Interview

Interview with Mark Smith and Denise Paolucci

zvi LikesTV

Maryland, United States

[0.1] Abstract—Interview with Mark Smith and Denise Paolucci, conducted by zvi LikesTV.

[0.2] Keywords—Blog; Fan community; Legal; Open source software

Smith, Mark, and Denise Paolucci. 2009. Interview with Mark Smith and Denise Paolucci. Conducted by zvi LikesTV. Transformative Works and Cultures, no. 3. https://doi.org/10.3983/twc.2009.0166.

1. Introduction

[1.1] This is an interview with the owners of Dreamwidth Studios, LLC (http://www.dreamwidth.org/), and founders of the Dreamwidth open source software project, Mark Smith (http://xb95.com/) and Denise Paolucci (http://denise.dreamwidth.org/tag/about+me). Mark and Denise were both employed for several years by former owners of LiveJournal.com, with Mark working on software development and Denise heading user management.

[1.2] LiveJournal and other Web sites that run on its code, known as LJ clones, are widely used by a portion of both media fandom and speculative fiction fandom as a blogging platform, newsreader (known as the friends list), and social media platform.

[1.3] Dreamwidth, the software, is a fork—that is, an independently developed version of existing software with functionality that deviates from the original—of the LiveJournal code. Mark, Denise, and approximately 40 volunteer coders have transformed the software in a number of important ways. Perhaps the most important change that developers face may be the massive rewrite that reorganizes and simplifies the code's structure. Similarly, the most important change for users may be the separation of the friends list into a reading page and an access list. On a LiveJournal-code Web site, friending someone means to both allow them to see secured entries on one's own journal and to have them appear on one's friends page—that is, one's newsreader, where journal entries are aggregated for reading. In the Dreamwidth software, these two functions are separated into giving access and subscribing. This change had been requested by LiveJournal users for years. It is a fundamental change in the privacy capabilities of this journaling model, which is significant because the finely granulated privacy controls have long been one of LiveJournal's unique features.

[1.4] Dreamwidth.org is the installation of the code owned and managed by Mark and Denise. In contrast to the practice of LiveJournal's various owners, Mark and Denise intend that anyone will be able to set up his or her own installation of the Dreamwidth code in a straightforward manner, with all of the functionality required to host users and administer the Web site. (Much of the code required to fully administer a LiveJournal clone is not available as free software; an LJ clone has to reimplement some fundamental software, such as a payment system, on its own.) The Dreamwidth code and the Dreamwidth.org Web site are currently in beta testing.

[1.5] This interview was conducted via IRC and the text then edited for brevity and clarity. Zvi (http://zvilikestv.net/), who conducted the interview, is project leader for the Dreamwidth Cool Hunters and a former Webmaster for the Organization for Transformative Works.

2. Software

[2.1] Q: What's your goal in starting Dreamwidth?

[2.2] DP: I think one of the major things about Dreamwidth that I personally find interesting is the fact that we're getting a chance to do things again. Both Mark and I are approaching this project with a major eye toward our professional experience—things that went well, things that went poorly—and taking that experience to heart. People have called DW "a bunch of ex-LJ staff rebooting LJ from the outside," which is a slightly glib, but reasonably accurate, description.

[2.3] We're trying to avoid framing things as anti-LJ or LJ-opposite, but we really have learned so much from our time on LJ about what makes for a good online community, and I think that it's really helping us as we go.

[2.4] Q: Is that familiarity the reason you went with what is, if you're only looking at the open source stuff, an incomplete and out-of-date software suite, instead of integrating a modern open source blogging tool and a modern open source bulletin board, or whipping something up from scratch?

[2.5] DP: Joel Spolsky (of Joel on Software) has a whole rant about why it's important to never throw out your code and start over again (http://www.joelonsoftware.com/articles/fog0000000069.html), essentially boiling down to this: when you reuse code, you get the benefit of all the work that's already been done on that code, both in terms of architecture and bug finding and fixing. When we really set out to build an online community site, we knew that if we "rolled our own," we'd be in development for years and the product, when we finished it, wouldn't have that benefit of 10 years' worth of bug fixes, architecture, and design.

[2.6] Part of the final decision was, of course, that both Mark and I intimately know the LJ code, and part of it was that we knew we'd be able to make it do what we wanted it to do. Having to learn from scratch would delay us considerably. It took us nearly a year to get DW production-ready as it is, and that was with an already established code base. Part of it was a really emotional decision: we both knew LJ, sweated LJ, bled LJ over the years, and having a chance to fix all the things that always annoyed us was really compelling.

