This is a guest post by Frank Degenaar (@ProMashUp), author of the book, “Do Way, Way More in WorkFlowy”.
This is the first of a 2-part interview with Jesse Patel and Mike Turitzin, WorkFlowy’s co-creators. Today’s post is a throwback to the early days of WorkFlowy’s ideation and inception… while the next post will take a look at some tougher questions about WorkFlowy’s vision and behind-the-scenes development.
FRANK: I went fishing for WorkFlowy’s genesis and unearthed the following from a 2012 blog post elsewhere on the ‘net:
The idea for [WorkFlowy] grew out of Jesse Patel’s work at a nonprofit, “a job that was really overwhelming, where I had to manage a bunch of moving parts for 30 different projects.” While at that job, Patel tried many different programs to help him get organized. “The biggest problem with all of them is that they don’t support flexible data structures—they don’t let you define things how you want,” he says. “Instead they make you work in a specific way. Everything was super-janky and hard to use. So I was like, I’m just going to start creating a hierarchical interface for myself to manage this stuff.”
I’ve got a bunch of questions from that alone. For instance, how many man hours/ days, roughly, went into getting your first prototype up and running? And could you give us something… anything… more about those early days?
JESSE: I remember where I was when I decided to make WorkFlowy. It was in 2008 at some point. I was sitting at a red desk in the attic of a beautiful apartment, and I’d been trying and searching for a hierarchical, zoomable solution to my project management needs.I was teaching myself to program at the time, and tinkering with a lot of little thing, and decided to just try to make a hierarchical, zoomable interface to manage all my info and projects.
Given the fact that I didn’t know how to program very well, it took me a long time. It was in June, when I was living in Berlin for a month, working out of the Soundcloud offices (they had maybe 13 people at the time? They’d just raised $1M) that I really started working on WorkFlowy a lot more.
I think I got to something super basic I could use pretty quickly, but it literally stored the html in a big blob in the database and rewrote it every time I saved an edit. For those who know web development, you’ll understand how insane this is. It was also a pretty clever hack that let me, as a really inexperienced programmer, make something usable for myself pretty quickly.
I thought of this as my learning-to-code project because while I desperately wanted something like it, I thought it was a stupid idea that was bound to fail. After all, every programming tutorial starts with a todo list or note taking app, and everyone seems to think they can make a better one. Therefore, I viewed it as extremely unlikely that the idea I had for WorkFlowy was really novel or interesting, and that I was deluded in the same way everyone who made a notes/todo app was. So I thought, “This is a good starter project, because I am terrible at programming and I won’t be ruining a good idea with bad code.”
By around September or October of 2009 I had quit my job and was working on WorkFlowy full time, and had made something other people could use. So I guess it was about 6 months until I had made a real thing, but it would have been a lot faster if I actually knew how to program at the time. I had been teaching myself on the side for years, but still hadn’t put a ton of hours into it.
I named WorkFlowy in about 10 minutes, because I didn’t think it was important and didn’t think the product was going to be a real thing. If I’d known I’d spend this long working on it, and this many people would use it, I would have thought a lot harder about the name. I literally wrote down five names, asked my girlfriend (now wife) which she liked best, and registered the domain.
I was living in Geneva at the time. My first user was a friend name Shafqat Islam, who was working on his own startup, Newscred, which is now a pretty big success, hundreds of employees, offices in NYC, etc.. At the time, it was him and a friend working out of his apartment or mine, I ran a little co-working space for a bit. He used the first version of WorkFlowy religiously for a few months, then his usage petered out. His enjoyment of it was the first real sign that the zoomable hierarchy was interesting.
At the time, WorkFlowy looked dramatically different than it does now. I hadn’t yet arrived at the single piece of paper metaphor, and was instead using a column oriented approach similar to what you’ll find when exploring folders in a Mac finder. Here’s a screenshot of my actual account in an early version:
As you can see, even from this early stage, the development of WorkFlowy, which from this screenshot I guess I was calling “WorkFlow” at the time, was managed within itself. So, it really has been since the very beginning that WorkFlowy has been built in WorkFlowy.
Before I moved back to the US from Geneva, I rewrote WorkFlowy in about a week, using the single page interface that you see it in now. It had gotten kinda complicated and I wanted to make it simpler, get rid of my unbelievably crappy code and replace it with something better based on my learnings, which was still terrible code. You couldn’t click to edit, so it didn’t feel like a word processor, but that’s when it made the basic shift to looking like a text document, and that metaphor has driven much of the product direction in the years since.
At the end of 2009/Beginning of 2010, Mike and I started talking about working together on something. I had started to go slightly insane working by myself and was reading all sorts of stuff on Hacker News about how solo founders didn’t do as well. I also realized I wasn’t a good enough programmer to do something real alone.
Anyway, we started working together, but not on WorkFlowy. Got into Y Combinator, and ended up flailing around a bit as various projects didn’t seem to have legs. A few people in YC started using WorkFlowy at the time, and around half way through, we decided to make WorkFlowy our main project, so at least we would have something to show on demo day.
Then we finally launched it in November of 2010. Because we were part of Y Combinator, we got an article in Techcrunch automatically. Then Lifehacker wrote about us the next day, and after a day we had over 10K signups, and people seemed to really love the product. They stuck around and kept using it over time, told other people about it, and filled our email inboxes with nice emails saying how much they liked what we’d done. Although we had just launched it at that point, this was really the end of the beginning, because we now had a real product out in the world.
