Molly Holzschlag and Roger Hudson discuss web standards, interoperability, HTML5, IE8 and web professionalism at the Webstock Conference in New Zealand.

Video and audio versions of Molly’s Webstock presentation, “Why Web Standards Aren’t” are available on from the Webstock site.

Molly Holzchlag in conversation with Roger Hudson, Webstock 2008

Duration: 19:42
February 2008, Wellington, New Zealand.

Molly Holzschlag Interview Transcript

[SHOT OF CONERENCE VENUE: Mike Brown introduces Molly]

TITLE: Molly Holzschlag in conversation with Roger Hudson, Webstock 2008

Roger: Nice to see you again Molly and welcome to Webstock 2008!

Molly: Well, thank you very much.

Roger: Is this your first trip to Wellington?

Molly: It’s my very first trip to New Zealand, yes and it’s just fabulous, I’m enjoying it.

Roger: The conferences been going well?

Molly: Oh I’m really enjoying it, but one of the things that’s really special about it I think is that is that it’s got a great sense of passion and this unity of who we are and what we are in the web and workers of the web, I’m liking that passion a lot. I’m also enjoying the fact that there are speakers here from all over the world many of whom you know who are colleagues of mine for many years, and the fact there are a lot of local people and so it’s really bringing a lot of different people together, so it feels very much of a “webstock”, – you know like a get together of very diverse individuals with one greater common goal which is to make a very useful web and continue on innovation for the web.

Roger: In your presentation you draw distinction between standards and best practice what did you mean by this?

Molly: What I mean by that is that when we talk about web standards, I am challenging the language of that, I am suggesting that what we are doing really aren’t true standards like a manufactured standard for example which is, a good example would be a safety feature within your automobile. There have to be certain levels of safety features in each automobile before that manufactured item is able to ship and the law oversees this throughout the world. So there are international standards for that sort of thing, what this allows for, is for interoperability and quality and a level of security for the individuals in terms of what they are selling. We don’t really have anything like that in the web, what we have is a collection of specifications and recommendations that come out of the W3C as well as some other organizations depending on who we are talking about, for example Javascript is actually being standardized under ECMA. So the W3C itself, the technologies that we see there such as XML, XHTML, HTML and CSS and all the various languages that we mostly work with as the lingua franca of the web, are specifications and recommendations, these are not standards in the truest sense of the word. When we talk about W3C specifications and recommendations, there is no real mechanism to check other than conformance, so if I write an HTML document I can validate that document. But that doesn’t necessarily mean that what is valid and conforming is what is useful, usable, accessible and really works. That all comes into this ideology that we have termed web standards, (the term really emerged out of the Web Standards Project) where the idea was really about best practices and how to use these specifications and guidelines and recommendations to create great websites that are living to this ideal. I can create a website that validates and that doesn’t live to this ideal very easily okay, and so I think it’s just very important that we make the distinction that what we are really trying to achieve when we achieve web standards is actually a level of professionalism.

Roger: So in all of this, how important is interoperability?

Molly: Interoperability is the most important thing in my opinion when we look at the web and the vision that it was meant to be, was that it would be any platform, completely platform agnostic, completely user agent agnostic so, in other words whether I’m using my little PDA device or I’m sitting at a huge screen at my computer station that I should be able to access that data and this is where interoperability (software interoperability) it means that the information data that I’m sending, can be viewed in some way by all of these devices. Now the web is really meant to be that way and it can be that way, the problem is that these best practices have not necessarily been applied or adopted at a rate in which we would like to see it, a lot of that is educationally related, there are problems in education and tools and things of that nature. Building websites is really not an easy thing to do, and so interoperability is what we need in order to get back to what that grand vision was originally so that we use our best practices in order to create that interoperability, but we also need in terms of our software we need that to be interoperability where we see the worst problems and the most challenge there as within the browser, so which browser is doing what is going to determine how we work. Which actually adds a layer of extra work for everybody that really shouldn’t be there.

Roger: So without a greater commitment to interoperability, is the concept of web standards relevant still?

Molly: I think that web standards are a part of the commitment to interoperability, it’s sort of , somebody put it very well when I was asking for public definition on my blog about what people thought web standards were and he said “Standards are putting the cart before the horse”. So the two are intertwined, but until peoples skill sets balance out, until there are some more of us of a professional standard involved, and I think we are going to be at odds with the interoperability issue because we have difficulties in getting the different companies to decide they want to work together and lease those baselines and compete in other areas. If we go back to the analogy of the car, what I would say is that various user agents (the browsers), need to be promising each other that baseline commitment to certain aspects of specifications and what we would like to be standardized within browsers are there, that is their baseline, everybody follows that baseline and then innovation is built on top of that.

