I managed to keep that promise and I’m glad I did.

I want to summarise the event. I found some talks particularly though provoking and  they will probably be the subject of more detailed follow up posts.

I arrived in Sydney on Wednesday, a day early for the event, and enjoyed the Sydney hospitality, and beer, with some friends. It was a nice relaxing start to the Web Directions experience and reminded me how nice Sydney can be.

Thursday 8 Oct 09: Day 1 of Web Directions South.

Escalante

Day 1 keynote
Matt Webb

This guy was interesting and inspirational, but I just don’t know what to make of it.
The Here and There maps were cool. And I liked the way he drew inspiration from science fiction.
I suppose the take home message here is to look at things from a new perspective.

Font embedding and typography.

Mark Boulton

I liked this guys idea that the historical poor state of fonts on the web has forced us to have better typography. Making the most of what little we had.

I’ve often been frustrated with designers who come from a strong print background and think that the web is the same beast as print. This guy was a passionate designer, but he was realistic about the web and it’s limitations. At the same time, he was not afraid to challenge those limits.

His main call to arms was directed at the main browser vendors and font foundries: Make new cool fonts specifically designed for the web.

The state of developer tools

Ben Galbraith

This was a great technical overview of some of the great developer tools available now for writing code. Tools have evolved  far beyond  the old alert(); dialog for debuging JavaScript.

In general, he says, the tool available in Firefox (Firebug), Opera (Dragonfly), Webkit (Webkit is the render engine behind Safari, Chrome and Adobe Air), and IE8, are all on par with each other. The webkit developer tools got a special mention for looking the prettiest. While the IE8 tools got a special mention for having a heap explorer so you could see how you code was executing.

An interesting point here was that IE8 provides low level developer access right into the heap. FireFox abstracts this away to the point where the same kind of low level access is not available. IDEs could use an IE8 style API to run debuggers right in the IDE. FireFox cannot provide this until the open up an API to this kind of functionality, although they are working on this, he said (Check out the Web Debug Protocol working group).

Something else he said FireFox was working on is Web Workers. Web Workers are like threads (in a browser, FireFox) but are stateless to reduce problems with concurrency.

I think the take home message here was the JavaScript libraries are good and mature, and the tools have developed to the point where using JS libraries is a lot less painful than it once was.

Canvas

Dmitry Baranovskiy

Wow. What a great speaker. This guy was probably the most entertaining speaker of the whole conference. And I enjoyed what he had to say as well.

He was critical of the Canvas element while providing a good overview of it’s capabilities. He suggested the best example of development within Canvas is  Bespin

His main gripes with Canvas seemed to be that it had no DOM associated with it, and that it’s API is non standard.

Since Canvas just provides an API for painting pixels on the screen, there are no DOM objects that can be inspected and manipulated the same way everything else in the browser can. Seems to me that this means it suffers all the same problems Flash has, but without being a plugin. But since Canvas currently lacks wide spread browser support, I’m not sure the distinction means much.

He criticised the API for being non standard. Well, it has an API and it can be used, but it’s a very verbose API. He made a good point that it looked exactly like the Java AWT Graphics API. I wouldn’t be surprised if it was a Java developer over at Apple who simply ported AWT to JavaScript and called it Canvas. Perhaps this will get better in the future, and the door is wide open for somebody to write a library to wrap all the calls in nicer JavaScript flavoured functions.

SVG would be a DOM oriented alternative to Canvas, and Dmitry has written the Raphael JS library for SVG.

The state of the web as a platform

(panel discussion)

As interesting as it was to hear the guys on this panel talk, I’m not sure if there really was a take home message, or if they ever actually agreed on anything.

Best line, spoken by Microsoft Guy after Adobe Guy complimented Microsoft on something: “Who are you? And what have you done with the Adobe Guy?”

Making Waves

Day 1, Closing address
Cameron Adams

This was all about the making of Google Wave, for a designers perspective. It was an entertaining presentation (is it a coincidence that the designers all had the best looking presentations?) and gave a great peek at the iterative development process around such an ambitious project.

After Camerons talk, at the end of day 1, they presented the winner of the McFarlane prize. I was pleasantly surprised to see my old employer Swinburne won the prize. I know a lot of people did a lot of work to get it to where it is (including Mark), but well done Caroline.

WorkFLOW

Day 2, opening address (9 Oct 09)
Kelly Goto

There were some big ideas in this presentation. I don’t think I can do them any kind of justice in a paragraph or two. I think this talk is certainly marked for further consideration in it’s own blog post.

