Hello Kevin. Your blog says "Hi, I'm Kevin. I build search engines for a really big company". Can you tell us more ? Not only what you do but who you are and where you live.
Well, that's not entirely accurate anymore (I've updated it because you pointed it out). I work for AOL as a Principal Software Engineer and spend my time working on whatever web problem seems to be the most important/problematic at that time. I just accepted a new position working in the AIM world, and expect to start spending more time rolling out new web products for AIM users. I spent the first six years as a web developer working on AOL Search, which I'll talk more about in the second and third questions. I've become something of a web standards evangelist in the last couple years, although it's only been a part of my job for the last year or so. It's a lot of fun, and means a lot of travel and training (you know, going to the W3C Tech Plenary, SxSW, etc).
How did you fall into the world of Web Standards ?
It's all AOL Search's fault. About four years ago, our product folks got this bright idea that a faster search results page would be a good thing. That meant that I needed to find ways to make the page smaller without removing any content. I went on a "quest" and for the next year, I spent all day working on reducing the amount of markup I used to produce the same result. This lead me to folks like Jeffrey Zeldman and web standards. The rest is history... I joined the CSS Working Group a couple years ago, although I haven't had as much time as I would like to spend on it, and started up a small grassroots standards advocacy group within AOL (about the same time) with Kimberly Blessing and Beth Epperson.
How can Web Standards help web-based applications in general and search engines in particular ?
There are dozens of ways in which web standards can help web applications. My favorites:
  • The possibility of speed: It's not a guarantee, but if you're going from complex table layout to semantic markup, you're going to see a major speed improvement because it should take much less markup to get the design result you're looking for.
  • Flexibility: There was a time when we redesigned AOL Search every couple weeks. During those designs, I rarely had to touch the markup because all the style was in CSS where it belongs. The ultimate goal of using standards, for me, is to keep things nicely separated. Markup should be about describing data. CSS is all about how that data appears. If you can keep those separate, your products will be faster and your users will be happier.
  • Conformity: I like being able to tell new developers that the documents they deliver have to validate as XHTML 1.0 Transitional. It requires them to do some quality checks before they deliver stuff, and I know that all tags will be lower case and all attributes quoted. We, of course, have other rules on top of that, but that's a great place to start from. The threat of error messages from the validator is a surprisingly strong motivator.
  • Findability: This is the search engine bit... As more spiders actually look at the markup in a document, semantic markup will become more and more important. Already, Google's indexer looks at markup, and I'm sure others will follow if they don't already. Having documents that are machine readable will soon be as important as having human readable ones.
How would you rate your influence inside your "really big company" as a standards evangelist ? Can you tell us what is your biggest problem trying to push standards ? Your largest success ?
How do I answer this without looking like a jackass? If I overstate my influence, I look like a pompous ass (I am). If I understate it, I look like a self-effacing goon (I'm that too). I think I've had an impact on the company's awareness of web standards. AOL has such an odd history with web standards. At Netscape, people were involved in shaping the standards people use every day: HTML, CSS, XML, SSL, RSS, etc. AOL forgot that for a while after the acquisition (don't cry, Daniel, I'm sorry... it wasn't my fault... really), and it's taken a bunch of people, not just me, to remind the company that standards are important and there are real tangible benefits from not only using them, but helping to create them as well. That communication has been the biggest challenge (which is what our panel at SxSW is all about) - how to craft the message around standards to get buy-in from management, design and any other decision makers. It's really hard for people who are passionate about things to take a step back and look at how to express that passion to people who aren't as enthralled with the idea as they are. My biggest success has been the management awareness of standards and the fact that I've been asked by senior vice presidents why I recommend XHTML 1.0 Transitional over 1.0 strict or 1.1 (that was a great day).
You seem to be participating more and more in Web conferences and standardization bodies. What are the must-see conferences these days and why are they important ? Can you tell us a bit about your activity as a W3C member ?
This is really my first year going to any conferences other than SxSW and the Technical Plenary. I still say that if you're passionate about the web, whether it's writing, designing, building or standardizing, that SxSW Interactive is the best place to go. It's folks from all of those disciplines in one place, ready and willing to share their expertise and wanting to learn what you know. It's great, and I can't wait for this year. I'm also looking forward to X-Tech and WWW, although I've never been to those conferences before.
Do Web Standards have limits ?
Unfortunately, they do. Standards are limited by their usefulness. For us front end folk, we're limited by browsers and their implementations of the standards. For example, I would love to be able to use CSS 2.1's display: table. It's awesome. I love it to death and have been playing with it for a couple years now, and would love to be able to use it in an actual shipping web product, but Microsoft Internet Explorer doesn't implement it and isn't going to in the near future. It's the same story with a lot of really cool CSS stuff, and others. I get really impatient with the speed of implementation, which is a personal flaw, but still... I wish things would move faster. Standards are only useful when they help you do something better than you did before. If the new standard isn't way better than the old one, and folks don't implemented, then not worth switching (see: XHTML 2.0).
Windows, Linux or Mac ?
All of the above? I love my Powerbook, but I have two linux boxes, and a Windows laptop. I much prefer OS X over Windows for multiple reasons that I won't get into here.
What are tomorrow's important standards ?
