to talk about each project and have a thread, forum thread, for every project. So, that happened pretty soon but there are still some things like, they struggle balancing their openness, like everyone’s welcome with sort of a quality control because they do want to have pleasurable, like listenable audiobooks but they don’t want to discourage any volunteers. So that was discussed a lot and debated a lot early on and sometimes still comes up. In the beginning it was kind of a free for all and nobody really knew what LibriVox could be or should be. One of the early projects was written by the volunteers, as a NaNoWriMo project for when National Writing Month in November. They wrote their own audiobook and recorded it chapter by chapter and it was kind of a random fun project and now that sort of thing would not be allowed on LibriVox. They want to record things that were previously published, already in the public domain. They don’t want to mess with personal like self-published things like that would be. So, there was kind of room for that kind of creativity and experimentation before they kind of figured out, “No. LibriVox is just going to do this and their prime directive is recording public domain audiobooks and putting them out for free on the Internet.”
[Weber] Interesting so they have exercised a little bit more control and one of the things you’re interested in is all volunteer-based. It sounds like there’s a man who runs it but it’s you know it’s a collaborative governance in some way.
[Chesley] Yeah and actually Hugh McGuire’s not involved as much as he used to be. He used to read and record and in the beginning he did think that he would sort of be the gatekeeper and he would say, “Okay our first book’s The Secret Agent. Next we’re going to do this book, next we’re going to do this book,” but it grew so quickly that he could not-he just realized, “I can’t actually do this. There’s more people than- there’s more people that want to record audiobooks than I could possibly manage.” So, he recruited, a lot of the early volunteers just sort of stepped up and started sharing, this is another part of the messiness too, is they shared their personal server space for uploading files before LibriVox had its own little corner. The people who started the project and were really excited about it and invested in it, they came together and negotiated, “What do we want our catalogue to look like? What kind of database categories do we need to build into this thing? How do we want to tag genres or authors?” and all of those technical behind the scenes sorts of things. People brought their expertise from wherever they, whatever they did in real life, they would donate all those skills to help get the project up and running. So, today you’ll rarely here from Hugh McGuire who started the project, but you’ll hear from like veteran volunteers who have taken on administrative roles.
[Weber] So one of the secrets was making it a little bit more decentralized than they had initially planned, is that right?
[Chesley] Yeah, another change that happened was they originally wanted it to be a podcast. Like every day we’ll put a chapter of the book out on a podcast and then again it just grew way too quickly. They’re like, “We’re going to have a backlog of millions of chapters. We can’t-we can’t say that we’re going to put one out per day.” So, they kind of strapped the podcast idea although they have the community podcast. They used to have like a recent releases, like sample podcast where they do excerpts of things that had come out, but now it’s just all like finish a project, put it all up in the catalogue, in all the file formats you could want, and they partner with Internet Archives. So, all of the final files are hosted by Internet Archive, and that cuts down on cost for LibriVox too.
[Weber] Great, so really the question you’re asking, we should mention that this is your dissertation project that you’re working on at Purdue, the question you’re asking is how LibriVox volunteers manage, control, and negotiate procedures and policies for their ongoing collaborative work? What kinds of strategies do they use? You’ve mentioned a few but are there others that they use to sort of get these volunteers to work together and ensure both freedom and quality control at the same time?
[Chesley] Yeah part of that has evolved over time too. They measure everything against that prime directive, so if a volunteer comes in and they have a suggestion or they want to do this particular thing, does it fit the prime directive of LibriVox to make public domain audiobooks and give them away for free. And if it does, great, and if it doesn’t, if it’s going to damage that or make it harder to do that then they’re pretty strict and from about like, “Sorry we just can’t do that here. Find somewhere else to do it if you want to.” They also are very flexible otherwise so between the constraints and the flexibility, they’re pretty open to everything. Like I mentioned they’ll accept any recording as long as it fits the technical specifications and as long as it’s loud enough, within a certain decibel range, and other than that there’s no quality control other than to prooflisten to every file. So, everything does get prooflistened just to catch major errors, like repeats or long silences, or background noise that would really get in the way. That was something that they didn’t have at first. They didn’t have proof listeners until like about six months into the project they realized, “Maybe it would be good,” and I’m just delving in now to how that was negotiated because some people thought, “No we don’t want to add to our workload. Like that’ll just be a logistical nightmare,” and other people thought, “Well we do need some way to catch these errors that are actual errors and not just preferences of style or something like that.”
[Weber] So the-this group, these volunteers that are making-that are doing all this negotiating, right that’s one of the things you’re interested in, is how do you negotiate with far flung people, you know who are on volunteer basis. How-what kinds of things do they do to sort of figure out like how, how they’ll arrive at a conclusion or a particular policy?
[Chesley] A lot of it happens in the forums, those forums that were set up and people post back and forth and share their side and typical sort of rhetorical debates about what-why I think this is best, making arguments. There is a personal message system. I really don’t know how much that’s used, that’s one thing I don’t really have access to but one thing I’m noticing is that negotiation may or may not be the best word to use because LibriVox has this grown towards all these very strict policies to not allow any unasked-for feedback.
[Weber] That’s interesting. This seems like you have to close those things off. That’s why I was interested in you saying, you know they use these forums because you know it’s always surprising what an Internet forum like produces the results, that’s shocking to me. So, this is an organization that’s been running for twelve years, that’s been crowd sourcing and getting things done effectively. What can we learn from this community to make other crowd sourcing efforts more effective, more productive?
[Chesley] This is the big question I’m still trying to really get my head around but think there are a few things that stick out right away and that clarity of purpose, that’s one thing that Hugh McGuire talked about when, when he had started this project and was so surprised that it was so successful because it wasn’t a very clear what that community was meant to be doing. I think another point is to trust the crowd. Trust the crowd can come together and do all these amazing things, so trust the people can make awesome stuff. I’m also learning that there’s a value to the chaos of it. So, you have that clarity of purpose on one hand but then you give volunteers freedom in how they go about that. There’s no two volunteers that have the exact same recording setup, or that use the exact same software in the exact same way, and yet it still, they can still all contribute to this awesome giant project. So, trust, clarity, some messiness, I think those are the three things that are coming out.
[Weber] Good luck with the rest of the project and thanks for telling us about your work so far. It sounds really interesting.
[Chesley] Thank you.
[End of Interview]
[Weber] I hope you enjoyed this interview with Amelia. As a special bonus, I’ve included a short poem that she recorded for LibriVox. I hope you enjoy.
[Chesley] Finis by E. Eastlin Cummings read for LibraVox.org by Amelia Chesley. Finis
Over silent waters day descending night / ascending floods the gentle glory of the sunset in a golden greeting / splendidly to westward / as pale twilight trembles into / comes the last light’s gracious exhortation lifting up to peace / so when life shall falter / standing on the shores of the eternal God may I behold my sunset / flooding over silent waters