[2.7] But more than that, there are a lot of things that LJ does right. Most of the major Web 2.0 features or functions that are de rigueur on other sites these days were in LJ circa 2002–2004 or so, either in nascent form or fully featured. To the best of my knowledge, LJ was the first to have threaded comments with e-mailed notifications of replies, for instance, and most of the community/group blogging tools that are standard on the Internet now started out on LJ.

[2.8] MS: LJ was the first service, and one of the only (to my knowledge), to allow voice posts. To have a reasonably integrated photo system. To have security for your posts (and still probably the only to have such a fully articulated security system, although Vox [http://www.vox.com/] comes closer). To have a built-in aggregator (friends page). LJ Talk, an instant messaging service. The list goes on; there are a lot of things that LJ does or did first, and still is the only site that does a lot of them.

[2.9] DP: And a lot of the reason why LJ never got the credit for having pioneered those features was due to either poor promotion or usability fail—that is, making a tool less useful by implementing it so that it is difficult to use. We really wanted to fix that.

[2.10] Q: One of the concerns in some segments of online fandom has been about that separation between LiveJournal and the rest of the Internet. Specifically, there's concern that the LJ code doesn't play well with others (neither its own clones nor other blogging platforms) in a way that a lot of other blogging software does now. Will Dreamwidth further that splintering effect?

[2.11] DP: I think (correct me if I'm wrong here, Mark) that was actually one of the first things I pointed out as something I specifically wanted to address with DW! Like, within the first 10 minutes of us talking about DW.

[2.12] MS: This was something we talked about extensively back when we were deciding whether or not to use the LJ code. Going with the LJ code base meant we would not end up with an interoperable site, by default. We're against that.

[2.13] DP: I'd say that about 50 percent of the work we've done has been to improve the interoperability functions. Our medium- to long-term goal is to make it so that anyone on any DW-based site can communicate seamlessly with anyone on any other DW-based site, with what site they choose to host on being totally transparent.

[2.14] MS: We have addressed it, so far, in the context of interoperability mostly with other LJ code sites. However, all of the core features we're working on have future growth in mind. Andrea Nall built the importer so it can import from WordPress, for example. It does not presently, but it's modular to make it do so. It's the same with the crossposter, which posts a Dreamwidth entry to another blog. Allen Petersen built that modular as well, so it's reasonably straightforward to support that.

[2.15] Q: So when you say interoperable, you mean DW code to DW code and DW to Internet in general. Do you have any feelings on whether LJ or the other clones are going to work with you for greater DW-LJ compatibility? How and to what extent?

[2.16] MS: The interoperability standard LJ has maintained for the past decade is woefully out of date and will not be Dreamwidth's policy. I want to support OAuth/OpenID 2, Open Social, and whatever other cross-platform functionality that makes sense for the community or for the Internet at large. I would rather us be more compatible with, for example, WordPress and Blogger than focus purely on LJ-code platforms.

[2.17] DP: LJ's been pretty good so far with working with us. Tupshin Harper, LJ's director of engineering, is really interested in working with us on feature development and the like. I think there's going to be a limit to the amount of cross-site interoperability that LJ is willing to uptake, just for business reasons, but I'm hopeful that we can achieve a better set of features over time.

[2.18] And I think that with our stated commitment to accessibility, we'll see a lot of people who start using DW as an overlay to their presence on the rest of the Internet, and I really want to support that. I have this pie-in-the-sky dream of DW eventually evolving into a centralized dashboard of our users' Internet presence, pulling together all of their various social media uses in one place.

3. Accessibility

[3.1] DP: We've already seen a lot of people saying that they find DW so much easier to use than LiveJournal and other blogging platforms, specifically because we work so hard to be accessible to people with disabilities or people who need assistive technology, that I think it'll be a major use case.

[3.2] Q: Why are you so focused on accessibility?

[3.3] MS: Because it's the right thing to do, in my opinion.

[3.4] DP: It sounds glib, but really, it's because it's just the right damn thing to do! I have disability issues myself, of a physical nature that don't tend to impact my computing use, but I know what it's like to live in an inaccessible world, and I know the relief that comes when you find someplace or someone who just gets it. Part of what we really believe, from the very beginning, is that DW should be welcoming of everyone. So the accessibility is part of that.

[3.5] Q: In what ways is it making the development process harder or easier? Who benefits from this push for accessibility, just people with disabilities?

