When I was a computer science student in college, everyone I knew read Slashdot. Founded in 1997, the site was dedicated to "News for nerds. Stuff that matters." It featured about a dozen short, user-submitted stories about the technology news of the day. In most cases, the stories consisted of a short summary and a link to content elsewhere on the Web.
The site was created by Rob Malda, who was then an undergraduate at Hope College, a liberal arts college in Michigan. Malda ran Slashdot until 2011. Then, after taking a year off, he joined WaPo Labs, a technology incubator owned by the Washington Post Company, the parent company of the Washington Post. (WaPo Labs is not among the companies being purchased by Jeff Bezos.) I spoke to Malda last week. The following transcript has been edited for length and clarity.
Timothy B. Lee: For readers who aren't familiar with Slashdot, how did it get started?
Rob Malda: Slashdot started in 1997. It grew out of my personal blogging, but you didn't call it blogging in 1997 because blogging didn't exist then. I was writing about sci-fi movies and tech issues. I was an active open source developer, though it wasn't called open source back then. I was contributing to different packages. And Slashdot really grew out of mailing lists and IRC channels that I was participating in. I would find things I liked and I would write about them. Over a few years, I wrote a content-management system, and it turned into something a little bigger.
My earliest audience primarily came from the fact that I was an open source developer. I was participating in mailing lists. My personal Web site had thousands of readers because they were downloading software I had written. I was really active in X11 window managers [which provide the user interface on Linux]. I had screenshots of desktops, and I was making graphics and icons. I was an art as well as computer science student. My stuff looked interesting. I didn't intentionally pull people from IRC and mailing lists. I think they were already there.
So you didn't build a community from scratch. Rather, Slashdot grew out of your existing relationships in the open source world?
That's the thing. People ask me why were you successful. The thing I tell them is no bulls**t. I never had to fake what I was doing because this is just what I was doing. These are the people I was hanging out with. These are things I was doing for fun. I was completely embedded inside of it, because it was my hobby. It was what I did for fun. I don't understand how reporters for a newspaper can say "now you're in charge of France." That doesn't click with me. I would be terrible at [reporting on] France. I don't know anything about it. The tech stuff was mine. I lived it.
Was there a moment when you thought to yourself, 'this Slashdot thing just might take off'?
My joke answer is I still wasn't sure about it up until about five years ago. There was a moment, and I don't think this was the moment that I realized that it was financially viable, but there was a moment that I had realized it was something that had grown well beyond my control.
One night we were sitting in my living room, myself and several other Slashdot folks. And we were tailing apache logs and we were doing DNS lookups on the IPs [e.g. monitoring who was visiting Slashdot.] And we said "Oh my God, someone from Microsoft is reading." "Oh my God, someone from Stanford is reading." "Oh my God, here's someone in France." As fast as I'm doing DNS lookups, people are coming from every continent. We had a really active user from the South Pole for a time. That was a really jaw-on-the-ground moment. We thought we were just a bunch of people screwing around. It's one thing to have tens of thousands of visitors. It's another thing to realize they're everywhere.
Slashdot was one of the first sites that had to deal with large-scale user-submitted comments. What did you learn about how to build a healthy online community?
I did not premeditate anything. I just winged it, and when I saw problems I triaged them. The Slashdot system was very much evolved and not designed.
One thing I learned is don't spend your entire life playing predictive defense against attacks that will never happen. Real people are very clever. If they choose to attack you they'll attack you in ways you can't predict.
People get bored fairly quickly, but the rare person never gets bored ever, and you have to be very careful of that person because they might be your best user or they might be a terrible person. And if you give them too much control they could do bad things.
Most people are just purely passive consumers, and that's okay. I think a lot of people in this space obsess too much over trying to convert passives into actives. Some people are just wired to be that kind of person. Others aren't. That's okay.
Are there new aggregators that have taken the place Slashdot once occupied?
Hacker News is awesome. It is probably my number one RSS feed right now. But it's getting too big. They've crossed the line that reddit crossed a long time ago. The value on reddit now is on the subreddits, not on the main page. Hacker News is wobbling on that space now. If I could just find someone who made a Hacker News digest, with the 10 best items from Hacker News, that would be a really good Slashdot.
Did Slashdot cross that line?
I'm the wrong person to ask that. I was too in the middle of it to look at it objectively. I tend to look at Slashdot not so much from the mix of stories that were on the site from one day to the next and rather with the big milestones in the site and how I felt about the site during that era. And for me, the sun set somewhere between the OMG Ponies April Fools Day [April 1, 2006, when the Slashdot theme color was changed to pink and the "news for nerds" slogan" was changed to "OMG!!! Ponies!!!"] and the 10-year anniversary [in October 2007].
What changes about an aggregation site to make it lose its early greatness?
To me, it's really about volume. [When starting a site] your first task is to just get a critical mass of content. When you have a really small group of people, that's really tough. But the small group has shared themes, shared interests. The output is really interesting. Slashdot in the early days was very focused on small subject matters--not particularly diverse. As the community gets bigger, there's a natural dilution of that original audience, but the amount of content increases, so then it's less an issue of acquiring information and more about editing it.
This is why Hacker News would be better if I could get the 10 best items. I don't care about a new version of CoffeeScript, but when tech culture stuff happens, the right story is there and it's fast. But I have to click through a lot of stories that aren't relevant to me.
