Interview With Lea Verou of the W3C
There's so much goodness happening on the web and as it continues to evolve, it's important that talented individuals step up into leadership roles to help shape the future of web development. And doing this isn't an easy task. Not only do you need to have the technical chops to help define new techniques and paradigms or create the next great technology, it's equally important to be able to effectively convey your message in a passionate and credible fashion so that your peers respect your direction.
Lea Verou is one of these new breed of leaders who is helping to push the web forward through her technical savvy and profound love for web standards. She's developed quite a following and her live coding sessions at major conferences are a thing of legend.
We had an opportunity to find out more about her in this Q&A.
Q Let's start with the usual. Could you give us a quick intro about yourself?
Q You've risen quickly to be one of the most recognized and respected web developers around. Has that changed the way that you view yourself within the community and the responsibilities you may (or may not) have in promoting best practices and specific technologies?
Not really, to be honest. I still do my thing, make stuff and put it out there in the hopes they will be useful for someone. I still speak my mind about the technologies and best practices I like and those that I don't. Whoever wants to listen to me, it's their call. I'm not going to censor myself because of the number of people who are following me. That would be counter-intuitive, since being myself made these people follow me in the first place.
Q You've been very vocal about the problem with vendor prefixes. Do you think that's been solved?
I think both browser makers and the WG (working group) have realized that vendor prefixes, although good in theory, do not work in practice. So, the way to go right now seems to be browsers implementing experimental features under a setting instead of behind a prefix. That way, developers will not start using it in production, forcing the WG to get stuck in early iterations, as was the case with vendor prefixes.
Q Along those same lines, how much responsibility did the W3C, the CSS WG and WebKit teams have in perpetuating what became an incredible hindrance to cross-browser development (especially mobile)?
There's no single cause, but I believe a big part of the blame lies with developers. Although we've endured the pains of a browser mono-culture before, we did not learn much. IE6 used to be really cool stuff 12 years ago you know, just like WebKit is today. I can see the CSS WG being at fault too, for not realizing the issues with vendor prefixes early on, which turned web development into a popularity contest. Last but not least, the WebKit team shares some part of the blame, as they shouldn't have implemented non-standard CSS features to get ahead in the browser game.
Q Developers want more modern features and they want them now. Is the pace of the standards bodies keeping up with the needs and wants of the developer community? If not, what needs to happen to change that?
I'm sure you are aware of the old project management triangle paradigm: Something cannot be fast, high quality and cheap, you need to pick two. I believe this applies to designing APIs as well. Budget is limited, as there are very few people paid to work on standards. So, basically, designing new features can either be fast or high quality, but not both. We can see the former when browsers decide to innovate: Usually, even when the original ideas are good, they are poorly thought out, since they were designed in the vacuum of a single company (examples: Drag and Drop API, -webkit-gradient()).
When the standardization process is followed through, features can be very high quality in the end, at the cost of taking a long time to be finished. Several parties with different interests need to reach consensus, a full test-suite needs to be written, it needs experimental implementations and several iterations based on implementor feedback. All of this takes time, but keep in mind that once a feature enters the open web platform, we're stuck with it for years, if not decades. Therefore, it pays off to invest that kind of resources in to it, and to be patient. Short-term pain for long-term gain ;)
Q You recently announced your departure from the W3C. How will that affect your involvement in the standards process?
I will still be involved in the CSS WG as an Invited Expert. The WG voted on it in a recent telcon and I was happy to see several people in support and nobody against it. :) In fact, I will be able to devote more time to it now, since I expect to have more free time in general, and having worked at W3C gives me a unique perspective and insight into the standards process.
Q You're renowned for your live coding demos, flooring conference attendees during your presentations. Aren't you concerned about messing up and affecting your flow? How do you even prep for something like that?
I have several safeguards preventing me from messing up. I keep my code examples concise, showing only what's needed. This also helps the audience understand, as the code is small enough for them to process. I believe that as the number of lines in a code sample grows linearly, understanding decreases exponentially. Most importantly, I practice a lot. I might not practice the delivery of the talk, but I always practice the live coding several times, even when I've given the talk before. Also, even if I mess up, which has happened a couple times, the audience is so glad to see the result gradually build up in front of them instead of being presented with a screenshot, that they can be incredibly forgiving of missteps. If something does not work, I will spend a few seconds trying to fix it and if I can't, I will explain what was supposed to happen and move on.
I often see people ruining live coding presentations by showing too much code, with too many distractions (e.g. a full IDE around it and having to switch windows to see the result) and long delays trying to debug their code when something goes awry. All three are very effective in getting the audience to lose focus. However, done right, live coding can be a great teaching aid and make a talk more engaging and fun.
Q A lot of devs long to be as multi-faceted as you. Most seem to be good in either JS or CSS but usually not both. What are the techniques or resources that have allowed you to become as proficient in both to where you can do live coding demos near flawlessly?
My two biggest interests since my preteens were design and programming. So, when I started making websites, I was studying the languages involved, as much as studying graphic design principles. I fell in love with CSS because it felt like a bridge between the two, which is why I specialized in it.
Regarding resources, I was always the type of person that learned through reading and practicing (in that order). I would read an entire book and then build something that helps me put what I learned into practice to create something that I wasn't able to before. I don't learn easily from lectures, and I even glided through university (Computer Science) studying on my own, rarely attending any lectures. However, this greatly depends on the person, I know some amazingly skilled folks who absolutely hate studying on their own without anyone teaching them. I think that's why I tried to take a different approach in my own talks, because I find conventional lectures so damn boring and hard to follow, except for the rare case where the speaker is as funny as a stand-up comedian (and while I like to think I have a good sense of humor, I'm by no means on that level).
More About Lea
Thank you Lea for taking the time to chat with us. To get to know more about Lea and her work on standards and web development, be sure to visit her website and follow her on Twitter. Also, if you have an opportunity to see her speak in person, definitely jump on it. Her list of past and future events can be found on Lanyrd which also link to videos of her previous presentations.