Level Up Your Life

Subscribe

I’m Anil Dash, And This Is How I Work

Nick Douglas Avatar
Nick Douglas

Glitch wants to do for web apps what YouTube did for videos, Flickr and Imgur did for pictures, and Twitter did for bad opinions: make them easy to create, share, and embed, and give them the chance to reach the right audience or even go viral. I talked to CEO Anil Dash about Glitch, his career, his internet fame, and how and why to run a socially responsible company. It’s a microhistory of how the internet took over the world, how the world ruined the internet, and how we get both back.

Take me through your background and how you got to here.

I grew up in rural Pennsylvania and in junior high I started making grading spreadsheets for my neighbour, who was a professor at the community college. In the 90s, if you knew how to plug in a printer, you could get a job. It was this or flipping burgers. A guy hired me to help set up networks and printers. I started a company the day after I graduated high school, he became my co-founder. That was how I got into tech.

We were a “What do you need? We got it” [kind of shop]. Some places, I was like installing Windows 95 for people. Some places, writing custom software to help update systems. I’d been around construction [work], so we built software for construction teams.

My parents had a house that they would rent out when I was a kid. And the tenants would invariably trash it. So I got to do a ton of construction as a kid. I would do everything, hanging siding, putting up drywall, doing plumbing and electrical and HVAC. So I became kind of handy. And my business partner had been a reseller for a construction accounting app, which is the least glamorous thing in the world. But it was fascinating to me, it was how I learned business. He was a great salesman, still is, and he gave me sales training at 17. It blew my mind, it [wasn’t] the stereotype of the sketchy used car salesman, it was problem solving. It wasn’t persuasion at all.

I would work with these construction companies. When I was done installing a printer driver or whatever, they’d be like, can you swing a hammer? You’re skinny, can you run through the docks and pull some cable? It was this weird hybrid thing, but it was great. I had a roster of customers in four states.

So I did that for seven years. I learned to run a business. I learned that doing tax paperwork stresses me the fuck out. That was college for me.

[One] company had not made the transition to Windows, everyone was going to Windows 95. I pitched a deal for this software porting thing, four months, 25 grand. They said yes. I [felt like] I had just hit the jackpot. I’d put on the suit, gone there, pitched them and got it. That was the dream.

I started having clients in New York, spending more time in the city, and I’d always wanted to be here. The summer I was 21, I moved up here. There may have been a girl involved. I didn’t know anybody and the company wasn’t doing great. I was not some master of this stuff. Then it was brutal. I got a sublet in the Spanish Harlem, was pretty much broke, the company was going sideways.

I started applying to jobs. Dot coms had started arriving. There was a startup doing online video streaming for music videos. It was a farcical title, but I was CTO. Most of my day was transcoding videos into Quicktime, Windows Media Player, and RealPlayer. Building systems to do that. We would go to record labels and get their videos online. At that time the clips were like 20 seconds long. You couldn’t put a whole video on dialup. But I built a pretty good system, and I taught myself to code in PHP.

It’s funny, but it was a video portal like YouTube. Like here’s a bunch of video clips. They weren’t the whole video, and you had to have the player, the experience wasn’t great. But it was super, super instructive. The place had been a video production house before. They’d pivoted to dot-com world because everybody was. So it was wildly mismanaged, but it was very educational. I was living in Manhattan, making 26 thousand a year. Not a good way to live. But I was here and I was making a go of it.

BMG Records was one of the big players at the time, and their office was right at Times Square, overlooking the square. We’re on the top floor in this big conference room. I got to go to the meeting with the bigwigs, but I’m the kid, obviously, so I’m in the back of the room just sitting quiet. In the end they asked for questions. So I asked, what are you guys doing about Napster? He said what’s Napster?

In that moment I was like oh my god, they don’t know about the internet. We’re fucked. I’ve got to get out of here. And the place imploded anyway.

I talked my way into a gig at the Village Voice. There’s a Prince song on his 1999 album called “All the Critics Love You in New York.” I’d taken it to be about the Village Voice. So I saw the banner in front of the Voice office at Cooper Square and I hustled my way in and I was a web developer. I got to work with editorial a lot, the people reviewing the events. Like what do you need if you need to say who’s playing Webster Hall? How would I list that they’ll play every Thursday for three weeks but not on odd weeks? The technical stuff was really interesting to me.

The second day I was there, Craigslist launched in New York.

One of my primary responsibilities was the apartment listings, which was how they made most of their money. I didn’t realise what a breach in protocol this was at the time, but I went to my boss’s boss and was like, what are you going to do about Craig? And she was like, who’s Craig?