In a nut shell, Kelly Goto spoke about flow. Not work flow as in the mechanical steps of getting a job done, but flow state. That state of mind where things just click and you get great things done. Kelly suggests flow state is achieved at a sweet spot between challenge and skill level. If the challenges of a job are too easy, you get bored and can’t achieve flow. If the challenges are to great, you achieve anxiety instead of flow. If you skill level is too high for the job, you also arrive at boredom. So you need to find challenges appropriate for your skill level.

RDFa

Mark Birbeck

This wasn’t the most exciting presentation of the conference, but it was interesting enough.
I’ve long preferred microformats because of their simplicity, but this guy has got me seriously thinking about RDFa.
First and foremost, he made it look easier that I thought it was. That’s good, since it’s complexity is one of the things that was scaring me off. Perhaps there are too many overly technical descriptions for what RDFa is, and not enough grass root “here’s a simple example” sites.

Microformats is growing beyond it’s means. It started simple, and as it grew, people were attracted to it’s simplicity, but then began to run into it’s limitations. So they wanted it to do more. The more they ask it to do, the further it moves away from it’s simplicity. Since it works by hijacking regular markup and applying a non standard semantic to them, it can only ever go so far without either running out of way to twist the normal markup language, or needing to introduce name spaces to define it’s own semantics. It it has to resort to new name spaces, then it’s crossed the line and you might as well be using RDFa.

The other thing that RDFa has going for it is that it’s part of the specifications now. Moving forward, this could be a great boon to RDFa. Assuming enough people take it up to give it momentum, being in teh spec will give it a huge advantage over microformats. Of course search engine support, with or without specifications, may also make or break RDFa. If google doesn’ get behind it, regardless of whether it also gets behind Microformats, then it’s idea in the water.

I wonder if it isn’t too early decide this one?
In the bigger picture, both RDFa and Microformats show that the future of the semantic web is bright.

progressive enhancement

Elliot Jay Stocks

This guy used the term “progressive enhancement” in a different context to what I’m familiar with. I always think about it as it relates to JavaScript. Eg: only introduce a new JavaScript feature in a site if that browser can support it. But he expanded the term to take in CSS features as well. It was more about visual design of a site.

Some quotes:

Progressive enhancement = reward
Graceful degradation = punishment

I think he meant that enhancing a site was giving a reward to those people who were using modern browsers. Degradation means building a site with all the bells and whistles, then punishing users with old browsers by taking those nice functions away.

Don’t underestimate the importance of beauty

the nicer your site looks,
the longer your visitors will look at it
the longer they look at it, the longer they stay on it.
the longer they stay on it, the better your chance of selling them something

“forward thinking code is future proofing your web site”

Pervasive Computing

Rob Manson

I should have taken note of the fact that this session was listed in the Business track.

If we are talking philosophically, I like the idea of pervasive computing. IT is one of those things that nobody notices when it’s working. But when it’s not working, all hell breaks loose. I personally think that technology will one day fade into the background and become wallpaper, stuff that’s all around us, but which we rarely take any notice of. And that’s a good thing. Who thinks about the gas and electricity supply in our homes? Clean running water? It’s just there and we take it for granted and that’s okay.

I think products being promoted on the strength of this or that technology (802.11 abcdefg? Can I buy a vowel?) show that we are still hung up on the whizz bang cool factors. But that will fade in time and people will just take technology for granted.

That’s what *I* think. That’s what I was hoping for when I went to this session.

After listening about all the metrics used to measure the “distance” of technology, we reached this statement: “it’s all about the value proposition”.

That’s when I took a walk. Kevin Yanks talk on CSS Frameworks was standing room only, with people bulging out the doors. I’d clearly gone to the wrong session, so I took a quick walk around Darling Harbour.

Test Your JavaScript

Rob Mitchell and Mike Williams

Maybe it was because it was the second day and all the talking was starting to get to me. But this session disappointed me a bit.

He took way too long trying to sell me on the idea that testing your code was good. I know that. Just show me the latest and greatest ways to test my JavaScript. Show me the code!

There was some code demo’s, but it was just enough to whet my appetite, not satisfy it.

Rather than repeat everything, here is the presenters list of links to available tools.

Closing Address

Dan Hill

Wow. This guy was deep. What a mind.

At first I did wonder if they had picked the wrong speaker and he might have been at the wrong conference. But he seemed interesting enough, if a little off topic. But the more I listened to him, the more impressed I was with his work. And the more I started to see why he’d been asked to speak.

This guy is looking at a seriously big picture.

I should also note that this guy provided a wonderfully rounded compliment to the opening speaker. Nicely done.