The SitePoint Podcast: HTML5 is a (Beautiful) Mess

The SitePoint Podcast

Kevin Yank is joined by Opera Software’s Bruce Lawson, SitePoint author Ian Lloyd, and Kyle Weems, creator of the CSSquirrel web comic, to discuss the latest uproar from within the W3C HTML5 Working Group. Is progress towards the HTML5 standard at risk of derailing, or is this just par for the co.

Running time
0h48m
File size
45.00MB

Download Original File | View original post

Episode synopsis

Episode 44 of The SitePoint Podcast is now available! This week, Kevin Yank (@sentience) is joined by Opera Software’s Bruce Lawson (@brucel), SitePoint author Ian Lloyd (@lloydi), and Kyle Weems (@cssquirrel), creator of the CSSquirrel web comic, to discuss the latest uproar from within the W3C HTML5 Working Group. Is progress towards the HTML5 standard at risk of derailing, or is this just par for the course in the wild, wild world of standards development?

Listen in your Browser

Play this episode directly in your browser! Just click the orange “play” button below:

A complete transcript of the interviews is provided below.

Download this Episode

You can also download this episode as a standalone MP3 file. Here’s the link:

Subscribe to the Podcast

The SitePoint Podcast is on iTunes! Add the SitePoint Podcast to your iTunes player. Or, if you don’t use iTunes, you can subscribe to the feed directly.

Interview Transcript

Kevin: January 15th, 2010. A panel of experts considers the latest clashes within the HTML5 working group. I’m Kevin Yank and this is the SitePoint Podcast number 44: HTML5 is a (Beautiful) Mess.

Today I am joined by a number of prominent voices in the web standards community. We’ve got Bruce Lawson, who’s a web standards evangelist with Opera Software Developer Relations.

Bruce: Hi Kevin.

Kevin: Hi there. You’re one of six authors collaborating on html5doctor.com, is that right?

Bruce: Six and counting, yeah.

Kevin: Six and counting, it’s like a Q&A site focusing on HTML5, answering people’s questions, demystifying the process. Is that fair to say?

Bruce: It is yeah. It’s because specs aren’t written for web authors, specs are written for great gurus, so we’re trying to do the translation.

Kevin: Good. And it is not a medical advice site.

Bruce: (Laughing) No, no, no, definitely not.

Kevin: I am also joined by Ian Lloyd who is the author of SitePoint’s book for beginning web developers, Build Your Own Web Site the Right Way Using HTML and CSS. The second edition is out now.

Kevin: Hi Ian.

Ian: Hello.

Kevin: Ian is also the author of SitePoint’s The Ultimate HTML Reference and the online version of that was just recently updated with the latest browser support information and that is available for free online at reference.sitepoint.com.

And finally today, we have Kyle Weems who is CSSquirrel—or is that pronounced CSS-squirrel?

Kyle: Just CSSquirrel, which is a number of huge confusions with people, yes.

Kevin: C-S-S-Q-U-I-R-R-E-L on Twitter and the site is cssquirrel.com. Kyle is a standards minded web designer, web standards commentator and creator of the always biting CSSquirrel web comic.

Kyle, your comic this week, The HTML5 Show is sort of the impetus for me for putting together this podcast, so welcome.

Kyle: Oh, thank you.

Kevin: So as I said, this is a timely podcast. In the past week, the web standards community has been consumed by this … I guess you could call it a scandal. On today’s show we’ll be focusing on this huge issue that seems to have pre-occupied every web developer I know this week and that’s Conan O’Brien threatening to leave the Tonight Show.

Kyle: (Laughing)

Bruce: Who?

Kevin: (Laughing) I think Kyle may only be the only one who got that one as a US participant, apologies to the Brits on board today.

Ian: Yeah, all we want to talk about at the moment is the snow.

Kevin: Snow and more snow up there. Is it true that your country is entirely white this week?

Bruce: Absolutely true.

Ian: Well it was a few days ago, there was an absolute awesome photo that the NASA satellite had taken of us completely covered, so it just feels like it’s never ending and it’s only been a week. We’re a bunch of wimps.

Kevin: Making your way around the tunnels underneath the snow, I understand.

Ian: Absolutely.

Kevin: But seriously, last week’s podcast here at SitePoint, we touched on HTML5 and somebody experiments that developers are trying to play with this spec that is still very much under development. But in the past week, we’ve had fresh controversy erupt from within the W3C HTML5 working group both from within and without—and that’s part of the problem—so I wanted to tackle that on today’s show. I’m trying to paint a picture of what that is, what it means, whether we should be concerned. The development process of the HTML5 spec has been tumultuous from the beginning I guess it’s fair to say. Certainly Kyle here has made a brisk trade of drawing comics about the clashes of personality and opinion that have come about. Bruce for those who don’t follow HTML5 closely, can you give us a rough idea of where that spec is at?

