Britt Selvitelle from Twitter talks about passion

Paul Boag

Britt Selvitte from Twitter talks about enthusiasm, passion and just getting your web application up and running.

Paul: So joining me is Britt Selvitelle from Twitter, good to have you on the show.

Britt: Thank you very much

Paul: I have just listened to your talk at the Future of Web Apps, which was excellent by the way, and just wanted to have a bit of chat with you about passion. Because that was the thing that really came across in your talk. You obviously started with this very melodramatic, on your knees “I love Software development” moment, which is always a good way to start when people have come in with hangovers.

Britt Selvitelle

Image Source

Marcus: I can honestly say I have never done that, gone down on one knee and said “I love software development”, ever.

Paul: Oh I thought you meant just generally speaking. I hope you did when you proposed.

Marcus: Yeah, maybe once or twice after a heavy session

Paul: “I love Alcohol”, “I love everything”

Britt: It is really is something I remember from years ago at school a professor of mine got down on his knees and spread his arms to the whole classroom of college students and said “I love Algorhythms!” and it really stuck in my mind.

Marcus: I wonder how many teachers are like that.

Britt: I know it was fantastic. He made such a great impression on me and I think that I would like more people to proclaim their excitement for creating these innovative bits of software.

Paul: Yeah and you gave us a quote as well which was really interesting, just tell us about that.

Britt: Yeah, Lars was one of the two authors behind Google Wave, which is of course getting huge amounts of press right now. And he said “I have been accused of being pathologically optimistic about it (google wave)” and that quote really struck home with me because, how funny or ridiculous it that being optimistic has actually become a derogatory term to people and something that people consider as being a jab.


Image Source

Paul: But that should be the ultimate compliment in a way.

Britt: of course.

Marcus: Sorry, if you are born as English person then that is just normal. We are all cynical and pessimistic people, although I am not personally…

Paul: But most English people are. You have this American dream thing going on.

Britt: Well y’know, I tell you in software development and in general there’s this fear of being overly optimistic because people have dealines and they don’t wanna promise too much because they will miss their deadlines but you have to have deadlines and you have to have structure. But the kind of energy about not being creative and excited about what you are doing is really stifling to people.

Paul: Yeah, that is what drives you and motivates you. If you don’t believe in what you are doing then it is going to be a job.

Britt: Exactly and I think a lot people end up not doing the thing, like they have an idea for the start up and they do one, and people accuse them of failing, then they do another and it doesn’t really work out, and they have this third idea, and maybe that is the idea that is really brilliant

Paul: Actually it was interesting in another interview where I gave this quote where Churchill talk about success is ‘going from failure to failure, with no loss of enthusiasm’

Britt: Yeah, absolutely. I couldn’t agree more.

Paul: And it doesn’t just apply to start-ups and things like that. A lot of the people that listen to this show are in-house developers in some corporate entity somewhere. And so often when we go in to companies and start working with them their in-house teams are ground down by politics and stuff like that and they need to re-kindle that enthusiasm.

Britt: Yeah, and I actually have a suggestion on that, something that works well for us is we currently work with a company called Pivotal Labs. And there are some contractors that are friends of ours who are really fantastic developers and of course twitters code base has been around since mid-2006. And so there is definitely some legacy, like when there is a new person they are pairing with they are always apologising, “I am sorry about this code, it is so terrible” and what’s really fantastic is when that new person (young blood energy) and say to the jaded person that has been looking at this code for three years and say “Chipper up little fella”, instead of grumbling about it, lets take thirty minutes or an hour out of our day, write some new tests and clean it up and actually touch on little bits of the code so that it improves overall and after three months, that guy will be bitter and jaded, so you need some new excitement. And often we find by moving people around, this isn’t something we have totally solidified yet because we are still working on a lot of our processes, but by kind of having some motion in the company and people looking at different things you often get some of that excitement.

Paul: Yeah, the other thing that really came across from your talk was, almost giving people permission, not that you are gonna have crap legacy code. It was this idea that it doesn’t all need to be perfect out of the gate, just get it out there, get going and lets streamline it later. That kinda thing.