Mobile's going to be big, but it's going to take a long time for things to shake out. If you think getting standards implemented on desktop browser is slow... try baking the browser into a phone that might not ship for two years. It's a huge challenge, but one that's incredibly important. I think things like XBL and the new Rich Client groups are doing really interesting stuff that I unfortunately haven't had time to keep up with (hopefully next week I can sit in on some meetings).
Did blogging change the way you work ? Does it impact Web Standards at a conceptual level, and if yes how ?
It did. I played with all of the standards-based development stuff I eventually rolled into AOL Search on my blog first. My blog was XHTML compliant long before I got all the approvals and code through QA for AOL Search. Blogging also got me in touch, though their own blogs, with the folks who really pioneered standards adoption. Through their blogs, I learned more than I'll ever be able to repay them for. As for their impact on web standards, I think that blogging is where a lot of people come into contact with standards-based development for the first time. A lot of great developers and designers have gotten their start or redefined their careers through blogging, which I think is excellent.
Why do you have a blog ?
It started out as an experiment, but then I realized that it was a great way to keep my family updated on what was going on with me, express myself to the world, and eventually, a way to connect with a bunch of new friends. I don't think I would have found SxSW if I wasn't a blogger, and I wouldn't have been introduced to web standards when I was if I weren't blogging. I do it now out of habit, and because I still like doing it. I'm even considering starting a new blog just for longer essays about web development... when I have the time (so yeah, it'll never happen).
What your take on "web 2.0" ? On microformats ? On XHTML 2.0 ? On CSS 3 ?
That's a long list, you sure this is one interview?
  • Web 2.0: It depends. The hype is just that, but, there's real palpable excitement in the developer community that I haven't seen for several years, even at AOL. People are excited about working on the web. People are interested in standards and good design practices. If that's web 2.0, then I'm all for it. I still want display: table though.
  • Microformats: Love 'em so much I helped create one for a new AOL product (see http://iamalpha.com). I think they're a great way to fill in the gaps in existing standards, and the perfect application of what HTML should be. That we can come up with a way to agree on what markup should be used around types of content gets us closer to that separation I was talking about, and to a truly machine--readable web, without the need for clunky complex markup standards like RDF (not that it doesn't have its place too).
  • XHTML 2.0: I'll use it when I see that it provides real benefits over XHTML 1.0 Transitional. I haven't seen them yet, and I don't think any major browser has actually implemented it yet (might be wrong about that, but still... not useful to me yet).
  • CSS3: Can't wait. I can't wait for advanced layout to get implemented. I love playing with multiple backgrounds in Safari, opacity, etc. If the world can convince Microsoft to go implement some of the CSS3 stuff (oh, and the rest of 2.1) in Internet Explorer 7.x or 8, we'll see some truly amazing stuff come out of designers and developers. I feel so limited now, like I'm missing half the palette of tools I should have to work with in CSS. Need... advanced... layout... now!
Kevin, you're an employee of the company that nested Mozilla during its first years of life but also embeds Microsoft Internet Explorer in its products. Can you tell us how you see the current work on IE7 and the forthcoming Browser War 2.0 ?
I'm worried about IE7. I'm ecstatic that Microsoft is developing IE again, but I've played with the betas and it feels half done. I know it's still beta, but I've seen the posts on IEBlog about what they're actually implemented from CSS2.1 and CSS3 selectors, and it doesn't come close to what's already in Safari, Firefox and Opera today. With the way they're removing CSS hacks from IE7, they're not implementing enough of CSS2.1 and 3 to achieve parity with their competitors and are creating a new milestone in the middle that's going to require a whole new set of hacks to be developed without giving developers the standards they're clamoring for. It seems like a lot of trouble for a browser that's still not as standards-compliant as Firefox, Safari or Opera. And that makes me sad. Of course, that's my opinion and has nothing at all to do with AOL, which loves IE in all its forms - last I checked. I'm really curious to see the adoption rate for IE7. People didn't upgrade to IE6 in droves until there were numerous "fatal" security warnings about IE5.5. With so many "average" users using versions of Windows that IE7 won't be available for, and with no plans to upgrade, I think the adoption rate for 7 will be even slower than that of IE6 (it took at least two years for IE6 to overtake IE5.5, right?). We're looking at having to support IE6, which is already five years old, for at least another five years. That REALLY makes me sad. I think users of older versions of Windows will start switching to other browsers when the adoption rate of IE7 reaches a point where developers can start dropping support for IE6 (for some that will come sooner than for companies like AOL, I'm sure). When that happens, and IE6 users' experience becomes like that of Netscape 4.x users about four years ago, we'll see people pick up Firefox or Opera to be able to use those sites. I think. I could be wrong, you never know. I still want to know when IE8 comes out and if it's going to support display:table.
If there were only one restaurant on the planet, which one would it be ?
That's hard. Because L'Ermitage Du Riou (http://www.ermitage-du-riou.fr/home/english/) is all the way in France, and aren't open for lunch... OK, I'd probably still pick it. Just look at the menu, imagine how good it is, and then double that. The reality is way better than anything you're imagining. And, the best part is, I'm going there next week - at least once!
Thanks for your time, Kevin !
No problem. Anything for you, glazou.