Werd I/O

It’s fun to think of the work I’ve done in terms of Taylor Swift style eras. Hey, I might not have the musical talent, good looks, or legions of fans, but the work I’ve done has required a series of overlapping re-inventions.

So, why not. In roughly reverse-chronological order, here are my Taylor Swift eras; what are yours?

Super-serious journalism supporter.

Distinctive look: open button-down shirt
Distinctive food: Austin-style breakfast taco
Distinctive activity: karaoke

I got into media through a lucky encounter with the founders of what became Latakoo, who attended a talk I gave about user-centered social network design at Harvard’s Kennedy School in 2009. We collectively designed Latakoo to be an easy way for broadcast journalists to get their footage back to their newsrooms using commodity internet connections, in the video format the newsroom needed. It’s the way organizations like NBC News send much of their recorded video today.

I was the first CTO at The 19th, a non-profit newsroom reporting on gender, politics, and policy, and was an active participant in its Senior Leadership Team across all areas of organizational strategy. I’ve also contracted with other non-profit newsrooms to provide tech leadership support.

At Matter, I invested in media startups — but the cool thing about Matter’s fund structure was that the LPs were all media organizations like PRX, KQED, the Knight Foundation, the New York Times, the Associated Press, McClatchy, Tamedia, CNHI, and, yes, tronc. I got to regularly meet with teams from those organizations and (as part of the Matter team) help them through innovation problems they were encountering using a design thinking led approach. I also got to participate in their own internal innovation processes, like giving feedback as part of the KQED Lab internal accelerator.

Startup bro.

Distinctive look: branded hoodie over a t-shirt that was also branded; socks were also often branded; third wave coffee mug also featured logo
Distinctive food: kombucha on tap and espresso using the imported Italian machine
Distinctive activity: offsites

I was the Head of Engineering at ForUsAll, which was my only foray into fintech. I was drawn to it because of its original mission to help increase access to retirement savings for more people. There was a lot of pressure to raise subsequent rounds of funding, and a major culture shift as the in-person company moved to a remote-first company during the pandemic. This allowed me to hire people I ordinarily never could have, in every US timezone.

I was also a Senior Engineer at Medium on its publications team. It was my first experience working at a company that had, frankly, so much money, sometimes alongside people I’d been following for years. I got to work alongside people who had previously built fundamental tools like Gmail as well as core pieces of web technology. The change in context meant I started off terrified: everyone was so completely on top of their respective games, and I had the biggest imposter syndrome of my life. It was also, for reasons I still don’t completely understand, the most fashionably stylish team I’d ever worked with.

Open source utopian.

Distinctive look: the jeans-tshirt-and-blazer look, because we were trying to look fancy and legitimate
Distinctive food: poké, for some reason
Distinctive activity: long, long walks, sometimes to save money on transit fares

I worked with Julien Genestoux on his Unlock Protocol: a way to help independent creators make money on their own terms without a middleman. Fully open source and decentralized, the protocol has taken advantage of various blockchains as they’ve become available, allowing the protocol to become as fast and cost effective as possible. Julien and I are both open-web-first evangelists, and this attitude shows through in the project.

With Erin Richey, I built Known: a kind of social news feed that you host yourself. Any number of people can publish to a Known feed (my site is a news feed of one, but some have had hundreds or thousands). We built an award-winning site with KQED and people around the world are still using it to power their websites. For a while, Known allowed you to directly syndicate your content to third-party websites, which saw us get coverage in Wired, among other places.

With Dave Tosh, I built Elgg: an open source social networking platform that was used by the Canadian national government, Fortune 500 companies, and organizations like Greenpeace and Oxfam. It was, in retrospect, one of the first private social networks and social intranets. We built the first social network ever run at a university, and I’m particularly proud of the social movements that used it. For example, the Spanish Movimiento 15-M anti-austerity movement used Elgg to organize. We also built the first open data definition for social networks, which helped inform the subsequent design of ActivityPub.

Institutional web developer.