I [thought] I’m sure they’ve got this. They’ll figure it out. So I kept my head down. Two months later was 9/11. So I was like, I’m not going anywhere, I’m not messing with this. I hunkered down.

In the next month, classified listings dropped [about] 70 per cent. [They blamed it on] the attacks. I was like, I think people still want apartments! That month, my friends launched [blog publishing system] Movable Type. I’d been blogging.

How did you know them?

The blogging world was small. You could just see anybody who was doing it. I’d started in ’99. [Twitter, Medium, and Blogger co-founder] Ev Williams asked me to maintain a featured list of cool blogs on the Blogger site. So I was looking all the time at new blogs. I saw a site called Dollar Short, which [Movable Type co-founder] Mena Trott had made. It was incredible writing, incredible design.

And I beta-tested [Movable Type] as a developer. It came out and it was a hit. I was like, you guys are not charging enough, you’re going to run yourselves ragged. I remember going home from the Voice at the end of the day and making spreadsheets, like here’s how to charge for this, here’s how many users…I just informally started helping them.

A couple months later, Nick Denton reached out. Said “I’m building a gossip blog and I want to show something to you.” He’d asked Mena to build Gawker. [He] got [designer and blogger] Jason Kottke to make the logo. Mena did all of Gizmodo, and Jason did the design of Gawker.

We met at KGB bar on Fourth. Nick had this IBM laptop that was like three inches thick, dim screen. Nobody would take a laptop out in a bar back then. He had all these mockups on it and he clicked around. He had like a secret URL for Gawker.

[At the Voice] I went down to editorial, and I found [gossip columnist] Michael Musto. I showed him Gawker. He wasn’t rude, but he’s like, yes, I’ve been doing this shit for years. I’m like, they’re going to update like five times a day. He’s like, “OK.”

Well that didn’t go like I thought it would! I went back upstairs, embarrassed. Like aw shit, now I’m not cool with the cool guy. He was very polite about it but he was disinterested, like so what? I think I’d gone deep on explaining what a weblog was, and I hadn’t hit the cultural significance of it.

Around that time, there was a MetaFilter meetup on Bowery. Nick was there and [Gawker’s eventual founding editor] Liz Spiers was there. She’d been doing an economic blog. And I introduced them.

I started working on Movable Type because I’d gotten laid off at the Voice. I knew they were trying to get funded, and none of us knew what venture capital was. They started paying me out of their pockets. I helped them put together their pitch, and they raised money from Joi Ito.

That A round was $US600,000 ($874,056) on a post-money valuation of $US1.8 ($3) million. (Our Glitch round last fall was $US30 ($44) million.) And I thought 600 grand was all the money in the world. I think we immediately sent $US400,000 ($582,704) of it to Dell to buy servers. There was no Amazon Web Services. So we didn’t see any of that. The company was already cash flow positive, it had to be, because the funding wouldn’t have done jack shit.

I was there at [Movable Type owner] Six Apart for seven years, and it was wild. I [had moved to] San Francisco. Obama had just come into office. I ended up building a nonprofit researching how social media would impact politics. The MacArthur Foundation gave us half a million dollars. I started working with [Lifehacker founding editor] Gina Trapani again. Nick had said he was looking for a developer for [Gawker Media platform] Kinja, so I’d introduced them.

But we designed a nonprofit, Gina was leading tech and I was leading engagement. We went to the White House, the State Department, everywhere, to show social media was going to affect policy. At the White House they were like, we can’t go on Facebook because we have Internet Explorer 6. It was very Wild West.

The open source project took off, so we turned it into a startup, built a crowdfunded campaign. At the same time I had founded Activate, a consultancy. I’d never imagined doing management consulting, but the idea that it was predicated on the concept of competency, that you’re here because you know your shit, was exciting to me. I learned a lot about the rigour of the analysis, and these MBAs who could break things down into their components.

We worked with Dropbox, you name it. Did a ton of work with Condé Nast. And that echo was there: When we brought back Gourmet Magazine, we had Paul Ford there, we had Liz Spiers there.

Doing that in tandem with a startup, I’m pretty good but I’m not that good. The death knell was, we were dependent on the platforms to give us access to data. Facebook, Twitter, Instagram. They never said no, they would just string us along. “We’re going to give you that API back in three months, two months.” We couldn’t scale up the business if features are going away.