[3.6] DP: I think we all benefit, really, both on a practical and on a social scale. Socially, it teaches everyone to be aware of the people around them and their needs, and practically, a lot of the functions and features that make DW work well with assistive tech also make DW more usable for everyone, and make DW display better for low-bandwidth users, people on mobile devices, and people on older computers.

[3.7] It does make the dev [development] process a little harder, both in that we have to constantly be thinking about it and in that it can take some extra time to add assistive tech or code around some of the inherited accessibility problems, and there have been times when it's taken more time to finish a feature or when fixing something for accessibility purposes means introducing new bugs.

[3.8] MS: Designing an accessible Web site is a harder task than making something just shiny and pretty with the latest whiz-bang features. Practically, we have had to, and will have to, spend more effort to build a site that is more accessible. I do not believe it has slowed us down any, however, since by making the decision to support accessibility we have gotten some superb volunteers who are contributing to that part of the project who probably wouldn't have contributed to DW otherwise.

[3.9] DP: And, of course, it's a huge shift in thinking and design. I know that I had a whole bunch of things that I never used to notice until someone would point it out: "Hey, this isn't gonna fly." But we learn from our mistakes.

[3.10] And yes, by being more accessible, and by making it such a core focus of our dev process, we get so many incredible people coming to volunteer for us, because they see that we're welcoming them and valuing them as people.

[3.11] MS: I think that everybody benefits, too. Especially in today's modern world, having a version of the site that works well for people who use low-resolution screens and large fonts also makes it more compatible with mobile devices, which are increasingly more common.

4. Users

[4.1] Q: Welcoming everyone sounds warm and fuzzy and terrific, but "everyone" includes a lot of people who are unpleasant to other users because their ideology is explicitly hateful, or they have poor social interaction skills, or they just like to make other people miserable. Is everyone going to include, for want of a more precise term, jerks?

[4.2] DP: John Gabriel's Greater Internet Fuckwad Theory (http://www.penny-arcade.com/comic/2004/03/19/) is always going to apply.

[4.3] MS: Yes, jerks are welcome on Dreamwidth, but please mind the terms of service (http://www.dreamwidth.org/legal/tos).

[4.4] DP: I think, given my experience on LJ's abuse prevention team, that I've probably seen pretty much every possible variation on "people being jerks to each other on the Internet" that it's possible to come up with!

[4.5] Our general philosophy on users being jerks to each other is to go for a technical, instead of a social, solution: make stronger antiabuse tools, and put their control in the hands of the users. Basically, sites can go for one of two methods when it comes to antiabuse tools. They can either use administrative solutions (site admins banning or blocking people, cutting off IPs [Internet protocol addresses] at the source, issuing warnings, generally behaving as the stern parent figure), or they can code around it.

[4.6] A very basic example of a code solution is LJ/DW's ability to ban a user from commenting in your journal, for instance. Or a community maintainer's ability to remove someone from a community, or screen their comments so they're only visible to the commenter and to the maintainer of the community.

[4.7] Generally, with my experience on LJ's abuse prevention team, I can pretty conclusively say that as soon as you have a site administrator issuing a "hey, stop being a jerk" warning, or putting down boundaries, the person in question is going to start looking for ways to dance right up to the line, or (if blocks are set) looking for ways around it. There's something about a rule that makes people want to break it!

[4.8] So our ideal is less us saying "that's not okay" and more letting people tailor their individual DW experience. There are really fine-grained privacy tools, for instance, as well as different commenting settings and different levels of access.

[4.9] I actually think that our split of LJ's "friends" concept into its two component parts is a major step forward there. On LJ, you can't subscribe to people without giving them access to do things in your journal, such as commenting if you have your settings set to only allow friends to comment. On DW, there might be someone whose writing you find interesting, in a "makes my blood boil but I want to keep tabs on it" sort of way, and our setup allows you to do that without giving them any sway in your space.

[4.10] Q: So what do you say to the people who say that because some particular subset of everyone is on the service, they can't use Dreamwidth? I've seen the suggestion, for instance, that if Dreamwidth allows fictional depictions of underage sex, that schoolteachers won't be able to use the service.

[4.11] MS: It seems like an impossibly slippery slope. If we ever say "x or y is not welcome on DW because it offends z," then we are going to become arbitrators of what are valid values for x, y, and z, and that's not a position I want us to ever be in. If we ever tried to do that, not only would Denise and I probably disagree, but we'd also end up making a site that suits our particular worldviews to the exclusion of others. The only reasonable limit that I feel we can ever have on content is the one we have now: what's restricted under U.S. law, and what is harmful to the service itself, such as spam, viruses, and exploits.