Britt: Yeh, a lot of people, Front-end development specifically, at a certain point has to scale, but it doesn’t matter when you are by yourself. When you are doing a small project then the best thing to do is, like it’s the same in software, in general some people are for and against Google Optimisation, I am very much against it, I think you should just do something and rock it out and then worry about optimising it if it takes off. With scaling your front-end development team, that is not something you have to do until you have five or six people. The problem with Front-end development is that the language that we use is so expressive that they allow people to write, not necessarily good or bad code, but different code. Javascript allows you to write in hundreds of different styles, but they are all different. So it is not until now that we are all writing guidelines for how you should architect the look and feel of your JavaScript code within Twitter. But if you are just starting something, do not worry about that, get it out there for people to use.

Paul: Because you can get bogged down in that kind of stuff and you lose the momentum and the enthusiasm again. We are back to the same subject again. And talking about the whole thing of enthusiasm you struck me as someone who gets inspired by other stuff you are seeing online and you talked about different places and different things. What is it that really inspires you and what is some of the stuff out there.

Britt: I can give you one really geeky example, I don’t know what the limit is either 8 or 10 pages of apps on your iPhone. I have them all full, I have downloaded more apps than anybody I know. And those are the ones I have got on my phone. I have got several hundred sitting on the computer at home. I download these things because the concept of the iPhone and the way the user interface is constrained such that you have to innovate. I download them and I don’t really use them after that, I just want to see how they have got around these kind of constraints. Because constrained development is fantastic. That’s what I love about Twitter, because 140 characters makes you really choose your words. I sometimes spend 30 minutes authoring a 140 character tweet.

Paul: Okay, that’s obsessive!

Marcus: Basically, you hit enter and then you do the next one.

Britt: I start with a paragraph and then I slowly wittle down, until I have that core function that I want to get across.

Paul: But it is funny because so many people grumble and moan about the constraints on them. “Oh, the client wants this”, ” the client wants that”, “I can’t do what I wanna do” but, actually those constraints can be inspiring as well.

Britt: Absolutely, they are really inspired and I have done consulting and like we did a start-up in Kentucky where we did consulting for contract work and I think those constraints are necessary because the person that you are dealing with in those sort of situations just doesn’t know what you are working with so putting these constraints on them early on is often a really successful tactic.

Paul: Share the Tweetie 2 example, because I really loved that.

Britt: That’s just fantastic. Lauren is brilliant and his Tweet 2 is in Beta right now and I am helping him Beta test it. He has got this one feature where you are composing a reply to somebody, often the reply takes up the entire screen of the iPhone you lose the context of the message. So if you are trying to reply to a specific bit of the message it is only 140 characters but still you might want to check what you are replying on. And on Twitter that is fine because you can just look at the page. So he has this thing where you literally drag down the edit box, you can see a video of me doing this on the talk online, but if you drag down where you are typing it show in a de-emphasised form the tweet you are actually replying to. And then letting go it pops back up. It is something you really have to see to appreciate. It is an example of beautiful design and development that is what gets everybody at Twitter excited about what we do, creating these things.

Paul: You must find things people do with the Twitter API amazingly inspiring, because some of them are wild aren’t they.

Britt: Crazy, how inappropriate can I be?

Paul: You can be inappropriate, we can always remove it.

Britt: Yeah so, at SXSW in 2007 somebody had made a dildo-couch that vibrated whenever you got a tweet. So you put in your user name and sat down and whenever you got tweeted the dildo vibrated and I was like “Oh my God, what have we created” , it’s fantastic but good Lord.

Paul: I was in South-by and I don’t remember that

Britt: and something like that is really important to us, not only because it is inspirational and it feels good to have all these things that are changing the world. I had a guy came up to me when I was in a Rails conf in Germany in 2007, and he came up to me in a bar afterwards and said “I just wanted ask permission, because I don’t know if you allow this. I am using Twitter to send AIDS medication reminders to my patients in South Africa.