We pivoted to Makerbase, which was IMDb for apps. It was cool, I wish it existed, people still wish it existed. But Gina was burnt out. And she went to work with Paul at Postlight. In April 2016 on a Monday, we talked and said [the company] is done. On Tuesday I emailed our investors. I thought the worst [reaction] would be they’d say “You fuckup, why did you lose our money?”

On Wednesday I found out the worst is, you get an auto-response that says, “Thanks for your message. I’m not looking at my email right now. If it’s important, call my admin.” Oh, they actually don’t even give a shit about this thing that I’ve been losing my mind over. Thursday of that week, Prince died.

I was like…“I’m not enjoying this.”

Two weeks later Joel Spolsky called, said I’ve got something to show you. The prior two times he’d said that it was Trello and Stack Overflow. I came by the office, I saw the prototype of Glitch. It looked cool as hell. I said, “You need to do blah blah blah,” this list. “You have to be able to remix it. Not fork, remix.”

Joel was like, “You should come run this thing.” I was like, “I just killed a startup, I’m not doing this again. No way.” So six months later I joined.

Joel and Michael were so busy running the other companies. And in the span of a year or two we moved the headquarters, sold off the business that had been the bread and butter for the prior 17 years, including a third of the company going with it, which was very emotional.

We renamed the product Glitch, renamed the company Glitch, and then raised the largest consumer A round of funding in New York City last year. That was our first year and a half. And while we did that, Glitch took off and people built the first million apps. And we took the team from essentially 100 per cent homogenous to as representative as we could get it.

Let’s talk about what you’re known for now. You’re famously the tech guy who loves Prince, you write, you’re politically involved.

My public persona is somewhat incoherent. Most people who know me, know me because of social media. There’s one way of performing being a public intellectual. Or a writer. Which is closest to what people know me as.

And then I also am a weird pop-culture obsessive. People know I like Prince. I think by normal standards, I would be a huge Star Wars fan. I would be a huge—until recently—Michael Jackson fan.

And then the last piece is I am the CEO of a very successful startup that raised 30 million dollars and has dozens of employees and millions of users. That’s the part I’m least known for!

People are surprised to see me in the room when I’m in a roomful of activists. People are still surprised to see me in the room at a tech conference! They’re like “I see you talking shit on Twitter and also calling out tech leaders and VCs, so how could you get invited?”

I’ve written more visible stuff for a longer time than a lot of people that identify as writers. But nobody introduces me as a writer. Which is moderately frustrating. Not for nothing, but I had a column in Wired.

I’ve spent twenty years being pissed off that [Postlight co-founder] Paul Ford wrote a thing before I did. And not just that he got there, but he’s a better writer than me. But that’s a writer. Paul is a writer, I’m a person who writes. But he’s the CEO of a company too!

Let’s talk about Glitch. We live in this time when all the well-known startups, we all agree they’re basically bad for the world, and if we use them, we’re trapped. There’s a correlation between wanting to be a big startup, and doing shady things.

Even intentionally destructive [things]. What, say, Uber does to the labour market is egregious and intentional.

And people, even those who know him, can openly say the CEO of Twitter is terrible and enabling Nazis—even as we use the product because there’s no other game in town.

This is the reckoning in tech, which is that we have a narrative that everybody big or successful or visible, is corrupt and complicit. And yet.

Glitch used to be Fog Creek Software. Over the years they’d created Stack Overflow and Trello. A billion dollars’ worth of companies had been spun out of this company even before it became Glitch. Trello exited for $US425 ($619) million. They’d only take a tiny bit of funding before that, so it was a very successful exit. They sold to Atlassian. Those guys are pretty thoughtful, too. I mean, I don’t love using Jira, but that’s not an ethical failure. And I do love using Trello.

And all of those companies collectively, the ones I just rattled off, is probably $US50 ($73)-100 billion in value. And half a dozen different CEOs, many of them straight cis male white men, all of whom have made public statements about what they’re going to allow in their platforms, what they’re going to enable, how they’re gonna make their money. All their business models are “We sell goods and services in exchange for money,” not “We secretly broker your shit behind your back, steal your data.” But they’re not the narrative. Uber is. And the worst parts of Facebook and Twitter or whatever are.

Why isn’t the narrative that Slack is a success, be a success like Slack? Or Trello, GitHub—companies are very big, very influential, and are not predicated on being destructive.

It literally constrains my opportunities to lead as the CEO of a tech company. More tech events want me to come and say “Facebook bad!” than want me to come and talk about Glitch. Even though there’s like 3 million apps on Glitch. There’s like 50 per cent more apps on Glitch than on Apple’s app store.