[4.12] DP: I think that no matter what we choose to do, someone's going to be happy about it and someone's going to be offended by it or negatively impacted by it.

[4.13] To loop back to a previous discussion point, this is one of the reasons why I really want to improve our federated blog network features: that way, people who feel like they don't want to or can't support Dreamwidth can set up their own instance of the code, or find another instance of the code that's more to their liking, and still participate in the greater community.

[4.14] I'd really question your particular example of underage sex and schoolteachers. I don't think that just sharing server space with someone constitutes an endorsement of their activity, really. Do you know what everyone who's using your e-mail provider is doing?

5. Legal relationship with users

[5.1] Q: I've seen a lot of people questioning what "welcome" will entail, a lot of speculation that, bluntly, Strikethrough (note 1) could happen here, too. In practical terms of not deleting accounts for violation of the site's terms of service and responding to letters of lawyers, what does "welcome" mean?

[5.2] DP: In practical terms, we're not going to take down content unless it's illegal. And by "illegal," we mean, "a court has ruled that this specific content violates the law." This is another case where my experience with LJ really serves us, since a lot of what fannish content creators are concerned about is the possibility of a DMCA (note 2) takedown notice from the rights holder of the specific property that they're fannish about. The DMCA is perhaps the most abused piece of legislation on the Internet.

[5.3] The DMCA does set forth what we have to do, yes. But even within that really strict set of obligations, there's some wiggle room, as outlined in DW's policy (http://www.dreamwidth.org/legal/dmca.bml).

[5.4] Our policy was specifically written to mitigate some of the worst abuses. For instance, the law says we have to deny service to repeat offenders, but it doesn't say what "repeat offender" is. So we wrote the policy as something that a court would likely find reasonable, but which legitimate use (that is, not someone who's using DW solely to spread cracked copies of software or something) isn't likely to trip.

[5.5] Another thing that came from experience is that most services don't forward the takedown notice to the user in question. The law doesn't say "you have to forward the notice," so a lot of places don't. We do. We'll also send it to the Chilling Effects Clearinghouse (http://www.chillingeffects.org/), which is a group that specifically seeks to document abuses of the DMCA.

[5.6] Also, a lot of people don't realize that the law sets forth a way for the alleged copyright infringer to dispute the notification, and again, a lot of services don't educate their users about that right. We're going to be really clear about that too.

[5.7] So if, say, Marvel Comics filed a DMCA takedown notice on someone's X-Men fanfic, the law says we have to disable access to that material. But we'll tell the user what we're doing, forward on the takedown notice, and give them information about where they can learn about their rights. Our DMCA policy actually explicitly says that people should contact the Organization for Transformative Works (OTW) (note 3) if their disputed work is also fan work. I checked with OTW's legal committee, and they were all for us including that link.

[5.8] Q: So, to clarify, you're not offering any legal assistance, but your concern is more about providing information to your users than avoiding having an intellectual property holder send you lots and lots of letters—or, even if not an IP holder, just a loud and obnoxious third party.

[5.9] DP: Yes, exactly. We can't offer someone legal assistance or help defend their use, but we can tell them what their rights are and point them to where they can get more information. We're very used to loud and obnoxious third parties shouting at us.

[5.10] MS: People who are worried about DMCA should protect their posts (http://www.dreamwidth.org/support/faqbrowse.bml?faqid=54). Don't leave it public, where the great Google can turn it up and the DMCA-happy lawyers can find it!

[5.11] DP: Or even just turn on the "view with discretion" bit (http://www.dreamwidth.org/support/faqbrowse.bml?faqid=146), which is another layer of protection.

[5.12] Q: Dreamwidth is a for-profit business, and you both to intend to earn a living wage from the site. However, a lot of the work on the site—programming, answering support questions, creating CSS designs for journals—is done by volunteers. Why do you think that profiting from the work of your volunteers is a moral business plan? What rewards, tangible or intangible, do volunteers receive from working on the site?

[5.13] DP: Our volunteers are going into it with the same attitude we have, really, which is that we're all coming together to make something special and long-lasting, and everyone participates as much or as little as they have the time, energy, and experience to participate. I think our volunteers are mostly people who are looking for an Internet home that treats them with respect, where they can really make a difference and get things done, and we're offering that. Mark and I aren't trying to be the power-hungry dictators or the profiteering creeps—our role is more as the people who are organizing and guiding everyone's efforts, and facilitating everyone's work to work together to benefit the entire service.

