With the release of jQuery UI 1.7 just out the door, Paul Bakaus – creator of jQuery UI has kindly agreed to answer some questions about that launch, his life as a programmer, his involvement with the jQuery UI project and the world as he knows it. Without further ado, let’s get on with the session:

Can you give me a brief history of your programming days? Where did it all start?

Although it’s a bit embarrassing, my first programming experience was trying to build a customized and skinnable mediaplayer with Visual Basic 6. This probably was around 1999 or so, and although the media player was quite nice in terms of features, it had countless timers and totally killed every CPU by just playing a simple song.

After this experiment I soon realized that I wanted to do something with the internet. One of my hobbies was (and still is) japanese anime and manga, and there wasn’t really a great community around it in germany at that time, so I took the challenge and created a community called Anime-Domain as my first big web project. It was my little playground, and after 10 revisions and designs it grew so popular that it was featured in german print magazines and had some 10k unique visitors a day. This was the time where I knew that by doing web development, you could reach many people very quickly with what you do, so I decided to continue to go that route.

How did you become the creator of jQuery UI? What’s the story behind that?

I initially came to the jQuery project while searching for a good solution to power the web applications of a big german client. As I was specifically responsible for building out a lot of the frontend logic and interaction, I quickly found out about Interface, a collection of interface plugins developed by Stefan Petre. I soon realized there was a lot of work involved to make it stable for our environment, so I invested a lot of time into bugfixes and feature stabilization, and later even planned the next generation of Interface, Interface 2 with Stefan. However, it was then that Stefan moved on with founding his own business and ran out of time, so Interface was discontinued.

In May 2006, John Resig then asked me if I would build the first version of an official jQuery interface addition, since I already had quite some experience from working on Interface. He and the jQuery community wanted to have it done in three months for the Ajax Experience conference in Boston, which was nearly impossible after a quick analysis of my workload. I had a day job, and I estimated I would need to work on it 3-4 hours everyday. After some days of consideration, I finally said ‘yes’, and for a three month period woke up every day at 6am to work for 3 hours on jQuery UI and then go to my day job. Now I can say that it was worth it

With jQuery UI 1.7 being released in the last few days, what do you see as the key parts of jQuery UI 1.7? What are you most proud of out of that release?

The one thing I’m most proud of is that our framework has matured so much since 1.5. We have taken great care to unify our API, fix hundreds of compatibility and behaviour issues, and now we truly have a foundation to build upon without needing to rewrite the core functionality again and again. This will allow us to push our features in the next releases in literally no time.

What were the biggest challenges of getting the 1.7 release out there?

Our test coverage is still incomplete, and with every fixed bug, we introduced 2 others, which made the arrival at a stable level extremely difficult. Every week, there were some 50-100 bugs entered in the bugtracker, and there was literally no end. Luckily, we were able to triage the bugs to criticals and blockers and solve these in time for a release. You have to have a lot of guts to push out a release that’s still imperfect – but an imperfect release is better than one that gets delayed for months. You can always roll out 1.7.1.

Jumping off of jQuery UI for a second, Do you “release early, release often” with all of your projects? What are your thoughts on that strategy?

Speaking for myself, I often release too early. That has been a problem in the past, when we released versions as stable that weren’t, for instance. So it’s important to find a combination of both – a stable release must be stable, while development cannot be halted or blocked through stubborn processes. A labs section is great in that way – it allows developers to contribute freely and plan on an open canvas, with early preview releases, and the work can later be merged back.

Have any of your startup projects failed dismally – if so, why and how did you learn from them?

A lot of my side projects failed, as a matter of fact, while others succeded. Usually, the reason why some project fails is not a technical one, but the fact that one didn’t build meta data around it. That means a dedicated site, documentation, instructions.

One of the examples is my recent Transformie plugin ( http://paulbakaus.com/2008/08/16/bringing-css-transform-to-internet-explorer/ ). While I believe it gives you great technological advantage, meaning to be able to use CSS Transforms in Internet Explorer, Firefox 3.1 and Webkit/Safari, nobody adapted it yet. I believe the reason is that I simply forgot to also work on documentation and a dedicated site with a download, which could have helped a lot during the promotion.

Mac, Windows or Linux? Why do you love this platform?

I switched to Mac hardware around a year ago and I’m totally in love with it. All components work together nicely, and so far, I never had to return my Macbook Pro to the Apple Store because of an issue. However, I’m still using Windows through Parallels because OSX, while visually nice and stable, has fundamental usability flaws.

One of these flaws is the Finder. I recently worked on the jQuery UI Selectables in the labs version, and once again saw that the Finder had great flaws when it comes down to selection. For instance, if you select multiple items and click on one of them, the multiple selection isn’t cleared. Also, my tools that I love for windows simply don’t have an alternative yet (see below).

What are your tools of choice to get the job done?

For editing files, I love the e texteditor (yep, the name is “e”). It basically started as a textmate clone for windows, but since then grew into something much greater. One of the features I can’t live without now is there great multiple selection support. Hold down CTRL, select a couple words through double clicking and then type over all of them. Is that cool or what? On a sidenote, on one of my talks in Japan the audience asked why I’m so insane to use a Textmate clone through an emulated Windows on OSX – I showed them this exact feature, and they were all amazed.