Distinctive look: ironed shirt and trousers
Distinctive food: university canteen food (I was kicked out of the Edinburgh MALTS canteen after hacking the menu)
Distinctive activity: inventing acronyms for things

I ran the web properties at Oxford University’s Saïd Business School. The coolest thing about this job was getting to know the faculty and students; it wasn’t long before they realized that I knew a lot more about startups and web tech than a random guy in an IT department probably should. I ended up meeting visiting dignitaries and participating in MBA round-tables. They were very kind to me, and in turn, I believe I pushed the IT department forward in its relationship to the web.

And first, perhaps most improbably, I ran the web properties for what is now the St Leonard’s Land Pool at the University of Edinburgh: an Olympic-sized swimming pool set up with underwater cameras to analyze and improve the strokes and techniques of elite athletes. I started being loaned out to the Edinburgh University Media and Learning Technology Service, which is where I met Dave and started cooking up Elgg.

Proto-nerd.

Distinctive look: baggy sweatshirt, jeans, oversized glasses, leather jacket for some reason
Distinctive food: chips
Distinctive activity: putting 486 computers together

I helped build the first website for Daily Information, a local one-sheet newspaper for Oxford that included classified ads (it was possibly the first classified ad website in the world, pre-dating Craigslist) and reviews for local restaurants, movies, gigs, and theater. Before it became a website, I came on as its first BBS SysOp — my first ever job.

I ran a hypertext magazine called Spire, which I built in Windows Help Format because its capabilities at the time outstripped HTML. (We did move to the web later on.) I got to interview celebrities-to-me like Roger Ebert and Nicholas Negroponte. Distribution was via BBS initially, and then we started to be carried on the cover CDs of more professional print computer magazines (something I achieved by faxing them all in turn with a proposal, which blows my mind now). I was 15.

And I ran Rum and Monkey, a website that regularly got millions of pageviews a day and taught me all about social virality (this was 2002). I’ve written extensively about that over here.

I’ve used DeleteMe to remove my personal information from search engines and information hubs, but it hadn’t occurred to me until recently that I needed to also remove information about my home from listings sites. It turns out there are full photos, including video walkthroughs, just about everywhere. Particularly with a baby in the house, we felt uneasy about leaving these up.

USA Today has a quick guide to removing your home photos on the most popular sites, but it turns out there’s no public way to remove listing photos from MLS, the listings database that realtors use behind the scenes. You have to ask your agent nicely to do it on your behalf, which I didn’t know to do when I bought the house.

The selling agent also uploaded videos to YouTube — and there’s no defined process to remove those. I’ve had to send a nice email and hope that he has the time and inclination to remove. It would be nice if there was an automated way to remove my information there, too. (Updated to add: he very kindly removed it incredibly quickly.)

I post a lot, but keeping your personal data footprint on the internet clean is really difficult even if you don’t keep a blog or post to social media. Although other people shouldn’t post your personal information — it’s not legal, for a start — there’s no unified way to prevent them from doing so. There’s the threat of data leaks, of course, but there’s also the threat of intentional disclosure by someone who thinks what they’re doing is benign. In that way, it’s somewhere between an arms race and a losing battle: you can’t ever be sure that someone you’re dealing with in some capacity isn’t sharing more than they should about you on the web.

Searching for yourself and your other identifying information is a good way to figure out what’s out there, although the act of searching leaves its own insecure footprint. Zuckerberg was morally wrong when he said that the era of privacy is dead, but I wonder if he was, on a very practical level, correct.

I’m not a particularly vulnerable person. In contrast, for some people, these disclosures are life and death. Revealing an address or a home walkthrough has real implications for a journalist reporting on political corruption or someone fleeing their abusive partner.

We can build all the tools we want, but as I mentioned, it’s an arms race: there will always be more disclosures. Eventually this all comes down to establishing strong legal protections, and more importantly social norms, around privacy. The design of our internet tools and social networks, our standard patterns of use, and the way we think about organizing the data underlying the ways we search and share online are all organized around the principle of public-by-default. What if that all changed? How might it? And are the collateral losses — less sharing on the internet overall, fewer services around certain kinds of personal data — worth it?