Bruce: Who-hoo-hoo… Yep. First of all, let’s go back to how it began.

Kevin: Sure.

Bruce: Because if we start at the very beginning, it’s a very good place to start.

Kevin: You can’t understand how it got to where it is now without understanding the mess that it sprang from.

Bruce: Indeed. I think 2001 or 2002, the W3C put the finishing touches on HTML4 with a point revision called HTML4.01 and then announced that they were abandoning work on HTML because the future was XML, they said. Many believed them, and I did personally, but then they kind of went off and dreamed up this crazy thing called XHTML2 which was a spec—I’ve described it before as a spec of fantastic and beautiful, philosophical purity that had no relevance at all to the real world. It had no relevance to what the browser manufacturers were doing and little relevance to what authors wanted to do and it broke backwards compatibility so severely that at one point it even depreciated the <img> tag. They brought that back but that’s where they were going.

And so a group of people who worked for browser manufacturers—it started at Opera for whom I work but we were soon joined by people in Mozilla and then Apple—started writing in their own personal time a guerilla spec called WHATWG. I think that stands for Web Hypertext Application Technology Working Group. They called this spec Web Applications 1.0 because it was a spec designed to extend HTML so that it would work for web apps, you know things like Yahoo! Mail and Google Maps and Google Docs and Picnik, the photo editing software, websites that are applications rather than static documents. They were doing this stuff in their own time and then Tim Berners-Lee said “We made a mistake at the W3C and we want to go and reach out and re-start work on HTML”, and a vote was taken and they decide to ask the WHATWG if they could use their spec as the basis for what’s being called HTML5.

So HTML5 is a word dreamed up by the W3C and the WHATWG called it Web Applications 1.0 and the specs have been developed by the two groups at once. So it’s been developed by the W3C and by the original custodians, the WHATWG and it’s entirely fair to say there’s been quite considerable culture clashes between those two groups which has kind of come to a head now but the last week’s shenanigans and not really any worse than what we’ve seen before and we’ll see them again, in my opinion.

Kevin: So what was on paper a great reconciliation—or at least we hoped it was a push in that direction—it hasn’t really panned out that way. Is that fair? I mean it still seems to be two groups in a lose association.

Bruce: Absolutely. I don’t know that there ever was any great reconciliation. So for example when the W3C decided to end work on XHTML2 even though it wasn’t complete, they did so because they were kind of acknowledging that the WHATWG had it correctly and there were certainly some members of the WHATWG who were being, in my opinion, unnecessarily triumphalist, if you like, and kind of almost crowing about this victory over the W3C.

I work for Opera and we’re sitting on lots of W3C committees. I sit on a working group myself, I’m a web standers evangelist. I’ve got lots of time for the W3C but it’s fair to say that W3C took— they really dropped the ball with HTML and it needed the WHAT Working Group to give them a good kick at the backside to make them realize that HTML wasn’t dead and can’t die and it can’t break backwards compatibility but there was never a reconciliation. It’s always been a very uneasy meeting when the WHATWG and the W3C come together.

Kevin: So that brings us to this week, you said the shenanigans this week were no worse than what we’ve seen before but— Well Kyle, your comic this week, The HTML5 Show paints a picture of the whole process having degenerated into something akin to the Muppet Show. From your perspective, are things getting worse or better?

Kyle: I think they’re a little better today than they were when I first made it.

Kevin: Okay.

Kyle: But that said, I don’t think it’s going to stay that way. I think it’s going to continue to break down as a process.

Kevin: So what’s going on exactly?

Kyle: I mean, you have these two organizations as it’s been said with very different philosophies and what to do about the spec and it kind of feels to me like the WHATWG doesn’t really think of HTML5 as HTML5 anymore. Under Ian Hickson, the editor, they’re kind of moving forward. They have made their last call on this. They’re ready to continue to add things to a “versionless” HTML. Meanwhile the W3C which hasn’t hit last call with the spec is still trying to solve what are the targets are for HTML5, specifically.

Kevin: Right.