[5.14] DP: Obviously, we are taking in money, because servers don't run on idealism, but we aren't going into this expecting to get rich. Our operating agreement actually specifies that two-thirds of our profits have to be reinvested in the project somehow, and half of that money will get directed as the community specifies. I view it more as a collective project, where we, as the people who can devote full-time attention to the project, get paid a living wage in exchange for the time and effort we put in, while the rest of the income gets used to further the project as a whole—but we're all working together to build something. Mark and I just happen to be the ones legally responsible for the bills in the end.

[5.15] MS: Tangibly, we provide active volunteers with free paid accounts. We also provide them with the infrastructure they need to work on what they want: development environments, tools, and so on. Intangibly, there are many rewards to volunteering: knowing that you're contributing to a project you care about, feeling like a part of a team, helping out the community, feeling in charge of the site's destiny, knowing that the things you're contributing are actually being used by thousands of people. There's a lot of positive energy that comes from contributing to something voluntarily.

[5.16] MS: Volunteerism, especially as it relates to a for-profit business, is nothing new. I used to volunteer for LiveJournal, as did Denise, when Brad Fitzpatrick, the founder of LiveJournal, was making money and buying himself a house off it. I may not have derived a monetary benefit from that work, but the value I got in terms of professional experience was immense. Learning to deal with a development process, people, and the support team was an intensely valuable thing for me. I would not be where I am professionally without having done that.

[5.17] MS: One of the things I least expected when I joined the LJ volunteer team was that I would end up meeting people who would become good friends to this day—nearly a decade later. The people I share a house with, I met through volunteering. My life has been permanently enriched by the experience, and I expect people will have a similar experience with volunteering for Dreamwidth.

[5.18] DP: We also get a lot of volunteers who are excluded from the traditional workforce in some way, whether through disability or life circumstances, and a lot of them say that working on the Dreamwidth project really provides them a place they can use their skills that they wouldn't otherwise have. Our accessibility project coordinator, for instance, is totally bed bound, and she's one of the most brilliant people I know. We can offer her a project to work on that fits her schedule and her health issues. A lot of our volunteers are in similar situations, and they've said how much they appreciate having a place where they feel they can make a difference.

6. Dreamwidth and fandom

[6.1] Q: A lot of interest in Dreamwidth has been as a place for fannish activity. What kinds of fannish activity were you seeing on LiveJournal? What kinds have you seen move over to Dreamwidth? Are there any fannish activities that you saw on LJ that you are not seeing move to Dreamwidth?

[6.2] DP: One of the most prominent pieces of fannish activity, obviously, is fiction writing, and I've been really happy to see—even in our first 3 weeks!—how much of that is appearing on Dreamwidth. In the past few weeks, I've been seeing more and more stories posted to DW appear on fan fiction recommendation lists, and that's great to see. For instance, five out of every six stories I see recced in the New Trek fandom are posted to DW, and that's awesome! (I think our longer post and comment length limits, compared with LJ, is really appealing there.)

[6.3] One bit of fannish activity that isn't going to be switching to DW anytime soon is anything having to do with icons, graphics, or art—we don't have any form of photo or image hosting yet. We're going to have to do a lot more work to improve the image posting functions, though, since that's something LJ code doesn't really have.

[6.4] MS: I've dug a bit through metafandom (http://community.livejournal.com/metafandom/) to get an idea of how many links are on LJ and how many are on DW, and have been pretty happy to see how many we're seeing on DW.

[6.5] DP: We want to be a real home for all kinds of creative work—written and visual. But I have seen some people questioning the "creative work" element of things, and I do also want to be clear that we don't just mean fiction or art. We mean any kind of creation: essays, link organizing, recs, recording your life, everything.

[6.6] MS: I think that's part of our drive for accessibility too. I believe, at my core, that everybody has something useful to contribute to the world, even if it's just who they are, their metaphorical voice. "Even if it's just" makes it sound like I'm setting that as a second-class citizen, which is not my intent.

[6.7] Q: I was wondering to what extent you'd seen migration among role-playing gamers, music fans, sports fans, or foodies. Were those organized communities on LJ? Are they moving as communities to DW?

[6.8] MS: We've had a few people asking us for invites for their role-playing game (RPG) group. We've granted most of those.