One of the newsletters I subscribe to ran a sponsored post for Paintbrush, a firm that gives idea-stage founders a $50,000 loan to prove out their idea. The pitch on the front page is, “No rich aunt or uncle? No worries.”

My initial reaction was positive: I do think access to capital for founders from non-wealthy backgrounds is important. We’re missing out on so many important businesses by perpetuating an ecosystem that works best for people with deep pockets (who, in turn, tend to come from a narrow set of demographics). But the more I dug in, the more I think this is a bad deal, and I wanted to talk about why.

Based on their literature, Paintbrush provides a $50,000 loan with a very low-friction application and a fast decision. But the total repayment amount can be as much as $75,000, tied to a personal founder guarantee. That means that if your startup doesn’t work, you as a founder are required to pay that amount back at an amount pegged at 15% of your pre-tax income. For example, if your total income was $150,000, you would pay back $22,500 a year. That amounts to around 22% of what your post-tax takehome pay would be before payments like health insurance and rent.

Investor and founder Erik Severinghaus, in a piece entitled Never, Ever Personally Guarantee Your Startup:

Remember that 75 percent of even venture backed startups fail. Behind every one of those failures is a story of heartbroken entrepreneurs trying valiantly to extricate themselves from a challenging situation while retaining some modicum of dignity. Putting the money aside, that emotional hell is one that you don't want to live through, and it's exponentially worse if your creditors can come after your personal assets in addition to the corporate ones.

Not only that, but if you want to follow the VC path — or, for example, take part in an accelerator — you should know that investors take a close look at debt that you might have on the books. At an earlier stage startup, debt is a higher percentage of a startup’s total value, so early investors may take a particularly unkind view of it.

I expect that the founders of Paintbrush are trying to do the right thing. And in some cases, it may well still be a good solution! But I’d warn entrepreneurs to think about it very carefully before plunging in. Even if they provide a quick answer about your “funding”, you need to take your time and consider your options — and particularly the consequences if, like 90% of startups, yours fails. A fast process can lead to emotional decision-making where you’re all signed up before you consider the consequences. There may be better routes forward.

I’ve been thinking a lot about how to build high-performing teams: specifically, teams that build great products that I would also enjoy to be a part of. An incredibly productive team that also happens to be full of jerks is not something I’m particularly interested in replicating; I care about building meaningful things well in a resilient, nurturing environment. As well as being nicer places to work overall, these kinds of teams tend to have lower churn (people tend to stay for longer) and higher quality end products (the people who build things really care about what they’re building).

One of the most important things I learned working for Corey Ford at Matter Ventures was that a culture of open feedback is a core part of building a supportive culture. If people are to do their best work, they have to receive constructive feedback from their colleagues well; they also have to be able to give it openly. A team that’s stewing about friction they’re encountering without being able to talk about it in a way that might lead to resolution is one that’s highly likely to burn out.

One of the tools we used at Matter, which I believe was inspired by the famous Interpersonal Dynamics class at Stanford Business School, was a simple way to give and receive feedback on a regular cadence. I’ll describe the Matter version, which was face-to-face, and then discuss how I’ve adapted it for remote working.

By the way, Corey is an expert at this; he now runs Columbia University’s Sulzberger Executive Leadership Program for news executives, which is a giant opportunity if you’re in the industry. Regardless of the kind of organization you work in, you want him to help with your organizational culture.

In-person feedback.
Time to complete: 30 mins

Two people — Person A and Person B — sit opposite each other. Each has twelve square Post-It notes of a particular color; Person A might have twelve yellow Post-Its while Person B might have twelve blue Post-Its.

They set a timer and spend roughly fifteen minutes writing privately:

• Three Post-Its giving themselves positive feedback. What’s something that went well?
• Two Post-Its giving themselves deltas: what’s something they wish they could change?
• One Post-It describing how they’re feeling about their work overall.
• Another six giving the other person feedback in the same pattern: three positive, two deltas, and one that describes their overall feeling about their working relationship with that person.

Post-Its should always be written in a thick pen like a Sharpie, which forces brevity. Each one should be as simple as a headline, with the author’s name in the bottom corner.

Then the participants take turns to reveal their Post-Its.