With Slashdot, I was able to control that to a certain extent because the home page was rate-limited to humans making the call about what goes on that page. reddit can have 200 items on the home page. Slashdot we kept bounded to 10 or 15 [posts per day in the early years] and by the end we kept it bounded by 20 or 25. At some point that 31st story isn't really that much better. With more and more voices, you tend toward broader subjects. Eventually it becomes less and less interesting.
But what's interesting to [me] isn't always the thing that's most lucrative. I had a pretty strong belief that the population of Slashdot users was relatively fixed. That makes for a crappy business model. You want to have a million really good users and be done? How does make the chart go up and to the right?
Do you see any sites that have the potential to be the next Hacker News?
I don't think it's going to work that way any more. I think that the power has decentralized. Successful people on Twitter basically can fulfill a lot of that same role. You can follow Tim O'Reilly and Robert Scoble and Tim Lee and you can get a pretty good summary of what's happening around the universe.
But I think Twitter is fundamentally broken in certain regards. You get a lot of mediocre stuff along with that. Everyone thinks they're interesting. Part of the value of Slashdot was I could rate-limit it down to a few things a day, and still leave you with a sense you have pretty good sense of what was happening during the day. But if you follow 50 people, that is an epic time commitment.
So I'm not sure that there is a clear heir apparent. There's power in what reddit does, and what Hacker News does and to a lesser extent what sites like Slashdot still do. But the power has shifted to individuals. My concern is still that the tools aren't really there yet to make that as good of an experience. I don't feel like I get as good an experience out of Twitter as I got out of a good mailing list and an IRC chat room and Slashdot 15 years ago. It's different, chattier, I get more pictures of what people ate. It doesn't necessarily bring people insight.
What tech-related topics are you most obsessed with right now?
Bitcoin. Not a lot has been happening on Bitcoin lately.
[I'm interested in Bradley] Manning. I'm not interested in the verdict part. That was the aftermath. The interesting part already happened. The WikiLeaks thing, the giant cache of documents. All of that happened. [The verdict is] just cleanup.
I guess what matters to me in the Manning case is: will that influence the next person? Because we have [Edward] Snowden who did a vaguely similar thing. It remains to be seen what charges he'll ever face. If your only option for leaking government secrets--what did Manning get, 5 counts or something, and Snowden can never come back to the country again--that sets the bounds of what the punishment is for leaking.
The stuff that's interesting has as much to do with the technology as the information. I'm interested in the technology the government uses to spy on me. I'm interested in the fact of the anonymous file-sharing network that made the Manning stuff possible. That's the stuff that gets to my soft, gooey center. The policy parts, I don't feel like I have a say in that. I don't have a voice there. I know what I want to see happen. But I don't feel like I have a say or a voice so I choose to be interested in the technology and think about where that's going to take us next. All that stuff never would have existed 20 years ago.
A lot of people who are interested in computers and the Internet—the kind of people who read Slashdot—have a pretty strong civil libertarian bent. Any idea why that is?
I don't know that it's interested in computers specifically. There's a whole slew of overlapping factors. College-educated folks tend to have backgrounds in computers, so they tend to be slightly left-leaning.
A lot of the Slashdot readers came out of the open source/free software movement. There's this core kernel of folks, that "information wants to be free" nucleus of the free software movement. They are naturally fearful of authority. It comes out of this natural sense that Microsoft can't be trusted because their closed binaries control our destiny. It's easy to see someone who's fearful of Microsoft would be fearful of the government.
There was a lot of libertarian leaning in Slashdot, more so than the global population. I think a big piece of it is that the hacker ethos, the hacker spirit is the natural desire to tinker, take apart and take control. Anything that hampers that becomes the enemy. Which is why [geeks care about] copyrights and patents, that kind of stuff. "What do you mean I can't take apart my car engine? That's insane."
"What I can't reverse-engineer my CPU? I can't decompile a binary?" That runs against the grain of these people. That means you have to start taking steps backwards. That leads you to corporations and governments. That leads you to naturally distrust everyone.
How did you come to the Washington Post and what are you working on here?
Somebody posted my resume to Hacker News and I spent the following month doing about 50-odd interviews with 40-odd companies. It was really cool because I got to see what a lot of interesting startups were into and talked to some big companies that I'd been writing about for a long, long time.
But the Post had a number of things that I was very interested in being able to play with. The Labs team that I work for is a small group of engineers that has relative autonomy in the company, relative to the newspaper. We're not owned by the newspaper, we're owned by the parent company. Our charter is different. Our charter is interesting: take a big swing at the future of news. To that end, we have built interesting technology. The people here are all interested in the same things I am, which is the future of news and using interesting technologies and communities to address that problem. I found that a lot of the ideas being explored here are similar enough to the ideas I had that I'd be able to fit in nicely and not have to go hire 20 engineers of my own to go build something.
My current project is a redesign of social reader. The original version of it was conceived as a Facebook app that stood on top of a fairly popular tech stack that we'd been building for a few years. Over the previous year, Facebook throttled the "frictionless sharing" feature that they had rolled out that the Social Reader platform was originally designed for. I'm currently working on re-conceiving the product to have its own soft of natural community and not be wholly dependent on the existence of Facebook to drive it.
If you want to have a social news product, you need to be involved in every step of the loop, the aggregation of content, the curation of content, the sharing. You need to be involved in every step and originally Facebook was the primary controller of sharing, which meant we were powerless to control our own fate.