Some of that is, infrastructure is never gonna be as cool as a consumer facing app. That’s part of why I have this bias towards making Glitch a consumer app as much as it is a development tool for businesses. The industry needs, the world needs, an example of a tech company that is…our business model for Glitch is very straightforward. It’s very, very easy to make apps on Glitch. So use it at work to make apps. And then pay for it. Everybody’s like, that makes sense.

And then we say, by the way, since it makes it easy to make apps, there’s a consumer part, too. You want to make apps for your friends or for a funny art project or because you’re an activist or whatever. They’re all there, too. Nobody is confused by that.

I’m trying to broker my own visibility into building a narrative not just for Glitch, but for all of us who were optimistic about tech because it could do good things. It is possible to build things that are just the good parts. And I think that’s why people build on Glitch.

People don’t worry about being harassed here. They don’t worry about abuse. And they’re like “Every single day I find something kind of makes me laugh or it’s technically interesting.”

This app that went viral this week, by a young woman—I believe she’s Ukrainian—it uses computer vision and your camera on your computer, and if you’re slouching in your seat, [your screen] gets blurrier. It forces you to sit up straight [to keep] your screen clear.

It’s a gimmick thing, but it’s cool it’s possible! She’s using cutting edge A.I. stuff from Google. And because it’s a Glitch app, it’s all open source and it’s visible. So you know the data’s not going somewhere else. You know she’s not capturing your information and misusing it.

And that’s true of every Glitch app, they’re open source?

Yeah. It’s possible to close them up, but nobody ever does. I expect when we get more businesses using it, they’re going to, just for the sake of privacy, [lock] them down. But by default they’re all open source and they’re all remixable. But the cool thing is, people like to look behind the scenes, like, how does this work?

I am not a machine learning expert, I don’t know computer vision very well, but I can go in there and see oh, that’s how that works. And this is how you train the system to do these things. And this is how you would make the screen fuzzy because they’re slouching in their chair.

[That app] was in the Daily News, Daily Mail, Vice, all over the place. And [the developer] is a young woman, I’d guess in her early 20s. And all over the world, people are responding to this app. That’s gotta feel good. There is [something] pure about “I made something cool, and I want you to check it out.”

I remember back in the day I would go to Fark.com or MetaFilter, and there would just be cool stuff on the web. And it felt good! During my lunch break, to see a distraction and feel good about it and not be depressed by it, and also not feel like it’s just surveilling me and trying to steal my data. People want that. I want that. It’s not a retro throwback thing. I’m not like “We’re going to bring back ancient internet!” It’s just, that was the last time we had this.

Regarding visibility, I keep running into products made on Glitch. Are you able to see how these things first spread? Are they bubbling up through some social features on Glitch itself before they break out?

For the stage we’re at, it is primarily about what in the olden days we’d call viral spread, where apps take off in their communities.

I can talk about a “stop slouching in your chair” app and that’s a very mainstream thing. That also happens for developers: “This one JavaScript framework is the thing I love, and that cool demo is on Glitch.” Those are a little more within the community, because there are a lot of coders hanging out.

There isn’t this giant dichotomy between people who know some technology, and people who have something to share with the world. We found, especially amongst our core community, there’s a ton of people building music apps around this. They’re a coder by day at Google, they go home and they have something they want to express.

They’re making tons of web VR art. They’re like, “I’ve got this VR rig for my gaming system, but I want to make something expressive.” So Glitch is probably the largest community of VR and XR content in the world. We’re not trying to build YouTube for VR, but it’s happening on its own.

The phase we’re in is sort of YouTube circa Lazy Sunday. Nobody was typing YouTube.com into their browser. They were reading Gawker or whatever site, and it had Lazy Sunday embedded in it. And they’re like, OK, this is how I play videos now.

People are finding Glitch because there’s a data visualisation for an article that happens to be an embedded Glitch app. And if you’re nerdy, you’re like let me see that data about this, and you’ll click and you’ll be in Glitch. Somebody reading Reddit one day or on a news article, and ends up at a place for people to create with code.

You can definitely imagine creating a video, because you see YouTube videos embedded everywhere. Sometimes you’ll read an article and there’s a SoundCloud embed of a podcast about that thing. That’s mostly the space we’re in.

We are building more professional tools, for when you want to have your whole team collaborate on building something. You want to sign it with your Google account, all the stuff you’d expect. That’s the business we’re building. And I would be shocked if a year from now, that’s not the majority of our company attention. But they build each other. There’s no tension between getting business users on and building the community in a healthy way.