• If Person A starts, they start with their feedback to themself first, revealing each Post-It one by one, and describing it a little bit more than is written in the headline.
• Then they continue onto their feedback for Person B, revealing and explaining each Post-It one at a time. Person B must remain silent except to ask clarifying questions.
• At the end of Person A’s Post-Its, Person B just says “thank you”. No rebuttals are allowed.
• Then you swap: Person B presents their Post-Its in the same way, and Person A says “thank you” at the end.
• Each person takes the feedback Post-Its that the other person has written for them.

There are a few obvious pitfalls, which should be explicitly called out at the beginning of explaining this kind of session for the first time:

• Don’t go “over the net”. This means don’t make assumptions about someone’s motivations or causation for a particular event. It’s totally fine to say, “when you did X it made me feel Y”; it’s not okay to say, for example, “you did X because you don’t care about Z”.
• Be aware of other common cognitive biases.
• Don’t interrupt the presenter.
• Nothing leaves the room. No feedback should be discussed with anyone else.

Most importantly, when someone is giving you feedback, they’re giving you the gift of their inner mind: they’re speaking what might otherwise be unsaid, so that you can become aware of other peoples’ reactions and learn from them. The process should be taken and received in the spirit of gift-giving.

Therefore, protecting a safe space is vital. Crucially, managers should be prepared to receive honest feedback as well as give it, in the same spirit of gift-giving. If there is ever any blowback from feedback from a manager, or an adverse reaction, the space is no longer safe and the feedback is not effective.

This also can’t be a one-off, because comfort with giving and receiving feedback builds over time. So it’s best if everyone has a one-on-one feedback session with all the people they directly work with at least every few weeks.

Remote feedback.

Obviously, there are no Post-Its directly in a Zoom call, and collaborative whiteboarding services tend not to have a function that allows you to write in private and then reveal your sticky notes one at a time. It’s also awkward as hell to write on a paper Post-It and hold it up to the camera as you speak.

I’ve experimented with a shared Google Doc or a whiteboard space, and I think the best version of this that I’ve come up with works as follows:

• Each person starts in their own document. I prefer sticky notes a whiteboard space, but a Google Doc works pretty much as well with a little set up. You’ll want to make sure that positive feedback, deltas, and the summary notes are each well marked, perhaps with a “+”, “Δ”, and line respectively.
• There is also a shared document that both people have open. Rather than screen sharing, each person is looking at this document during the sharing step.
• Each person copies and pastes a note into the shared document as they are describing it, one at a time.
• At the end, both people retain access to the document. Next time, a new document is started.

Otherwise, exactly the same rules apply.

This is just one tool. Obviously, establishing a participative, open, supportive culture requires a great many techniques, and is about an overarching mindset more than it is about any one type of meeting. But I’ve found this to be a very helpful part of my toolkit when I’m running teams. I hope you find it useful too.

Kate McKean describes how she’s writing her novel:

Right now, I am getting up early (6ish, not bonkers early) and leaving my house about 7am to go to a local coffee shop to write for an hour or two before my regular work day. I do this Tuesdays and Thursdays as much as possible.

It’s honestly not a bad plan. I’ve mostly been writing in the evenings once the baby goes to sleep, but not as consistently as I’d like: there are sets of days where I get barely any words down at all. But then again, there are other days when I write thousands, and because I’ve become used to my own ebbs and flows I try not to be too hard on myself.

If I’m writing during the daytime, green tea is my crutch. There’s something about just enough caffeine, without the cortisol boost that coffee gives you, that puts my head into the right spot. I used to depend on brain.fm to tune out distractions, but I’m lucky enough to have an office with a closing door. The sound of the wind outside — or more commonly lately, a raging thunderstorm — works just fine.

It’s taken a very long time to get this far, but at this pace I expect to have a full first draft ready by the end of September. Obviously, I’m full of self-doubt about being able to do anything with it once I hit that milestone, but getting there will be an achievement in itself.

And that’s all I really want to say about any of this, because talking about something you have written feels much more meaningful than talking about something you will.

Nonetheless: worth mentioning that I’m still at it.