[6.9] DP: And I can think of a lot of interesting features that we can put in to make RPGs easier—things like switching back and forth between a "master account" and various "sub" accounts more readily.

[6.10] I haven't really seen a lot of adoption among sports fen, music fen, foodies, and so on. Mostly I'm seeing people who are using DW already who have these as secondary interests creating communities for them, rather than moving those interests to DW being the sole drive. But I'd certainly be interested in analyzing those communities, seeing what they want or need in terms of features, and adding some of those as well.

[6.11] One of the things I always loved about LJ was seeing what various subcultures used the tools for, and what kind of weird and funky adoptions came up. I've been trying to make lists of those types of repurposings, and am really interested in making things easier and more natural for people to use, based on that.

[6.12] Q: What about the communities that are connected to media fandom: Comics? Speculative fiction? Horror? And are you seeing original fiction writers, pro or amateur but not necessarily speculative fiction, moving to Dreamwidth?

[6.13] MS: We have briefly talked to the people behind scans_daily from LJ. They were an LJ community that posted scans of comic book pages with commentary; LiveJournal deleted their account because it decided that those scans were a copyright violation and thus violated the LJ terms of service. They are looking at DW as a possible resting place. I don't know about the rest.

[6.14] DP: We've been doing a lot of outreach in the speculative fiction community, since a few of our core project team have roots in that culture. There are a few pro speculative fiction authors who are really interested in DW and in the drive behind it. The majority of moves so far have been in science fiction/fantasy fandom, but we're seeing some others as well.

[6.15] Q: What other subcultures, interest groups, and communities are you seeing come to Dreamwidth, and what's attracting them?

[6.16] DP: Aside from fandom, a lot of our early adopters have been people in various alternative sexual or gender identity cultures: BDSM/kink communities, the queer community, the transgender community—subcultures with really strong identities and senses of community.

[6.17] And I think part of that is the feeling of unease that a lot of people get on other services, where they fear that their kind of expression might be less desirable or more "edge." Having our diversity statement there and public helps reassure people that yes, we do know about them, we're not scared of them, and we welcome them.

[6.18] MS: Explicitly welcome them.

[6.19] DP: Yes! The more varied the voices we have, the happier we are! I've always thought that the strength of an online community lies in its diversity, so it makes me really happy to see differing voices here.

[6.20] MS: We've had some people write in to ask if they would be welcome. One woman, from the BDSM community, asked us about moving her Yahoo! group over. She was concerned that they post pictures—and that they were R rated. I told her to come on over.

7. Fan friendly versus fannish

[7.1] Q: Your DMCA policy and your general welcoming stance is what you point to when you say Dreamwidth is fan friendly. But you've clarified repeatedly that the service is not fannish. You've also seen a lot of mistaken belief that Dreamwidth is a project of the OTW. Can you explain that distinction and its implications for fannish users and the service in general?

[7.2] DP: I think that part of the reason that people confuse us with the OTW is that we are, to the best of anyone's knowledge, the only two majority-female open source projects on the Internet, and another part of the reason is because they both spring out of the same vague social network. A lot of people who are real proponents of DW are also proponents of the OTW, and vice versa.

[7.3] Part of the reason why I really want to clarify that DW is not a "fandom" project goes back to my point about diverse voices, though. For a lot of people, their fan work and fan activity are an important part of their off-duty time, but it's not the be-all and end-all of their lives. One of the really neat things I always thought about fandom's presence on LJ was that you'd get to know someone for their fan work, and in the process, you'd get to know them as a person because the personal posts were mingled in with the fan work. So I think the really neat side effect of having fannish work on a service that's not fan specific is that breakdown effect—that people can socialize both in and out of a fannish framework.

[7.4] MS: For me, the distinction is one of focus and priorities. I want to prioritize what we do on Dreamwidth based on what our users need and where we want the site to go, not based on what x group or y group needs. Note that if our users are 90 percent from x group, then certainly most of our effort will go toward making the site better for that group.

[7.5] DP: I think, though, that there are things we can do that will be usable and useful for everyone, and that's the design goal I've been trying to keep in mind the whole way—not designing to a specific subculture's usage, but making features that are flexible and powerful and that can be repurposed for a whole bunch of uses. When I'm doing feature design, I always try to make sure that the features I'm designing can be used in at least two or three different ways.

[7.6] Part of where LJ always fell down, to me, was the content management tools. There's a really heavy emphasis on privacy and security, which I love, but it's focused on creating new content, not finding and categorizing old content.

