I Had 10 Dollars

This interview is a bit different for I Had Ten Dollars. For one thing, it’s the first interview I’ve done over chat, so it’s the first interview I’ll actually be editing in any significant way. That’s boring to everyone besides Andre, who has to worry about me butchering his responses though. What really makes this interview different is that Andre didn’t really want to talk about his project, MLKSHK, so I’m going to talk about it a little instead.

MLKSHK is a photo sharing community, which Rusty mentioned alongside MetaFilter in today’s Tabs as being, “the product of another pioneer of an earlier web,” and Matthew Panzarino accurately described as providing an “honest transaction when you visit. You go there to look at an image and the image loads first without a bunch of crap being shoved in your face,” in his TechCrunch post.

While MLKSHK opened to the public around the end of 2010 and may not be early-internet old, I see it as a sort of bridge to an earlier web. The site was bootstrapped from the beginning, and has remained solely under the control of Andre and his partner Amber Costley for its entire existence. MLKSHK functions as a beacon of tasteful restraint while still managing to maintain an exuberant style that can make me feel like I’m having fun as soon as I land on the page.

Collaborative collections like “computers”, “you had one job”, and “glorious hair”, allow for themed image sharing trends that are probably most easily described as being similar to submission-based themed Tumblrs.

Okay, I’ll let Andre talk now.

How did you come to tech?

My first real tech job was in the QA department of a startup in Los Angeles which I won’t name because it was started by Scientologists and it was all very creepy. We had creepy clothes we wore to conferences and they creepily tried to get us to go to a session at the Scientology Celebrity Centre in Los Angeles for an evaluation.

It was a startup though, so I had a chance to learn a ton of things and I took advantage of that. I would hang out with the developers and ask them questions and I learned how to make gold masters and a little bit of C++. I taught myself how to make stuff with MFC and wrote some pretty awful automation tools for testing. I wasn’t good enough to join their engineering team but I knew I could fake it someplace until I got better.

But then one Thursday they told us to take the following day off and I was at home enjoying my day off when a co-worker called and reminded me that it was payday and no checks showed up in our bank accounts.

I quit shortly after that and got my first job as a programmer.

I was kind of hooked on that lifestyle though. I wanted to work at places where you could move fast and try ideas out. I’ve worked at startups ever since—except for a few years I spent at Activision.

You mentioned getting hooked on the lifestyle and I’ve seen you mention being addicted to the feeling of passing your tests and shipping. What do you think makes that such a compelling feeling in a way that simply completing tasks isn’t?

I can answer this by describing how I work. I am not a zealot about any software development methodology. Like any productivity systems people come up with I think it’s important to take what is good about them and modify it so that it works for you. I am not a strict TDDer, but tests go hand-in-hand with writing code for me. It’s a cycle, and I feel like I can hop on at any point of it and get going.

Some days I write a lot of code and then tests, other times I spec out a bunch of tests and just leave them there to prime me to start working on code. But this is just a long way to say that I have a compulsion to be confident in my code because I have clients I am handing code off to and the last thing I ever want is them doing is coming back frustrated and unhappy with what we wrote.

I used to worry about code all the time. I’d worry about whether something I shipped worked. I’d worry if code my engineers wrote was buggy. I worried so much and now I worry very little. It’s no way to live.

There is no industry I can think of that can so very closely replicate an actual production environment for so cheap. We are so lucky we can spin up a few hundred clients and test out something we are about offer to our users in a matter of milliseconds. How cool is that? Who else can do that? Nobody. Write tests. Ship code. Worry about other stuff.

What are your favorite places online right now? Is there anything new that’s caught your attention as particularly fun?

I sit in three or four chat rooms every day. Each has a different group of people, and we usually commiserate on stuff we don’t want to tweet about or don’t feel comfortable sharing anyplace else. One is work based and the others are (for some weird reason) in different chat apps. [HipChat, Groupme, Telegram, Slack]

People look for safe places to hang out and be themselves and they will always do that. We did it on weblogs until everyone discovered our weblogs then we did it on Twitter and now we do it in chat rooms.

The second thing I want to say on this subject is that nobody does online chat right. I have been using chat apps since back when IRC was still king and nothing out there right now does it the way I want. I don’t include Slack in this because its focus is on businesses. It’s pretty damn close though.

[Editor’s Note: This kind of echoes something Choire Sicha said recently about all the most exciting writing happening in different chat apps. Everybody knows that all the best writing is actually happening in Venmo though.]

You’ve spent a lot of time working on advertising; or, more generally, bringing in revenue to back content. You even helped found a company around it in 2005. What attracted you to that initially and what keeps bringing you back to it?