Some of the stuff that takes off so prosaic, and you realise just how frustrated people get by how complex technology is. There’s a timer app. It’s literally like, set a countdown timer for 15 minutes. But it’s beautiful and it’s full screen. I think it was made by a teacher.

Classrooms, like my kid’s classroom, have a big smart board at the front of the classroom. It makes me nuts. His teacher would google “countdown timer,” find the first result. It was surrounded in ads. You have a roomful of seven-year-olds staring at ads on a browser all day.

That creeps me out.

It’s terrifying. And it’s normal to them, so they don’t realise it. But as somebody who knows how this happened to the world, it’s devastating.

Anyway, a teacher was like, that’s sucks, let’s just do this. It’ll have nothing but the time, and it works on a touch screen, all that stuff. And then people start using it at work. “Oh, we’re going to set timers for our meetings so they end on time. This is a horizontal need. Everybody needs this sometimes. It is not pushing the boundaries of computer science. It just looks nice and solves a problem. That’s alongside the apps using cutting-edge machine learning and AI.

I look at people solving problems their I.T. department will never solve. And we had that happen internally. We have a company handbook that’s public, and we publish all our policies there. And it’s, of course, a Glitch app.

We want meetings to be inclusive, so we don’t just force the most junior person on the team, or a woman on the team, to take notes every time. Fortunately, because we have a good gender balance here, there’s very seldom meetings with only one woman in the room. But we had the natural impulse of a bunch of nerds like us: Let’s make an app that randomly shuffles all the people in the room and picks somebody to make notes.

It keeps track that the same person doesn’t get picked three times in a row, all that stuff. Then we put it out there. Again, picking a random name from a list is like the first thing you would learn to code. This is not groundbreaking computer science innovation. It’s just a useful app that solves a problem.

And I can see every team that uses Glitch at work using that app. Because if you’re the person who gets stuck [taking notes], you’re so incentivized. And if you’re the person that wants to be good to your co-workers, you’re very incentivized. And it takes away the social friction. [You don’t have to] say, “I don’t want to do this in this meeting.”

We always talk about the unintended consequences of technologies, how they change how we interact, often for the worse. It’s nice to reduce the friction of being nice.

It only take a small nudge to free people from a lot of stress, and tech used to be great at that. It still is sometimes. And I think people have the impulse to create it.

But that app will never be a business. Nobody can build a startup around reminding people whose turn is to take notes. So the industry didn’t have a place to put that capability. Even if you made it for your own company, how are you going to share that with somebody else? You’re not going to be like “Find it on GitHub.”

Or you’d make an app and slowly realise it costs money to keep it in the Apple store. And therefore join an ad network, and now you serve tons of ads…

Exactly. There’s a million different points of friction. There isn’t a way to share culture through technology between communities or companies. And that’s actually what Glitch is doing.

My neighbourhood group wants to clean up the park, and I don’t want Facebook to be [where we organise that]. So I made a little mailing list app. Somebody could take that who’s organising in Iowa because for Elizabeth Warren, and they don’t want to depend on Facebook, but they can use the same app, remix it, change the logo.

So right now, there aren’t arseholes on the platform. How much is that already because of active choices, and how much is it that they haven’t arrived yet?

It’s both. For people who want to transgress, there’s a path of escalation. The first wave is, there’s people having a bad day. That’s always there. Someone’s rude because [their] dog died.

Then there are [people going,] “let me see what I can get away with.” They’re not even malicious. They’re just like, you know, teenage boys. They’re people being performative about transgression. Generally, these things are actually not that harmful or toxic.

But they can shade into, the form of transgression I want to perform is using this slur or this misbehavior. And if that’s allowed to flourish, that opens the door for the intentional harmers. It sets a tone. Somebody is using a hateful slur because they’re a teen who’s trying to provoke. And some else like “This is a word I believe in using.” Those are indistinguishable in a naive system. So you have to preclude it.

I think of when I was teenage dirtbag going to diners with my friends at 2 in the morning. And the rowdiest guy in the group would belch real loud. And then we were all doing it. And then you get kicked out. I don’t think anybody was ever like, why did you kick us out? They were like, you guys are being arseholes. And depending on how they did it, generally it de-escalated. And that’s actually the biggest thing we [at Glitch] do. I do a lot of what I think of as, “Come on guys.”

Community management online [is] the same as managing a physical space. You look at, what would you do in a park? What would you do at a party if people are getting heated up or somebody had one too many and is being belligerent?