[7.7] Q: One of the ways in which you've been clarifying the distinction to the fannish community is in your personal identities. Mark, you do not identify yourself as a fan, whereas Denise, you do. Can you both explain what you mean with these identifications and how (if at all) they may affect your interaction with your Dreamwidth user base?

[7.8] DP: It's less a question of interacting with the user base (although it does mean that I've got a good understanding of the vocabulary and culture that people will be bringing to communications with us) and more of a question of knowing what sort of features and functions "fandom" (by which I mean fannish use cases) will find useful and relevant, because they're usually things that I've wanted myself for a really long time.

[7.9] The question of "what is fandom" is one that will never be resolved in our lifetimes. But for me, when I self-identify as fannish, I mean that I identify as part of the Western-media-based creative subculture, where people engage with source texts in some sort of transformative way. (I personally write in the Stargate SG-1 fandom.)

[7.10] MS: When I say I am not a fan, I mean that I have never been a part of the fannish culture. I've seen it from the outside. I've read some fic, I have a lot of friends who are in fandom, and I think it's interesting. I don't think I'll ever get into the culture as a contributor or avid consumer. Also, I don't believe I am very specific with my terminology. I know enough to know that fandom is a huge, diverse group with many, many distinct overlapping and nonoverlapping groups.

8. Free and open source software development

[8.1] Q: I'm curious whether you've seen or experienced any parallels or perpendiculars with the software development community and open source process, as compared to any sort of fandom.

[8.2] DP: The most obvious parallel is that both fandom and the open source software (OSS) community is composed of intensely passionate people who are incredibly dedicated to their particular interest. Other than that…I'm not really sure. I have this vague and unformed idea that it's the difference between male space and female space, but I'm not entirely convinced of it, and I'm not sure I can talk about it without having to resort to insane generalizations and talking out my ass.

[8.3] Q: Would you characterize Dreamwidth development as different from run-of-the-mill free software development?

[8.4] MS: It's vastly, vastly different—and that's a conscious decision. The biggest differences are in who contributes, and how those contributions are received. Traditionally, most OSS projects expect you to come in knowing what you are doing. You are expected to be a master programmer and understand how it all works. Meritocracy feels like a great idea when you look at the surface, but underneath, it turns it into a competition, not a collaboration.

[8.5] DP: I haven't had a lot of experience with OSS development on projects other than LJ, but what I have seen is always something that its proponents describe as meritocracy, and the outside observers view as a small clique. We think that everyone can contribute. There are so many different tasks that we have, development-wise, that need to get done. Some of them are huge and sweeping and do require a lot of experience; some of them are literally one-character changes.

[8.6] By encouraging and teaching people who are less experienced on those one-character changes, we not only free up more experienced developer time and effort for the big things, but we also get people learning as they go until they are those experienced developers. We're never going to run out of the little things. That kind of attention to the small details also gives us a chance to showcase our commitment to excellence and detail. We don't want to fall into the trap of saying, "That's a tiny thing. We can work around it until someone has time to fix it." There's never going to be a shortage of the huge issues, but that doesn't mean we can't pay attention to the tiny things too.

[8.7] Q: How have people used to traditional free software development reacted to your process? Does Dreamwidth interact with the larger OSS world?

[8.8] MS: They've been reacting very well. It's been surprising to me to see the people who have shown up in IRC. People like Skud (http://infotrope.net/blog/), eagle (http://www.eyrie.org/~eagle/), and adamk (http://ali.as/) are fairly well-known old-school OSS Perl types, and they have been interested in the DW project. As to the development process we're using, Skud has been really supportive and interested in it and is helping us out. I don't have a good sense of how the greater OSS community views our process yet, though.

[8.9] DP: One thing that I find really interesting is that there's been a conversation in the OSS world for years about how to interest or attract newcomers, and so a lot of people are seeing DW and our success and getting interested because of that. I know Skud's been doing a lot of advocacy work in that direction, and a lot of the attention we've been getting is due to her promotion.

[8.10] Q: Based on the assumption that if an old school OSS developer wants a blog, they've already installed a self-hosted solution, is it fair to say they're more interested in your process than your product?

[8.11] MS: Adamk in particular has been evaluating Dreamwidth as a potential new home for the use.perl.org (http://use.perl.org/) content. Although DW is not yet suitable for what they need, we've started talking about it. At this point, I think the interest is more on the product, but Skud is doing a lot to advocate for the process we're using and attracting more awareness to what Dreamwidth is bringing to the table for OSS.