I don’t know what it means that I don’t have a good answer for either of those questions right now. I think I am struggling with both those questions at this moment and I don’t have good answers yet. I find building things for people to do their jobs highly satisfying. And once you have built a platform, being able to build new and better tools on top of it using the data is a lot of fun. I guess advertising provides a lot of opportunities for that because there is a lot of data and a nice feedback loop with sales, publishers, and advertisers to get them what they need.

Plus, it seems like you focus on smaller content providers. Federated Media initially worked with smaller voices (some of whom have become giants, if I’m not mistaken) and Tugboat seemed to have the same focus on creators with established, but still niche audiences.

Yeah. I care deeply about the independent web and that was definitely our original focus at FM to help independent web publishers. And Tugboat has similar goals.

“But more to the point, programmer-blogger Torrez is a classic idea seeder; a person who introduces concepts to the people who take them big.”

Okay, so Mat Honan wrote this about you in 2011. Does this jive with how you see yourself? You share ideas all the time and seem happy just to talk about them; was there anything you were like, “Crap, I really should have pursued that”?

Huh, I think I am suddenly getting very self-conscious about this interview. The truth is I don’t think about this stuff. I like talking about ideas and sharing ideas with people, and I don’t really think about the future that much. I built Dropcash and Dropload because I needed them and other people needed them and I didn’t think about pursuing them further because they did what I needed them to do and people seemed to think so too.

I guess what I am saying is I like to see projects as something to complete. Like making a crescent wrench. At some point it’s a crescent wrench and any new features and doohickeys are just going to make it a more expensive crescent wrench but not necessarily a better crescent wrench.

I brush my teeth with the most boring looking toothbrush in the world. It’s made by GUM and has a flat handle and some soft bristles at one end. That’s what I want to build: boring-ass toothbrushes that help people do what they need or want to do.

I also don’t apply time limits to myself about the stuff I make. I’ve been working on an idea for about 14 years off and on. I wrote it once in PHP, then again as OS X application, then again as a Python application. It’s on hold again and we’ll see if I ever actually finish it. [Update: I started working on it again after I wrote this response!]

I feel like you are constantly sharing ideas and tinkering on stuff like Muni-Bar or Tea Timer. How do you decide what’s going to be an idea you share just to talk about and the ideas you want to try working on yourself? The bar utilities were probably pretty quick, but you also made your own CMS a little while ago. Like, you took a look at Jekyll and thought, “I’ll just make my own static site generator”?

I think my brain just naturally gravitates to what is interesting to me or what I need and sometimes after building something it is no longer interesting or I no longer need it. Muni Bar was a thing I did after being frustrated all the time about missing my train. My office window overlooked the train stop and very often I would be getting up to leave work when a train would show up and I’d be stuck waiting 15 minutes for the next one. So I thought about it and built (in an hour?) a little app that displayed how many minutes the next train was from my stop.

Everything was hardcoded and I shared the code on Github in case anyone wanted to take it further, but I was done with it. And now I don’t work in that office or work that close to the train so I just leave work without thinking. I have no use for that code.

Maybe it is this: I usually write code for things I want to exist. I don’t write code for things that are going to become PRODUCTS. Though I am infected with that particular strain of disease as much as anyone else in this city who works in tech, I think, so sometimes I do just that and then feel bad about it.

One thing I’ve been thinking about a lot is flexibility in online communities. You’ve created a couple now that have created their own conventions and have very strong voices. How do you find a balance between guiding people to successfully complete tasks without boxing them in?

I do not know! I don’t know. I once sat down to write a blog post on communities but the truth is I have no clue. If anything, I do believe there are two rules that should be enforced in every community, one for the members and one for the person in charge of running it. And both rules are exactly the same: don’t be an asshole.

Sometimes I am an asshole! Sometimes the users are. People have bad days all the time and we shouldn’t hold that against them. It’s the habitual assholes that need to be escorted out of the way.

The other thing about communities is you have to show up every day and read as much as you can. You have to get to know every active member and store a little bit of info about them in your head. If that knowledge doesn’t fit in your one brain then get another brain to help. Keep getting brains until you have enough.

How do you typically structure your time for working on side projects. An hour a day after work? Weekend sprints?

Before we had kids I would spend my weekends hacking on ideas or staying up late after work. Now that we have kids I just think a lot about stuff. Sometimes I can bang something out in an hour or so in the 9-10pm hour before I have to go to sleep. I wish I could be more like Darius Kazemi. Here is how I wish I worked

You also write music news for 7x7 and SFBG. How did you get into that?

It’s actually a funny story: that isn’t me.

There’s some other guy named Andre Torrez who lives in SF and writes for the SFBG and CBS. I emailed him once to get a beer and he didn’t reply so I figure it’s probably too much to deal with, or he’s busy, or he is constantly fending off emails from people named Andre Torrez trying to get a beer with him.

[Editor’s Note: I feel really silly about this because I have taken recommendations from Not Andre a couple of times.]