My first post on this site will be about the Fronteers 2009. You could view it as a symbol what this weblog and my doing is all about: the Web. First I wanted to get into much more detail about each talk, but it happend so much on this day that this will only be a small roundup what happened on the first day of the Fronteers and I try to provide you with links where you can get more information..

The first session was held by Molly E. Holzschlag about the

h3. “State of the Moblie Web”

Molly opened the Fronteers with a talk about what goes on right now and tried to awaken the people in the room with a lot of questions and polls where people can lift their hand so they don’t fall asleep. In one of the polls it seemed that two third of the listeners use the mobile web. She stated that most of the time we are working around issues instead of creating great software.

The mobile web has more constraints then the whole web world, like a more limited typo set, fewer colors, fewer memory in the devied just as fewer bandwidth and battery power. But there are also some advantages like location-based services. This talk was a nice warm up for the day

Then Peter-Paul Koch a.k.a. PPK gave a talk about

“The Mobile Web”

PPK started his talk really at the beginning. Back in 1588 when the spanish armada tried to conquer England. PPK pointed out that they didn’t know how to do a modern naval battle and that the same applies to the mobile development today. Then after giving a good overview about what browsers on mobile phones are out there, he commented about the mobile browsers situation. He said that the browser for Blackberry is dead, mozilla has much to catch up, because they are really late to the game, NetFront just isn’t good and IE is just IE (IE6 to be specific).

He also talked about W3C Widgets as the solution for making it possible to develop applications on mobile devices with tools we already know: HTML, CSS and JavaScript. The only thing that is to learn is the config.xml and that can’t be that hard. Weak spot of the widgets approach is, that there is no device API available right now. So it is not possible to use device specific functions like filesystem or geo locations and implementing that is a hard problem.

Then Cristian Heilmann talked

“Of Hamsters, Feature Creatures and Missed Opportunities”

In his talk he first analysed human behaviours of gathering knowledge and stuff and that we solve the same porblem over and over again. Then he revealed what this talk should really be about: YQL. The Yahoo! Query Language is basically a wrapper for a lot of APIs and data gathering stuff like reading out some content of pages you want. You create a SQL-Statement where you can specify what you want Yahoo to get for you and in which form it should be retrieved. It is there so we can focus on remixing stuff to create great new things, because “the web is about Data”.

The slides are also available on his website.

Lunch

That wasn’t the best I have eaten so far, but it did what it was supposed to do: Fill my stomach.

After that Lunch Martin Savelkoul presented a case study at which he showed how it is done:

Building and maintaining a complex frontend codebase

He recommended to split all code into components and even create a component library for code that is used across the boundary of a project. Also he stated that we “want strict seperation of frontend development and the optimization of code release”. With what he means that every code release — for example testing or implementation — needs its own optimizations and that we should use the right buildscripts, documentations and compressions for each release.

One of my favorite talks on the this Fronteers and definitely one of the most inspiring ones (on this day) was Stephan Hay talking about

The Future of Layout

This talk was about CSS and how different layout models can support creating a website using real grids. It is how designers think. He presented in short the three grid layout models out there: CSS3-Grid from Microsoft, Mozilla’s CSS3-Flexbox and the CSS3-Layout by the W3C. Each of them having its own advantages and disadavantages. So making all of them work in all browsers so we can use that one, that is best for the current problem should be the best approach.

In short he also said something about some other CSS3 features. Especially the media queries have great potential in shaping the possibilities of the future (mobile) web.

The slides of this talk are also on his website.

Then it was John Resig’s turn:

Javascript Testing

He stated that testing Javascript is “not the same as Desktop or Server-Side testing”. You have to test not only on different systems, you even have to test in different browsers and browser versions. In a poll he found out that the most used unit testing frameworks for Javascript are (in descending order) QUnit, JSUnit and YUITest.

Besides of talking short about functional testing, server-side testing and distributed testing, he also mentioned a category he called manual testing, what in my opinion is a genious idea (if it would work). Distributing testing instructions to volunteering users who have to test a system and say if it does what was expected. I guess you can say it is some kind of asynchronous remote usability testing.

Of course he get into more detail about the testing frameworks, and especially about his QUnit and TestSwarm. For interesting stuff to this topic are Selenium

The Slides are available on slideshare.

The last session of this day was held by Steve Souders:

Fast by Default

Steve made much advertising for his book. He stated that “fast companies are successful”, so we have much reason to make our sites perform even better. He explained some techniques about how to create a fast website such as progressive enhancement, progressive rendering and how to load scripts that do not block the loading of the content. He also suggests using YSlow and PageSpeed to get a clue what could be improved in loading your website.

Tools you probably should look at for getting “Fast by Default” are the Aptimize Website Accelerator, SproutCore and some feature release of Rails. They will do some of the optimizing Steve speaks and writes about by default so it isn’t hard to get them.

Fotos

An awesome photostream by marvos is available on flickr.

Elsewhere

While many great people are gathered in the Felix Meritis to talk about frontend stuff the google team released their Closure library, including a Compiler that optimizes the code and makes it smaller, a library of various Widgets, utility functions, testing functions, etc. and a templating engine for Javascript that also works in Java.

I will add links to slide, fotos and media about this event to this post when they are available.