Roger: Is the explosion of mobile devices bringing us to a crisis point now?

Molly: I think that point of crisis has been with us since 1994 actually, as soon as the web became visual. Naturally, we can’t necessarily look at all of that and go “oh! this is a terrible thing” because of course it advanced the web, it made the web very useful for people right away, but what happens is again “the cart before the horse” we were building very complex websites without really understanding what was going to happen over time as they needed to scale, as they needed to be modified as they needed to be managed for the long term, as they began to grow right, so we have all of these challenges now in the infrastructure that came out of this experience in this huge burst of growth, you know, in the evolution. But this is part of evolutionary technology and we have to I think accept that, but we have also to be involved and constantly learning and improving and really caring for our profession.

Roger: 1994 also was the year of course that the W3C came into existence. It seems they deliberately avoided, when they were setting up W3C, to have any responsibility or control, they sort of shunned the notion that it was going to be an organization that would somehow be in control of the web. Was that a right decision back then do you think?

Molly: I think it’s really a philosophical issue, because the web is supposed to be an anyone anywhere other people kind of thing, the democratic vision, if you will, is part of that decision and I don’t want to see that changed, I think that’s beautiful it should be that way every person should be able to have their say if they wanted. And that’s really some of the value – it has tremendous amount to do with the value of what we have on the web today. So I think that “No it was not inherently wrong for them to take that ideology”, but what never really seemed to emerge was a professional, a very strong professional industry organization and we start seeing little ones coming up in different areas, Australia for example, here in New Zealand there is a web standards new Zealand there is various groups around the world that are starting to do this professionalism. None of these things really ever rose to that position where there is a professional organization that could be helpful to the people in the web profession, by providing educational resources, by providing different things that such professional organizations do.

Roger: Not wishing to sort of belittle the democratic side, but also since 1994, the Web has grown a hundred fold. Maybe it is time – I can’t help feeling – that someone was responsible; there is a need for a responsible organization:

Molly: I think yes, and I think we’re going to start to see that happen, but it’s not going to be one organization. As you sit through this conference we see that. For example Simon Willison was talking about OpenID and the various groups that are working together; this is what we really have to see. We have to see different specialities coming together in this huge mashup of technology and creativity. And I think that in order to sustain that we’re going to need to have little pieces of the pie come together and make up the whole thing. So the W3C will always have a role, I hope, but certainly its infrastructure can improve but, again, I am an advocate for professional organizations, I would also like to see some external to W3C, non partisan groups come up so that interoperability issues can be discussed freely amongst various browser vendors without the agenda’s getting in the way, so be they adjudicated or something of that nature.

Roger: In recent times, the development of HTML 5 has generated a whole lot of heat, has there also been a bit of light do you think?

Molly: A whole lot of heat?

Roger: Yeah, a whole lot of arguments and passion.

Molly: Ah, arguments and passion. I think HTML 5 certainly, when you look at it as a spec, it’s really fascinating. I mean we’re really moving away from just a markup language, to a language where there’s the creation and the relevance of API’s and all of this, so, you really have to look at it as something different. I think it’s an evolutionary, again it’s an evolutionary step; the way that people in the HTML 5, well originally the WHAT Working Group before HTML 5 came under the auspices of W3C, the folks in the WHAT Working Group really set things such as HTML 5, if it were coming out right after HTML 4.01, it probably would have been a progression, but now, it’s like a reworking of HTML because of all the different things that have happened, and the focus on application development. So we’re looking at in HTML 5, at a lot of things that will empower us on the client side with applications so that’s really good. But, the implementation part is going to be where we’re going to see problems. A good example is client side storage, right. So I mean, that’s a very, very good idea and HTML 5, a whole part of the spec is based on how to do that, but who’s going to implement it? You’re going to see again, the disparities here, and this again turns into a competitive issue on the functionality and the baseline of a browser and that’s not where the competition should be. So, there’s no unity in that piece, and that’s where interoperability in the software gets lost, and I’m very concerned about that. So whether it’s HTML 5 or whether it’s JavaScript, we also have an issue with JavaScript right now – in browsers and browsers evolving along different paths. I really would hope to see some point where the user-agent implementers get together and get those baselines into place; agree on those baselines, abide by those baselines and then find their own ways to compete.