In my early twenties, somebody would be like, “Hey, man, you gotta get the hell out of here.” And it would escalate. And a drunk guy is not going to back down. And then there’s my late twenties. Somebody’s like, “I think you had one too many. Let me take your keys. We’ll get you a ride home. It’s not your fault.” And that’s the maturity that hopefully you get.

Every major tech platform is not even at the beginning stage yet. They’re at the like, “Why don’t we let these drunk guys run the party?” And they are just now arriving at, “Hey bro, that’s not cool!” And it’s exacerbating it. It’s becoming a proof point. Part of it is, the alt right and their cohort are gaming the refs. “We’re being censored. I have as much right to be at this party and be drunk as anybody!” No you don’t! It’s my fucking house!

So we have [an approach of] “Look man. That’s not what we do here.” That’s our community moderation style. And what it’s done is at the scale of millions of apps, precluded almost every kind of common abuse. Because people are like, “We’re not gonna win there.” And I think that’s going to scale.

It’s a lot of work. You have to have a temperament about how we flag things and how we communicate, and that there aren’t templated messages. “Your app has been reported for violating rule twenty six dash four.” We’re like, “Hey, saw this thing. We don’t do that here.” Almost literally those words. “Your app, the description of it has this term, we don’t use that here.”

I think I’ve had two things escalated to me as CEO, where somebody was being transgressive in some way. There was a guy, I believe in Turkey. And his avatar was Hitler on his Glitch profile. I think he’d signed in with one of his other accounts and it brought his profile picture over. I am not fluent in what kind of shitposting a Hitler avatar represents in Turkey! I don’t care to be. So I was like, yeah, we don’t do that here. I think we hit him up at some other platform and said hey, please don’t do that. He just took it off. There wasn’t a dialog.

There was something else like that, a message about self harm. It was supposed to be ironic, it was bad, we didn’t like it. And the creator of it turned out to be a young woman. She was like, oh, I was being shitty to my friend and we were joking around. I didn’t think anybody would see it. I’m sorry. She said “I’m sorry!” And she deleted the app.

And by doing that a couple times and being proactive, it stops the cycle, the rapid metastasization that happens with this sort of toxic stuff. So we don’t have to police a mob of people.

And also, we haven’t built a ton of social features. We haven’t built [a way to] have a private group where you can message each other to do these horrible things. I think the only groups that will have those kinds of features will be paid teams. The companies that are using it right now are Google and Microsoft and big, big companies. They have their own issues. But that’s not going to be one of them. They’re not going to build widespread hate apps.

Those things are all intentional design. And I’ve been doing this for 20 years. I know how we fucked it up on LiveJournal back in the day. I’m on the board of Stack Overflow, and it wasn’t as welcoming a community as it should have been. It wasn’t fundamentally broken, people were able to get help. And that freeing of information was really powerful. But the first experience was people are jerks, because it turns out we built a game that rewarded gatekeeping. So we [had to] change the game. They’re doing that. I wish they’d done it faster, but the team did it and owned up to it. And the community is on board with the shift. That taught us, let’s not even get to that point, [let’s] have nothing to fix. It’s totally doable. It is 100 per cent doable. And at the scale of millions and millions of apps.

I think about Apple’s app store versus Google Play. Apple sets a tone. And I would disagree with some of their choices. But they actually don’t get a ton of transgressive apps submitted, because people know they’re looking.

Now, I don’t like the submission model. I think that is not compatible with the open web and with activism and journalism. And in fact there are acts of journalism that Apple banned from its store. But we can set policies that are clear enough around what we don’t allow in terms of harassment and targeting, and not have to filter the content.

Part of the reason I don’t want the approval model… So there are adult toys that are internet connected. The majority of those sex toys are proprietary, terrible apps, [just] like bad smart-home devices. So there’s a movement of open source, open access sex toys. And they’re like, Glitch is perfect for this.

What sprung up was a conversation between one of the lead developers in that community and a teacher, who was like, more power to you. That’s great. But I’ve got grade schoolers learning to code on Glitch. And they can’t search for butts and find your smart butt plug. Because they’re going to search for butts.

It was great to watch this conversation play out between these two people who were both invested. One, they believed in Glitch. But also, they’re both like, I love and respect what you’re doing. Very mutually, “What you’re doing should exist in the world.” So how do we solve this? And the conclusion really quickly came to this feature we’re building.

Everybody’s conception was like, oh, make a “not safe for work” flag. [But] that is not what this is about. Neither grade school nor butt plugs are work. What I want us to articulate is “Please don’t show this to kids.” That can be because you’re doing journalism on police violence. There’s a million reasons.