Kyle: And when they’ve made decisions via contributions of a large amount of people regarding what should be in the spec, what shouldn’t be in the spec. There’s been some pretty severe disagreements at least from what I’m viewing from people like Ian, and since basically he’s the editor-in-chief as it were, if he disagrees with that he’s chosen that, ”Well, it’s still in the spec but you guys cannot have it in yours over there if it makes you feel good.” I mean it’s still moving on. It sounds like there’s a breakdown of communication where it’s like even if they do understand each other there’s no kind of like closing the gap to make nice anymore. At least that’s how it feels to me from watching the various mailing lists and chat logs and whatnot.

Kevin: It’s a bizarre situation that we say there are these two camps, there’s the WHAT Working Group and there is the W3C working group, but in fact Ian Hickson is the appointed leader of both those groups.

Bruce: Yes, although he co-edits— The W3C spec, is co-edited with Dave Hyatt of Apple, I believe.

Kevin: Right.

Bruce: Although I think Ian is the sole editor of the WHATWG side. Although I’ll be frank, my interest is primarily the tech and not so much the politics because I don’t understand politics very well and it’s a full time job just following the politics.

Kevin: Well Ian, as someone who regularly writes for beginning web developers and people actually getting work done as opposed to people obsessing about politics. What effect does the tone of the debate that we’re seeing here have on your approach to HTML5?

Ian: Well, if Bruce says he’s not sort of following the politics then I’m completely in the dark because I’ve kind of made a slight conscious effort really to sort of steer away from the very fine details of HTML5. I definitely have an interest in it, I definitely have an interest in it from a point of view that I have an interest in doing anything I do right and following the standards but I’ve tried not to get too bulked down in fine details because I know it’s something that’s always shifting. I’m just really, really looking forward to the day when I can actually pick up a book that has everything in it and I can trust it and rely on it and that’s ultimately when I’m going to really start to kind of throw myself right in with HTML5.

From my point of view, I’m in this situation at the moment at the place where I work, where I’m trying to put together some standards for projects that will not see the light of day for probably a couple of years and what we’re talking about here is something that will have a web front end but it’s part of a much, much bigger project. So I’m faced with this situation where I’m trying to define some standards, not knowing where we’re going to be in two years’ time and knowing what can I safely say now. What can I say is the standard that we want to go with? And when I say standard I mean sort of an internally documented set of recommendations rather than sort of a proper standard if you catch my drift.

And so this has been the problem I’ve had with HTML5. What can I say now that is going to be safe? And every time I hear about HTML5, it’s not because I’m watching any of the IRC logs or anything like that because I don’t. The only time I ever really hear what’s going on with HTML5 is through Twitter and unfortunately, most of what I get to hear from Twitter is the moaning and the bitching which was obviously what’s happened in the last week. It’s been really interesting and funnily enough that the page that I really found out what was going on with regards to the most recent shenanigans was what Kyle wrote.

Couple of days ago, Dave Shae on Twitter, he was really looking for someone to summarize what all this was about and so it’s good to finally find out what the shenanigans were about but I don’t think I really fully understand the politics and I probably don’t actually want to.

Kevin: Right. We certainly see a lot of, sort of— I see a lot of useful summaries coming out of the WHAT Working Group. They like to post things like weekly summaries of what they’re working on. But that as we’re seeing, they are turning their focus towards the future. It’s like they— I don’t want to be unfair, but it seems to me like that WHAT Working Group, what they’re interested in is blazing the trail, setting a direction for the next thing and working out all the details about how the last thing that they created is going to form a coherent standard, isn’t as much of interest to them.

Bruce: Don’t forget Kevin that a lot of what we can call HTML5 was actually about documenting the actual reality that had never been written down before, so part of the effort for HTML5 was the first ever specification that documented XMLHTTPRerequest which Microsoft invented and we all use every day with Ajax and nobody ever written it down, things like Canvas which Apple invented and everybody reversed engineered into the different browsers was specified, things like ContentEditable, which was a Microsoft invention was specified because all the different browsers had implemented it. So a lot of what they did actually documented and specified and codified what we already have been using for a long time. There are definitely fantastic new ideas in the HTML5 spec as well but it’s not as if they’re just dreaming up new stuff and running on. I think they think that a lot of things they’ve done has been writing down what we currently use and now they’ve got that foundation there, they can start to move on. Again, I’m not part of that group. This is my personal take from outside that group.

Kevin: Go ahead, Ian.

Ian: Sorry, I was just going to say that there’s a nice little phrase they always use for that, isn’t there Bruce? The “paving the cow paths”.

Bruce: Indeed.

Ian: I believe that’s what’s always put forward.