Roger: Do you think some of the lack of communication has led to, I know a lot of people in the accessibility community have been very upset by some of the stuff in HTML 5

Molly: Well, it’s something very specific, it’s really for the canvas issue and other things in HTML 5 that accessibility folks are worried about and I think… where you have that, you have… There are problems in the HTML 5 process from the get go. You heard me speak about the authorship as being largely driven by one single person, so this isn’t to say that there aren’t collaborators and it’s certainly under the W3C. There are a lot more people involved in a more formal way. But, the fact of the matter is that the majority of the spec was written by one person and a handful of supporters, and that would be Ian Hickson and he’s also the person that has written Acid2 and Acid3 tests. Certainly a very bright guy, but it worries me that one individual would have that much sway over a given specification when they are not working toward an interoperable environment. I think that’s more critical so that if it’s only one person, that also means one agenda and I don’t know what the agenda is, though I’m sure to Ian, it’s making the best Web that it can be because he’s a very hardworking dedicated person. But I don’t think that is the way specifications should evolve and I don’t think it’s going to help with interoperability because it cuts the dialogue and it creates cabals, right. It creates these separations where “I’m with these guys” and “I’m with these guys,” and “but, no, I’m over here”. So there’s separation not integration. That’s a huge part of the problem.

Roger: I want to look for a moment at the recent sort of area of the development of IE 8. What’s been going on there?

Molly: Okay, so essentially what I can tell you a little bit about, in so far as my understanding of the hot topics of IE8 and what’s going on there, is that we’re seeing a shift away from the ‘Trident’ engine. So the ‘Trident’ engine has some core problems with it. It’s an older engine, of course there wasn’t any change or innovation for that long period of 5 years, and that caused a tremendous problem for Microsoft in terms of software, engineering, and trying to figure out how to catch up with everybody else. So one of the things that they have done, is that they’ve now begun to create a very strong engine, that is capable of supporting in this case a lot of stuff, like all the CSS 2.1 stuff, Acid2 rendering, and all that, so a tremendous amount of support for things in IE8 but there’s still these legacy issues and for Microsoft, that’s a critical piece because so many intranets, and so many public websites have been built using that as the model. What Microsoft risks, is that it will “break” the Web if it changes its browser too abruptly. So what ended up happening is an opt-in switch. So that individuals, so the creation of this opt-in switch if you add a meta element that says if I want my browser to operate in compliance mode, which is the standards mode, very similar to something we used to have, not used to have, it still happens in doc type switching, standards mode and quirks mode, basically, this is a different level of that; a different layer of that; where if we have the meta elements in there, then we can force that browser to render in the best way possible. Otherwise, it’s going to default to an earlier version, probably IE6, I’m not clear yet whether that is factual, but that is what I’m led to understand at this time: Which of course has caused some very big concerns in the industry. Why isn’t it going to default to whatever the latest and best would be beneath that, but this has to do with the ‘don’t break the Web’ ideology in the IE group within Microsoft because they have that responsibility to many, many billions of customers.

Roger: The last few years, there’s been a huge boost in social networking sites, is that presenting us with some challenges?

Molly: Oh well, I think clearly it does. I mean when you talk about security, you talk about identity, all of that comes into play so what’s really fascinating about social networking to me is that it is extraordinarily powerful, and yes, in the wrong hands, extraordinarily dangerous. So we have to be very responsible I think, again, this is where I’m hinting at the idea professional organizations that set a level of professional standards for people working in the industry that say we’re only going to do these best practices because we want to keep the Web safe. You saw Simon talking a little earlier about OpenID and some of the ways they are doing that through encryption. So, there are some technologies out there that are really going to assist us with that. But, there are a lot of concerns no doubt.

Roger: So how do you see the web developing over the next few years?

Molly: It’s interesting because I’m not a fortune teller; not a very good one anyway, and I tend to be leery of making predictions, but one of the things that I think is certainly going to happen is that we’re going to see clearly a lot more video, audio, more mashups, and more complex things happen technologically, and perhaps I hope a great improvement in user experience. And, of course with more broadband over time that even will increase more. But, my concern for the future has less to do with what is actually going to end up in the technological realm, than what is happening in the professional side of things. We need to keep the community learning, we need to keep the passion alive, we need to keep the ideologies alive, and that happens through social network, through conferences such as Webstock, bringing the world together in a greater ideology of professionalism. Because if we have professional understanding and set the bar higher for us, ourselves, then we are challenged to create better things; to solve problems more efficiently, and to create more innovative products. So I think that’s really where the focus is, it needs to be on the education, the community, the social network and the professional network as we move ahead – supporting each other via community.

Roger: And, what’s next for Molly?

Molly: I think a couple of days of rest.

Roger: Many, many thanks for your time today.

Molly: Thank you so much

Roger: And thank you very much for a great presentation

Molly: Thank you, it was truly my pleasure

Roger: Thank you.