In vetting this with users, they were like “I love that you’re letting me articulate my intent for my creation.” Not “I’m adjudicating as a platform that you don’t get to be a work tool.” I do want people on their lunch break at work to read about police violence and journalism that’s happening on Glitch. And also, sex work is work.

Of course, immediately people were like, what if somebody was being a jerk and put something terrible up and doesn’t flag it “please don’t show this to kids”? Well then they’ve expressed their intent to us. That’s very easy for me to moderate!

So by designing this system to allow a creator to express intent, now we’ve got a dialog. Now we’ve got a community, not a platform where people can inflict themselves on upon each other. You can talk about what you were trying to do, and we can respond.

Nobody’s having that dialog about toxicity on Facebook, because who’s gonna believe that Zuck’s going to listen to them and see it through into the product?

It seems like a fundamental way to make sure it can scale later. You’re figuring out a lot now, before you suddenly need to reroute millions of people doing something that is now not going to fly.

You cannot undo. You can only anticipate. And we won’t catch everything. I want to reduce the surface area of harm so much that we can spend our time on catching the things that are unanticipated or unprecedented. But everything that is a known issue, shame on us if we don’t handle that.

A while ago I was talking to a person who runs a very large, very visible platform. And I said, oh, this mutual acquaintance of ours was doxed for their work. And they were like, what’s doxing? Same conversation I mentioned swatting. And they didn’t know what swatting was. The isolation and the lack of accountability is so extreme.

I think the biggest thing I have going for me is an advantage as a leader is I’ve been targeted, and I still get targeted. And so for my personal safety, wellness and taking care of my family, I have to be fluent in vectors of attack. Multiple senior leaders on our team have been in that position and are fluent in it.

If we said “We don’t have anybody on our team familiar with security vulnerabilities,” you’d be like “That’s batshit, they’re doomed.” But if you say “We don’t have anybody who’s familiar with massive major cultural vulnerabilities,” that’s routine. Not here.

I’m really interested in the idea of a public handbook, and how that’s helped you build a better team. At our publisher G/O Media, our union contract is public.

I grew up in a union household and I worked in the trades and construction for years. And it’s very simple when you know how comp works, that things are more fair and people who are on the margins are better taken care of. My mother, as a woman of colour in a town where there were basically no people of colour, working a job where she was protected by a union, transformed our lives.

The company started as Fog Creek in 2000. Joel Spolsky was our founding CEO, and Michael Pryor was his co-founder. And from day one, [they were] super transparent. Joel made his name by blogging about how they were running the company.

Very early on, Joel made a salary ladder. Here’s how compensation works and how you can advance. Michael Pryor, the other co-founder, he became CEO of Trello. And he was the first, like, white cis male CEO in tech that ever reach out to me about how to do inclusion training for his team. I was like, so Michael, what [incident] happened? Like, why did you need to do this? And he was like, what do you mean? This is the right thing, that’s what we just do.

When I joined here, we started capturing, “What are our values?” And we’d found so much value over the years in sharing things publicly. People always said, “Fog Creek has private offices for every coder, and they wrote about it, so now we can do this.”

Stack Overflow, our sibling company with the same founders that I’m on the board of, they built a salary calculator. You put in all the factors, your years of experience and what role you’re in. And it was informed by data from the whole industry. And people loved it. So we spent a year working on salary transparency. We realised, we have other H.R. policies and company policies that are valuable.

And everybody found it empowering to have them documented. In my background as an activist in communities I’ve been a part of, [I’ve learned] if it ain’t in writing, it’s not real.

When I introduced the climate leave policy, we saw the power of that. Half our team is remote. We had people displaced simultaneously by wildfires and hurricanes.

We simply said if you’re displaced by weather, climate, which is increasingly common, you’ll still have your job and we’ll take care of you. Same as sick leave. We track it the same way.

And that blew up. Obviously activists want to be able to point at it. Here’s a thing we need to be mindful of. But it was also just pragmatism. It was not contrived.

I think people are like, this is this social justice warrior CEO, performing. I was like, we have multiple people on our staff currently displaced! So we have to write a policy! And since we are going to write a policy, let me write about the policy and put it out there, because it’ll help other people do the right thing.

Maciej Ceglowski who does Pinboard [was] very cynical. He didn’t say it this way but he came just shy of saying this is virtue signalling. Like “You’re just saying these platitudes, of course you’re not going to fire these people.”

And I said to him, your view of who’s gonna get fired in tech is shaped by your experience as a white guy. There are a ton of underrepresented people on our team who are like, I don’t care what [the company] says, if I don’t show up for work, I’m not trusting that I’m gonna still have my job. You might have that confidence, but I don’t.