Bruce: Yeah, that’s the essential philosophical and practical difference between HTML5 and XHTML2. HTML5 said for better or for worse, this is where we are. Let’s document it, let’s codify it, let’s standardize it so if somebody else wants to make a browser tomorrow, it’s all written down for them. Everything could be interoperable and then let’s build on where we are rather than go off into a fantasy world in which XML rules the world and we can design the Web from scratch and break backwards compatibility because there’s umpty-nine squillion web pages out there and we can’t afford for them to break. We need them to continue working so we are where we are.

Kevin: So thinking about Ian’s effort to sort of set his own company’s standards for what they’ll be building in a couple of years’ time. Bruce as the official standards body representative here today, it sounds like you’re advice would be to ignore the noise and the grand plans of people like the WHAT Working Group and focus on the serious standards that are being built at the W3C right now. That’s the stable core that you can focus on if you need something to hold on to.

Bruce: Yes. First, let me say I’m not representing any standards body. I work for Opera but I’m not part of the team that’s writing the HTML5 spec. My interest is as a developer, I came from a developer background.

Kevin: I will call you the defender in spirit of web standards, then.

Bruce: Okay, okay. I’m one of the WaSP, but I’m not a W3C guy, nor a WHATWG.

Kevin: Sure.

Bruce: Yes, if you look at the WHATWG version of the spec on WHATWG.org, you’ll see handy little status bars against each element or section that tells you where implementation is in the browsers and some things like Canvas are excellently implemented across four of the five big browsers, other things like ContentEditable, you can use across the board. Other stuff, like Web Sockets, aren’t implemented anywhere and so it would be folly to…

Ian: Microdata.

Bruce: Microdata. …it would be folly to suggest those as an internal standard but it’s not difficult to go through and work out which parts of the specs are close to real world or actually just merely document what we’ve all been using for ages. About the noise, I mean Ian mentioned that he was following a lot of stuff on Twitter. I think there is… We live in a noisier environment than we did when HTML4 was being written because we can all Twitter intimate thoughts 900 times a day like I do but also I think it’s naïve of people to assume that there is more politics now than there used to be. The thing is with the W3C, it was always a member organization in which people sat down in smoky rooms and argued with each other and arm wrestled and smiled while they kicked each other in the balls in hotels, face to faces.

What we’ve got now with the WHATWG is they started off a process in which anybody can join and anybody can comment and Ian Hickson has always said he doesn’t care where a good idea comes from. It can be from a W3C member group or it can be from an individual in Bangalore or Berlin or Birmingham. It can be posted on IRC or Twitter or a blog post. If it’s a good idea, it’s a good idea. So I think the extra noise and the idea that there’s more politics is actually a function of a) our Twitterized society, and also b) the fact that there’s just a hugely open process whereby bickering happens out in the open rather than in Silicon Valley hotel lobbies.

Kevin: Having spoken to people who’ve been on W3C working groups before, when they did this closed version of the process, it definitely seems like the new HTML way of doing things, if nothing else, is better for the mental health of the people involved in the debates. But from your point of view, is this openness damaging the spec in the eyes of the greater web community?

Bruce: In my personal opinion, if there is damage in the wider community, it’s very transient and it leads to a vastly superior spec in the long term. That’s my personal perspective. I don’t know what the other guys feel.

Kevin: Would you go along with that, Ian?

Ian: It depends what day it is you ask me really. I mean some days I’ll be very enthusiastic about what I can actually implement with HTML5 and what’s safe to use now and then other days you do have these kinds of flaming arguments that go on and make me think, (sigh) “Why do we bother?”

Kevin: What I’m wondering is would you rather hear all of this noise and take it or leave it or do you prefer the way things used to be that we just got these nice well-considered, well-worded specs for us to consider as a whole in draft form and submit feedback to in a civilized way. Is it nice to know what they’re arguing about or just that they are arguing on our behalf?

Ian: I think overall, it’s probably a good thing that we are hearing what the arguments are about, I mean you could always say use the expression “ignorance is bliss” if someone puts something in front of you and says this is how it should go, you say, “Oh, I’ll accept that. That’s fine.” But then if you have a problem with it then it’s just too late to do anything about it. So it’s good that people are having these arguments or discussions about where it’s right and where it’s going right and where it’s going wrong. If we cast our minds back to what was happening with the accessibility guidelines with the WCAG2 and that went through absolutely massive, massive changes and it wouldn’t have ended up being the documentation that it is now had they not been the almighty storm that was kicked off about it. And so when this happens with the HTML5, you got to think well, there are a few bumps in the road but ultimately it’s got to be good for the final result.

Kevin: Yeah. I know Kyle, you must be all for this public process. You wouldn’t have anything to draw comics about otherwise.