Paul: That’s incredible!

Britt: and I was like, “do I mind?, I couldn’t imagine a better way to use something”

Marcus: So it’s not just inane banter.

Paul: No, and it’s not just inane banter and dildo’s.

Paul: …and that’s amazing isn’t it that you start off with this really simple little idea. Because on the surface 140 character, because you put an API behind it’s being used in all these unexpected ways that you could never have planned.


Britt: and that’s a really good point because there are people now doing things that we could never have possible envisioned. I mean it would be ridiculously egotistical to think that we are going to come up with all the great ideas, or even a fraction. So, empowering other people and being able to feed off that and get back inspiration from those ideas!

Paul: On the subject of API’s on your talk you did talk about Alpha 1, the very first version you put out there, make sure it has an API. So, obviously you see it as a fundamental part of any decent web app.

Britt: I really do, and it something, that just came up last night when we were talking about API’s and I said “really it would be a shame to release an App, even the first release without an API” because, even if you release it and you don’t get adoption based on however you promoted it or something and then some other guy finds it and they’re like, this is what I have been waiting for my entire life. They can have access to that data and work with it and I really think that from day 1 it can define success or failure for that project. It is so easy with modern web-frameworks to toss in a basic API. It is worth the hour that you would delay the launch.

Paul: And that was the interesting thing about what you said, was that, y’know, it doesn’t need to be fancy full featured API, just do something basic, get that up and running,

Britt: It’s just amazing how people take twitter data. Even when we were small, everyday there would be an email sent out about some crazy thing somebody found online randomly. People do not give enough credit to how many people get excited about these things, they think ‘it’s just me, I’m weird’ and they do it and all of a sudden there is this big community being built around it.

Paul: and it’s quite interesting, obviously you are very much focussed on Web Apps and maybe a lot of people listening to this show aren’t building Web Apps and building more traditional web-sites, but I think even on a lot of these sites have got a lot of valuable data they could be opening up, for example we do a lot of work with Higher Education web-sites and they have lots of data in terms of courses that they run and stuff that’s going on, even that kind of stuff.

Britt: There is no doubt, I mean Twitter by no means now is a Web App, we are moving in that direction, but for the longest time we just had static pages with a bit of JavaScript spruced in. We are moving towards a sort of G-mail-esque sort of App. But, education, how many Colleges have really shitty web sites and where you have to dig through all the stuff, and how many students that are working on projects and could take this stuff and make that API and maybe create some brilliant way to, I mean, I remember being in College and trying to plan my classes and getting overlaps and classes and teachers that I didn’t know I was good or not, education is a perfect example.

Paul: This whole area of API’s, a lot of people go ‘oh API’s, that is for Silicon Valley start-ups’ y’know it is their kind of thing but actually I really don’t believe that I think it goes a lot wider than that. Y’know everybody should be looking at it.

Britt: If you have data you can make public, I think that is what a lot of people look at. This concept of an API, the idea of a public consumable API on the web is relatively new, I think a lot of people see the effort to build in, they don’t realise they have the data to put out there. They think, that’s really simple there is no reason for me to write and API. But even if you have just a little bit of data it just amazing what people will do with that. So if you are listening to this and working on something, even if you think just a little bit of the data is there, just do an API.

Paul: Of course the other thing is, that people are combining your bit of data with stuff from elsewhere and all these mash-ups and you might be contributing to something bigger.

Britt: And the other thing is that people think their data is not going to be useful to anyone else and that’s also a huge mistake and you cannot possibly imagine, like I said, it would be ridiculous to imagine that I could imagine even a tiny portion of any cases that you could use Twitter data for, so even if you think it’s useless, it’s an hour of work to throw in an API and do it.

Paul: Excellent, that’s superb and a really good message to end on, so thank you very much for coming on the show.

Britt: My pleasure, thank you so much for having me.

Thanks goes to Ben Hardcastle for transcribing this interview.