[8.12] We have a ways to go. I don't want this to be construed as "hey, look, we've solved it!" because by no means is that the case, but we are fairly unusual in the world of development methodologies, and I hope that over the coming years, we can refine it and educate others on things that we have found to work.

[8.13] DP: I think one major benefit, in terms of the overall end goal, is that we don't view our way as having totally solved it. We know that we're going to be iterating as we go.

9. Personal identity

[9.1] Q: One of the big concerns of some parts of fandom is separating the fannish identity from the legal identity—that is, work or family responsibilities. Both of you are now self-employed or transitioning to self-employment, and you have drawn your significant others into working for the business, but I know you also both maintain separate staff and personal accounts, if not identities. Why are you keeping them separate? How willing are people to respect that separation?

[9.2] MS: I don't feel that either of us have a very strong separation of identity, honestly. On my "mark" account, I send them to my other identity; I say "hey, go read xb95."

[9.3] DP: For me, part of the reason for maintaining two separate accounts is just so that people have a sense of when we're speaking ex cathedra, so to speak. If people have an issue with the fact that half of the site ownership is writing porn in her spare time, they're probably not going to be the type of people who are going to be happy on DW in the first place.

[9.4] MS: I also think that maintaining the separate accounts will be useful for separation of work and personal time. Just like they suggest you maintain an entirely separate room of your house if you actually work from home, I think it's useful for me to be able to be logged in as xb95 and not see official journals, people sending me messages, responding to comments officially, and so on.

[9.5] DP: Yes, absolutely. I know that I can go to my alternate account and just see off-duty stuff. It was always really hard for me, when working for LJ, to load my LJ friends page and see people criticizing something that I'd done or something I was involved with, or just people snarking about LJ in general. Having two accounts helps me maintain as much of that emotional separation as I can. Obviously I'm still going to run into people critiquing my baby, but if I'm viewing it in an entirely separate account, it helps me make the internal transition.

10. Conclusion

[10.1] Q: Was there anything you wish I had asked about—anything you would like to let the readers of TWC know about Dreamwidth?

[10.2] DP: I think one of the things that I find most interesting about DW is that in a lot of ways, we're an experiment, business-wise. We're explicitly trying to overturn a lot of the things that "everybody knows" about business on the Internet. Our goal isn't to make it big and sell out to another company, or make an initial public offering for millions.

[10.3] Mark and I are both going into this viewing it as something that we want to support ourselves for as long as market conditions make that possible. So in many ways, that attitude and that idea is informing a lot of our choices. We're designing for the long haul. I see a lot of people saying that our business stance is idealistic or overly optimistic, that "of course" we'll sell out or plaster ads all over the site when economics make it necessary, and that perception is something I want to combat.

[10.4] We built everything on the assumption that we never, ever, ever want to take money from anyone but our users, because we both think that the minute you go to outside funding sources, you give up an element of control over the project that we're not comfortable giving up.

[10.5] MS: I want to say, Dreamwidth is not a side project, it's not a clone, it's not a fly-by-night little thing we thought up one afternoon. There have been thousands and thousands of person-hours put into this project. We are here to build a stable, solid foundation, a useful service, and a vibrant, varied, diverse community.

[10.6] This is our business, this is our careers, this is our baby. We know the costs, we know the technical requirements, and we did all of our business design based around the notion of paying ourselves a living wage for the rest of our careers. We're in this for the long haul.

11. Notes

1. Strikethrough was a 2007 incident in which many LiveJournal user accounts were deleted without warning. LiveJournal initially claimed that these users had violated the terms of service, but they later restored some accounts and admitted that their criteria for deletion were overbroad and not a reasonable interpretation of the terms of service then in force. (http://fanlore.org/wiki/Strikethrough)

2. The 1998 Digital Millennium Copyright Act (DMCA; http://thomas.loc.gov/cgi-bin/toGPO/http://frwebgate.access.gpo.gov/cgi-bin/getdoc.cgi?dbname=105_cong_public_laws&docid=f:publ304.105.pdf) is a United States law that governs the rights and responsibilities of Internet service providers in relationship to intellectual property holders.

3. "The Organization for Transformative Works (OTW) is a nonprofit organization established by fans to serve the interests of fans by providing access to and preserving the history of fanworks and fan culture in its myriad forms" (http://transformativeworks.org/about/believe). OTW sponsors the Journal of Transformative Works and Cultures.