Kyle: Right, I’d still be trying to make up more jokes about Opera if HTML wasn’t so noisy – and no offense, Bruce. I think process is a good thing because in addition, it has given me something to draw every week. Yeah, it does give in theory everybody a say in what’s going on around them since HTML5 or versionless HTML or whatever we’re going to end up calling it—Frank, we could call it Frank if we like—is going to be used probably for several years if not way past what I would think a technology should last with billions of web pages. So you know if we don’t get our say in now, then naturally we’ll just be complaining farther down the road.

I’d say the only part about it that sometimes disturbs me personally and the reason I end up drawing these angry squirrel comics is when you have the appearance of the public process and then when there’s decisions made by the majority, then you’ll have like—I love to pick on them but—Ian Hickson, the editor will then say, “Okay, we’ve had the process. The process has gone through. I’ve decided that I’m going to do it this way anyway.” and that kind of hurt me when that happens because then the process is an illusion and it’s kind of shown for what it is and then it’s like well, what does it matter what we think?

Kevin: So in that sense your comics would be a peaceful form of protest?

Kyle: Absolutely. Maybe it’s just a principle because often HTML5 for what ends up coming out of it has been good so far so regardless of the process, it might be a good product but if you’re going to have a process I’d prefer it’s at least honest.

Bruce: I think Kyle’s hit upon something really interesting there because although I’ve been talking about an open process, it’s true or it’s my perception that the WHATWG and Ian Hickson run a kind of open source process whereby Hicksy’s a benevolent dictator for life and the W3C run a process which seeks consensus. And Hicksy’s on record is saying he doesn’t seek consensus. He will weigh up all the evidence given to him, regardless of where it comes from or what form it comes in, but ultimately he will take a decision because he thinks that you can’t have a consensus based approach when you’re looking at a thousand plus participants in a working group. I can’t comment, I can’t even imagine what it must be like to have to process the tub thumping and the ranting of a thousand people like me on email everyday but it’s true I think what Kyle says that there isn’t a consensus seeking approach. Openly it’s not a consensus seeking approach, it’s, “I’ve listened to you all and this is what I’ve decided.” from Ian.

Kyle said something interesting which was that the spec will probably last a long time, longer than maybe a technology should but I think that HTML5 has already prolonged the life of HTML because things like Flash and things like Microsoft Silverlight were snapping at HTML’s heels as the way to build web applications simply because it was too difficult or clumsy or clunky to do them in HTML and JavaScript. So I think the HTML5 process is already a victory for those of us who believe that they Web should be an open standard because it’s making the tools we all use more suitable for the things we need to build and hopefully, therefore, prolonging the life of open standards for the web which I believe passionately in.

Kevin: It seems like we have three levels of openness that are possible here if we divide it into strata. You have the closed technologies like Flash—as the perpetual whipping boy here—but Flash developed by one company and released as a published spec but just one company’s opinion of how things should be. Then we have the direction we see being taken by HTML5 where everyone gets a say but what ends up being decided on and published, it seems, comes down to what one person is convinced of by everyone getting a say, rather than the consensus that the W3C would hope to achieve in it’s typical standards process. And then finally we have this ideal of consensus whatever that means on any given day, whether it’s you get 75% of people in the working group voting the same way, you call that consensus and too bad if you’re in the other 25%. So it sounds like HTML5 is a middle ground. It’s open enough and everyone is getting their say and that makes it better than what we might otherwise be stuck with, but it’s not quite the ideal we might hope it could be.

Kyle, do you want to take that?

Kyle: Uh, sure. It kinda sounds like the wild west, now, you know? Uh, we should name HTML5 after it. But yeah, I mean one thing that Bruce said that I have to agree with completely is the bit about plug-ins like Flash and Silverlight, as a person who works with code everyday and as someone who’s gotten more aware of accessibility and you look at those plug-ins and it just creates this huge, obnoxious kind of glut that appears in a web page that’s hard to work with or do things with and of course are owned by the various companies that made them. So they decide how those work individually. I love the fact that with HTML, it’s an open standard for everyone and it’s also with HTML5 doing more things that developers do. Ian just, if we could let him jump in, but he just wrote a point about this Canvas-level accessibility and that’s a good one.

Ian: Yeah, I mean you mentioned obviously with Flash, and that Flash is, depending on who you give it to it can be incredibly inaccessible or it can be very accessible but finding a developer who can do a really, really good job with a Flash movie or application, it’s not easy to get someone who can do that. But from my understanding, my little understanding of what Canvas can do in terms of accessibility that is actually presenting a big hurdle at the moment so it may be using open technologies but it comes with its own baggage. Would that be fair to say, Bruce?