MIKE: I had worked as a programmer on the search engine at Google for a few years after college. I left that job wanting to strike out on an adventure, and I spent a year doing all sorts of interesting non-technical things.
I got back into the tech world in early 2010 when I worked with my brother Chris on some entertaining viral Facebook apps that reached over 10 million people in a matter of days (those were the days!)
Jesse and I had lived in the same residence in college. We started working together in 2010 because we had similar interests and both wanted a collaborator. We worked for several months on stuff that was completely and totally unrelated to WF.
As Jesse said, WF began as his “teach myself to program” project, which is obviously funny in retrospect.
As he said, midway through Y Combinator, we didn’t know what we were going to do next, and we needed something for demo day. The deadline loomed!
I had tried Jesse’s WF demo a bit, and in particular I thought the zoom feature was cool. I was already making multi-level bulleted lists in Google Docs all the time (and we did a lot of this when planning out earlier projects).
From my perspective, WF could completely take over my usage of Google Docs for project management and note-taking if we adopted an interface that was just as easy to use for adding and editing bullets as Google Docs (or any word processor) was. That became our first project when we started on WorkFlowy – to make editing as seamless as possible.
We discussed directions for WF, and we had similar visions of where it could go, so we decided to make it a collaboration. It felt like a fairly insignificant decision at the time, which is funny in retrospect.
FRANK: When you were conceptualizing everything and hammering out the WorkFlowy dynamic, were you really *confident* or really *hopeful* that the whole concept of WorkFlowy would catch on? Did you know right from the inception that along the way you’d pick up a crowd of crazed fans whose workflows depended on WorkFlowy?
JESSE: I am a person who dreams really big, about everything, to a ridiculous extent. Even if I’m making a peanut butter and jelly sandwich, I’m thinking, “This new peanut butter technique I just invented is going to revolutionize PB&J”. So basically, I always think everything I’m doing will be revolutionary and important, which is absurd and silly. I’ve been like this since I was a kid.
However, I’m also aware of this tendency to fantasize about the huge potential of things, so I have a big part of my brain that tries to be a realist. There’s kind of a balance in my mind between the realist keeping things in perspective and the fantasizer saying, “This is going to revolutionize everything.”
With regard to WorkFlowy, I simultaneously felt two contradictory ways about the project. First an enthusiasm around, “This is going to be huge, it is going to become the universal way people think things through, organize their work and communicate complicated ideas.” Second, a more realistic, “This is a throwaway project that will never see the light of day. I’m just learning to code here, and everyone makes a to-do app as their learning project.”
When we launched WorkFlowy, we were definitely surprised by the enthusiasm and usage that it got, and I still am. At the same time, I am honestly still disappointed that we haven’t succeeded in reaching a larger audience, because I think we really have made a better way to organize one’s thinking and one’s life.
MIKE: I was personally confident that our vision of where we wanted WF to go would produce an awesome product that I would love to use myself. So I was fully behind the idea, and a lot of my motivation has come from a personal desire to use the product.
Like Jesse, I did have some skepticism that the project could become a big thing. A lot of this has to do with the general crowdedness of the “space” that we are part of – there are so many note-taking, task management, and collaboration tools available that it’s easy to get lost in the noise.
We were both very happy to see all the positive feedback when we released the first version of WF to the world in November 2010. It was great to hear so much good stuff about something that both of us were aware was still very crude.
FRANK: What does your 2-man team look like, practically? Do you guys work in the same space in a 2-desk office or do you collaborate remotely? What do the 2 of you each focus on? How do you divide all that is the behind-the-scenes of WorkFlowy? Do you just slog away at a bug or a feature with dogged determination until it’s done?
JESSE: Mike lives in NYC and I live in San Francisco. We used to both live in San Francisco, until 2014 when he moved. We collaborate remotely via email, Slack and WorkFlowy.
Mike has traditionally done what I’d refer to as the parts of the product that require more hardcore coding. I work on the product a lot as well, but also on a lot of things that aren’t product related: payroll, taxes, forms, finding contractors, etc.. I have also done all the stuff that requires use of Photoshop and the like.
We both do customer support, with Sasha taking care of all the stuff that doesn’t require one of us to do something on the backend.
MIKE: For the first few years of the company, we both lived in SF and worked most days in the same room together. We are now remote, as I am in NYC. We both are very involved in designing new features and UI concerns.
I have generally been the one who wakes up in the middle of the night to respond to server outages and fight fires. Things have been pretty stable recently, but there have been periods where we’ve faced a lot of difficulties, many of which were growth-related or due to people using the product in ways we hadn’t anticipated.
FRANK: Are you sometimes surprised at what many WorkFlowy users squeeze out of your brainchild? I mean… do you see people using WorkFlowy in a way you never personally imagined or intended? I know this might sound a little over-the-top for an organizational tool… but have you ever felt like this creation of yours is bigger than you?
JESSE: Yeah, I’m really impressed and humbled with what people do with it. Especially the things people accomplish with it, that is what I find most exciting. And the fact that people take their valuable time to write scripts and styles and stuff to modify WorkFlowy is insane. And that you wrote a 250-page book about it, and that people actually wanted to read that book – that really blew my mind.
MIKE: I am blown away at all the things people use WF for. That’s one thing that I find very gratifying – we wanted to build a product that gets out of your way and lets you do whatever you want with it. So it’s neat to see people doing everything with it, which is exactly how it should be used :)
In Thursday’s post Mike and Jesse will be spilling the beans about the WorkFlowy they envision. Don’t forget to share some love below…