Obviously some of this is us showing off our products. The handbook is a Glitch app and we want people to use it. We’re proud of that. You can remix it and change it yourself. Another company said “We remixed the Glitch handbook, used it as a starting point for our own, changed it to what we’re able to do.” Not every company can afford to do the things we do.

I was on a high all day, because our team had made working conditions better for another team, and in this loosely connected way. We didn’t have to inflate the rat in front of their office.

In posting the handbook, you made a point that a good benefits package means sometimes salary is not going to be as high as a comparable job somewhere else. Can you talk about the effect of that?

I mean, we pay well. Nobody’s starving. We look at industry data—and this is slight shorthand—but there’s this 20 per cent of very high paying companies in tech. And generally they are paying for you to hold your nose as you walk in the door. You look at Uber and—well, being down here near Wall Street, a lot of the banks—and you can make a boatload of money. If you want to work on the self-driving car team at Uber, you’ll make a shitload of money. And we all know why.

We’re not asking anybody to take a pay cut, take a vow of poverty, but we’re not gonna be able to pay like Google. But we peg it to, I think, the 75th percentile on core compensation, on cash.

This doesn’t come up in Harvard Business Review or business management books, but it should: Our managers are freer to manage their teams because they’re not in a constant, passive aggressive battle about pay with their employees. Every other company including when I was a coder, I was constantly like, I can’t really tell my boss why I think about this thing, because I’m trying to get a raise next year and I don’t want to be a troublemaker. All that stuff goes away.

Benefits, we over-index on. We do 16 weeks parental leave whether it’s adoption or birth, whatever your gender is. And people take it. We encourage taking it. Same thing with vacation. We really encourage taking it and we’re good about it. But there’s also functional things. Our workday ends at the end of the workday. People go home at 5:30, 6 o’clock. Including me, I go pick up my kid from school.

There’s even expectations of boundaries around alerts on Slack. It doesn’t send off-hours notifications. We don’t do death marches. We’re building private offices for all of our coders. So there’s a lot of creature comforts, certainly no hardship.

And there’s a real alignment to what we’re doing. Everybody understands where we’re going and what we’re trying to do. We want everybody on the web to be able to build the web. Very straightforward.

So we have people from Google, from Snapchat, from the New York Times, from you name it. And many people who were independent entrepreneurs. They know the tradeoff. I tell people, I’m very appreciative, you could be anywhere.

We were a very homogenous team five years ago, when I was just an outside fan of the company. Now a majority of the team are women and non-binary folks, including in management. I think 30 per cent of our team identifies as LGBTQ. We have pretty much proportional representation on race across the team, including black and Latinx team members, which is in tech not very common. I think I’m the only South Asian man in the company. Pretty unusual at a software company.

I think there’s a long way to go. We still work in tech, we still have to work with companies. Everybody’s talking about right now within the tech industry, like, are you going to work with Palantir? Fortunately, they’re not using our product so I don’t need to worry about that right now.

But I can’t yet articulate why that’s different than working with Amazon or Google or Microsoft. And we’re going to work with those companies. And also they do amazing things! I look at the Chrome team, I’m like this shit is amazing.

So there’s going to be compromise. But can we form a rubric and rationale for how we make our choices and have everybody feel good about it, or at least be clear on it and not feel like it’s this creeping, incremental worsening? I think that comes from that level of trust.

We were very intentional, the investors do not have a board seat. And that’s fairly unprecedented. Especially at the scale of what we’re doing. They’re very supportive, we have a tremendously positive relationship with them. And where you get the money matters.

But I would like to diversify the board and have a community representative, somebody who’s a creator, who’s outside our usual circles and comes with a different perspective.

If Glitch fails—and most startups do—people are not going to say, “Their strategy around orchestrating containers in the cloud was wrong.” They’re going to say “This social justice warrior tried to prove a point and therefore couldn’t build a business.” I feel that pressure every day.

I’m really really driven to succeed at it. We’ve been able to do it because of the values we have, because of the team we have, because we are doing things the right way, because the community is not toxic and bad for the internet. So we have to win. Because we have to prove it’s true.

It’s not that others haven’t—like I said, Slack won without being bad for the world. But the narrative isn’t that. They’re “that weird messaging app” and they’re not a consumer brand.

I was losing my lid when Janelle Monae tweeted about Slack. “That should be us.” But, you know, give us a year or two.

This interview has been edited for clarity. 

Comments

Leave a Reply