Bruce: It would, yeah. I think the reason for that is that Canvas is a proprietary Apple invention they invented for their dashboard that was retrospectively codified as a spec. So no implementation of it is yet accessible but I do know that a couple of the guys I work with in Opera are interested in making Canvas accessible and I introduced them to the guys whose job it was in Adobe, or Macromedia as it was, to retrospectively make Flash 5 accessible, and they have been talking. And this gives me an opportunity as well to say that I don’t want to engage or I don’t want to be able to think I’m engaging in Flash bashing here. Flash is a cool technology.

Somebody said to me when I spoke at a conference last year, something that struck me as really profound and I’ve been looking for a reference for it, and this guy told me that this was research done on the top 100 companies in the world in the year 1900 and only three of them were still around in the year 2000. And I thought it doesn’t matter how good any one company is, if the Web’s going to last, it can’t be entrusted to any company, no matter how benevolent that might be because you just never know if that company is going to be around, and wouldn’t it be appalling if all of the history and all of the stuff that we built on the Web disappeared into the ether because one company went under. So that’s why I’m convinced that it should be an open standard not because I want to bash any one particular corporation just because I think the web is too valuable to be in the hands of any one company.

Kevin: So what then is the distinction between one company Adobe, building Flash and releasing it, as eventually releasing it as a file format that is documented for other people to implement if they see fit versus Apple bringing Canvas to the WHAT Working Group, proposing it there as something that any browser can implement. What’s the distinction because it seems very subtle?

Bruce: It does and the main difference I think is that Canvas for whatever reason, it”s before my time, was copied by the other browsers, it’s native to them so you don’t need a plug-in to use Canvas, whereas with Flash, the browser has to pass the data to a plug-in. So for example if you’re passing off video data to a Flash plug in, it’s a black box. If it crashes, you can’t control it, there’s nothing the browser manufacturer can do whereas with native video, you can manipulate it any which way you choose so there’s nothing in my opinion wrong with Flash as a product. It’s just for historical reasons, it’s a plug-in and Canvas lives natively in four of the five browsers and therefore it’s an open standard.

Ian: You mentioned something today Bruce on Twitter that one of the biggest hurdles that you saw with this kind of rich media was actually the lack of visual IDEs. So with Flash there are tools that have been there for a long time, Silverlight maybe, not everyone’s cup of tea but they’ve worked hard on developing the tools, whether it’s something that you want to use or not. But if you want to create something that’s going to be outputted on Canvas then where do you start?

Bruce: Well it’s interesting because immediately after I Twitted that, I got a Twit back from a guy who actually works for Opera who’s on the SVG working group which is a technology that’s similar but not the same as Canvas and he pointed me to a demo by an Adobe engineer of a guy importing an Illustrator file into Dreamweaver which Dreamweaver then turns into Canvas, which was fascinating and I didn’t know existed. So that’s really encouraging because the blog post that I was pointed to, the Adobe guy that made the point that Adobe doesn’t make the money from distributing the Flash player. Adobe makes its money from selling authoring tools, so if Dreamweaver and Illustrator and the Flash tool can produce Canvas just as easily as it can Flash then of course, there’s still a great incentive for people to part with the money to buy Creative Studio because at the moment you want to write Canvas, you got to write JavaScript and not everybody wants to do that.

Kevin: So bringing this back to HTML5, Kyle through your comics, I think everyone is more than aware of your opinions on the standards process but as far as the actual web development work that you do day to day, what is you approach to HTML5? I mean we’ve heard on this podcast from developers who have started using HTML4 with class names that match HTML5 elements as sort of a forward looking approach to their markup, while other people have actually taken the plunge and started using HTML5 markup despite the difficulties that that entails with today’s browsers. How do you approach it?

Kyle: That’s actually a really good point because it kind of ties into something else about advocacy for HTML5 which I’ll try to bounce back to in a second. I work at Mindfly Website Design Studio so I report to bosses with opinions and whatnot, and so HTML5 can be a tricky sell especially once you start doing the whole <!DOCTYPE html> bit because certain creaky old browsers that haven’t quite left the Internet yet can give you problems, severe problems when you start marking up in it. In particular Internet Explorer 7 and younger, or older, are fairly problematic when it comes to styling elements that didn’t exist when it was made like header and footer and what not.

Kevin: Right.

Kyle: So actually what’s happened is here by and large at the work place, I’m unfortunately still coding in HTML4, XHTML1.1 or whatever you want to call it and doing the class name conventions to kind of mentally prepare myself for things like header and footer and whatnot, and then when I can get away with it when we have a client who wants to take a risk or if there’s a function that HTML5 allows in certain browsers that isn’t a hundred percent vital in others then we’ll go ahead and take that risk and use the HTML5, prepare them for the fact that certain browsers won’t be able to use those features. So from that position, I use it on my personal stuff but when it comes to making things for clients, a lot of times I have to make sure the website works for everybody and so certain features doesn’t work for everybody regardless as opposed to some sort of graduated kind of experience.

Advocating HTML5, not only in the workplace but also in the web development community and then the online community of all web developers and designers, it can be tricky to get people to buy in to taking that risk making sites for clients in exchange for money when they’re not sure about the stability of the spec and I think that’s been the hugest issue with this public process. I mean I’m glad it’s a public process but when it doesn’t appear the groups are getting along, we have an item that’s in the spec, then it’s not in the spec, or you have something that changes in how it functions, these people become really shy at the idea of jumping in and trying to adopt these features and granted it’s a spec in progress so that’s expected but it can make it a hard sell.

Ian: And following up on the idea about advocacy, just to relay a little story about where Bruce and I have, well I wouldn’t say worked together, but what happened was I had a question from a colleague at work and they were asking about implementing video in HTML5 and I started to answer the question and then kind of stopped myself and thought, “You know I’m not really the best person to ask about this.” and I got on to Bruce and said, “Would you be interested in coming down to speaking to our developers here?” In our team, we’ve got a group of about 12 developers who are—well, not all of them are web-standards-savvy but we’ve got a good proportion of people that get it and they would be enthusiastic and would be keen to learn.

Bruce was happy to come down and talk about HTML5, talk about some of the Opera stuff and give us some demos and it was really, really good. We had some real enthusiastic responses to these people and I think Bruce was happy to come along to a company where we’ve got thousands of employees. It’s a big organization but there are people that get it. And we’re starting to make baby, baby steps towards using HTML5 and I mean the tiniest, tiniest little steps. For example, we now use the HTML5 doctype. It was a no-brainer. It was just such a simple little thing to change and I did it as part of a re-burning exercise that we were going through a short while ago.

Going back to what I said earlier in the podcast about trying to define some standards and picking up from what you were saying about using class names that kind of tallied up with the HTML5 standards. One of things that I’ve actually specified in these documentation that I’ve been writing is that we should be using header and footer elements instead of <div class="header">, or <div id="footer"> and on top of that, I’ve also created some little favelets that the testing people can use to test against it so if there’s a presence of a div with an ID that looks like it probably should be an HTML5 equivalent then it will alert it, alert them to it.

So these little things that we started to implement but it’s still a long way off before we get down the road of saying using the video element. I think ultimately, what we’re going to end up with is a real mish-mash of HTML5 doctype, XHTML Syntax, Flash and possibly Canvas. But another problem that we have is you’re talking about older browsers and I guess to come here work for is typical in a lot of companies where – well, we’re still using IE6 and we’re
not going to be off IE6, I know, for probably a good couple of years when we’re going to be jumping at a corporate level, we’re going to upgrading from Windows XP straight to Windows 7 but we’re still talking another two years of IE6 and that can make it very, very problematic as a developer to say, “Well, we want to use HTML5. We can come up with these things.” They look great in the new browsers but the people have got to sign these off are still using the cruddy, old browser from 2001 and that’s probably the biggest challenge that we have in trying to drive HTML5 forward.

Bruce: Aw man, I hear you. As you know Ian, I used to work for the Law Society in England and until 2005, the web team were only allowed to use IE5 to check our work and it was a disciplinary offense to download a different browser. So I understand where you are. And now I’m going to metamorphose into HTML5 Evangelist Man, but the beauty of what they’ve done philosophically is fantastic is it’s so backwardsly compatible. So with the video element you can put fallback between the tags and you can embed a Flash movie if you want to. So a browser that can use a video element will do, whereas a browser that can’t use a video element will just fall back and import the Flash movie and it won’t have necessarily all the loveliness that native video has, but something will show which I think is fab.

If anybody listening wants to look at that, there’s a guy called Kroc Camen, I think it’s C-A-M-E-N, his website’s camendesign.com. If you look at Video for Everyone in your search engine, you’ll find it and he’s got some beautiful code, which validates as HTML5 and yet will show the video in all browsers by using Flash as a backup for IE6, etc. That’s the great thing.

Canvas, it’s a little trickier. The header and the footer elements, Remi Sharp a friend of mine and a fellow HTML5 doctor adds a little bit of JavaScript which will make those stylable in IE5 onwards, I think. It doesn’t work in Netscape 4 because I tried it when I spoke at a conference last year and it was a bit wobbly.

Kevin: Oh gosh, well that rules it out. (laugh)

Bruce: Yeah, I think we can forgive Netscape— not working in Netscape 4.

Kevin: Yeah.

Bruce: I apologize for putting my evangelist hat on, but that’s why I forgive all the politics bull— B.S., because it can do this. They’ve designed it in a way that allows for maximum backwards compatibility and also Ian, it’s not a mismatch if you’re using XHTML syntax and some Canvas and some Flash and an HTML5 doctype, you’re doing what HTML5 was designed to do which was to reflect how real developers in the real world with real customers have to work, which is exactly the same way as we use CSS. You know, we have progressive enhancement and we use things that will work in the modern browsers but will degrade gracefully in those that don’t. This is the way we’ve always worked. There’s nothing to, there’s no need for anybody to feel shamefaced and say it’s a real mismatch. We’ve all been doing that for ages.

Kevin: So it sounds like HTML5, on the surface you see tags like video and you go, “Oh, that’s going to make video so much simpler. It’s just going to be one tag.” But in the short term, it’s one more tag that you wrap around the gobbledygoop of two embedded objects with perhaps conditional comments for Internet Explorer so it adds to the complexity now as an investment in hopefully reducing that complexity in the future.

Bruce: But from a language perspective, it’s very, very simple, there’s no need to wrap it around objects and embeds. From a having to serve IE6 and IE7 and IE8 perspective you have to do that but there’s nothing in the language that’s terribly complex there but at least it has the ability to be backwards compatible. When you compare that with what they were trying to do with XHTML2 you can see why the W3C effort, in my opinion, my personal opinion, was rather in the doldrums and it was so good that the WHATWG came along. (Reciting) This is my personal opinion and it does not reflect the opinion of my employer, etc.

Kevin: Well, it is clear that HTML5, if you eliminate all the noise, if you look through that. It’s a good thing and it will make the web better. Even if they stopped work on it today, it’s already much better than what we’ve had before and I suppose if that weren’t true people like Kyle wouldn’t be so upset when there were signs of the process going off the rails.

Kyle: That would definitely be right. I mean if it wasn’t a good product I probably wouldn’t be paying attention, people wouldn’t come read a comic about an angry squirrel talking about it. Regardless of the process it appears to be to me, there is no denying that I rather enjoy the product. It’s a tough sell right now to some of the shy-er developers out there which is more people than I would’ve imagined but I’m really excited for what can be done with it even today, let alone tomorrow.

Kevin: Well, thank you guys. I think that’s a great place to end and it sounds like we’ve concluded that the future of HTML5, while not smooth but it does at least seem to be safe for now. So thanks again Bruce Lawson, Ian Lloyd and Kyle Weems for joining us today.

Bruce: Thank you.

Ian: Thanks guys.

Kevin: And thanks for listening to the SitePoint Podcast. If you have any thoughts or questions about today’s interview, please do get in touch.

You can find SitePoint on Twitter @sitepointdotcom, and you can find me on Twitter @sentience.

Visit sitepoint.com/podcast to leave comments on this show and to subscribe to receive every show automatically. We’ll be back next week with another news and commentary show with our usual panel of experts.

This episode of the SitePoint Podcast was produced by Karn Broad and I’m Kevin Yank. Bye for now!

You can find SitePoint on Twitter @sitepointdotcom.

Have any thoughts about this interview? Please visit us at sitepoint.com/podcast to leave a comment on this show and to subscribe to receive every show automatically. Email podcast@sitepoint.com if you have any questions for us; we’d love to read them out on the show and give you our advice.

This episode of the SitePoint Podcast was produced by Karn Broad. Thank you for listening and have a happy, healthy, and successful 2010.

Theme music by Mike Mella.

Thanks for listening! Feel free to let us know how we’re doing, or to continue the discussion, using the comments field below.

Related posts:

  1. SitePoint Podcast #33: Team Opera at WDS09
  2. SitePoint Podcast #11: Interview with Microsoft’s Chris Wilson

You might also like...

Comments

Contribute

Why not write for us? Or you could submit an event or a user group in your area. Alternatively just tell us what you think!

Our tools

We've got automatic conversion tools to convert C# to VB.NET, VB.NET to C#. Also you can compress javascript and compress css and generate sql connection strings.

“Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it.” - Brian Kernighan