Other than that, I like TortoiseSVN, even if it slows down the Explorer, I love Photoshop, especially the new version with hardware acceleration, Trillian Astra for instant messaging, Gmail for email, Firefox 3 for browsing, VideoLAN for playing video, Keynote for preparing presentations, and WinSCP/Putty to do server administration. Phew, I guess that’s about it

Back onto jQuery UI… Can you see jQuery UI making more of an impression in the future, lead by the current successes of jQuery?

This question is difficult to answer because jQuery, other than jQuery UI, can be useful almost everytime. jQuery UI gives you a specific set of user interface widgets and behaviours, and many people think of it as loosing a kind of freedom. On the other hand, there’s definitely some connection – if you’re using Prototype already, and you’re looking for an UI framework, your choice is most often script.aculo.us. If you’re using jQuery, why not use the official side project?

What’s next on the hitlist with jQuery UI?

There’s a lot of movement right now. We finally pushed out 1.7 3 days ago, which we believe is a solid foundation for everything that’s coming in in the next couple of months. While 1.7 was a stability and foundation release, the next releases will concentrate around features, so expect to see many more components soon. Some examples are the colorpicker, menu, grid, tooltip and tree widgets.

Additionally, my personal goal is to target more platforms, for example the iPhone. Early test implementations I did show that it’s fairly doable to support the touch events, and therefore make all jQuery UI interaction compatible with mobile devices.

Additionally, I’m working on a brand new lab section to be able to push feature development without any restrictions. This allows us to work on anything we find is cool, but maybe not on the roadmap. Expect to see a lot soon in the jQuery UI Blog.

Is it too early to discuss jQuery UI 1.8?

There’s a lot on our list for the year. Filament Group did a great job to start a list of all widgets they could imagine being part of jQuery UI, which can be found and discussed in the jQuery UI planning wiki at http://wiki.jqueryui.com. There’s no definite roadmap for 1.8 yet, but some components are likely to make it into the next release. For instance, widgets that are already being worked on (grid, menu), as well as widgets that we had to kick out of 1.6/1.7 because they weren’t stable enough (spinner, colorpicker, autocomplete). It will be a huge feature release!

And finally, If you could give one tip to any new budding jQuery UI developers, what would it be?

Think different. No, seriously. For jQuery UI, we’re trying to take the same path than jQuery, and people that are restricted to thinking in classical OOP patterns will have a problem. If you think about it freely and give the functional, event-driven and progressive approach a try, you’ll find yourself getting the work done with jQuery UI in a fraction of the time needed than with other frameworks.

Paul can be found on Twitter as @pbakaus, or via his blog at http://paulbakaus.com/

Tools to help you learn…

Learning jQuery 1.3	Amazon Kindle 2	jQuery UI 1.6

In the first of I hope many articles, today I’ve been talking to Jörn Zaefferer of jQuery fame. Jörn is one the main contributors to the jQuery library and is based in Cologne, Germany – the rest you can hopefully read below. Catch Jörn on Twitter as @bassistance or http://bassistance.de.

1) First up, can you give us a brief history of your programming days? Where did it all start and how did you get from there to developing/contributing to jQuery?

My first programming involved typing of little programs from a book into my father’s Schneider computer, in Basic on a green screen. I also wrote a little program that outputted random symbols along with a beep of a varying frequency…

Time-warp forward about ten years, next thing was a Visual Basic program for managing Windows desktop themes, with a set of Diablo (the Blizzard game) based themes: wallpapers, icons, mouse cursors – I used some helper code from a friend to update the necessary registry entries. Lost it all during some reinstall.

In 2004 I actually started learning to program, starting with C for a semester, then a lot of Java and some PHP, SQL. During a two month project I developed a Eclipse RCP based application for a motion-sensor glove: Data was read from a USB device, filtered based on some calibration input, then outputted as graphs and as MIDI to control music synthesizer software. We presented that later at school and got a little award for it.

After that, early 2006, I started as a Java programmer in Cologne, programming JSR-168 portlets for the IBM Websphere Portal platform. Around that time I also started working with jQuery, using it to implement some JavaScript helpers in some admin portlet, ala a “Check All” checkbox.

Somewhere around summer 2006 I started contributing tests to jQuery. I also worked on the testrunner itself, which later got dubbed QUnit, and worked for about half a year a lot on jQuery’s ajax module. During that time I also started various plugins, with the validation plugin being the only one I started from scratch – everything else were ports of existing plugins, where the original authors usually didn’t continue working on them.

Today I’m maintaining QUnit, the testing framework, sort-of consulting on jQuery core (mostly involved in roadmap planning and some other discussions, not so much contributing any code), maintaining my plugins and also working on jQuery UI.

2) What are the biggest challenges that jQuery development faces in the future?

Trying to appeal to an even bigger audience. jQuery today is extremely appealing to designers, as there are basically no sane alternatives for them – or are there any other [framework]fordesigners.com sites I’m not aware of? But there are plenty of alternatives for “complex applications”. The challenge here is mostly to make “complex application” less ambiguous – learning what those developers actually need when they use that term, and why they think jQuery doesn’t fit in these cases.

3) What jQuery implementations have impressed you from an application perspective?

The great thing about jQuery is that you can use it to fill the usability-gaps, making things a little better. The side-effect is that there isn’t any obvious “jQuery application” – most of the time, you won’t notice jQuery in action at all. A recent exception of that maybe quakelive.com, which doesn’t work without JavaScript at all, and therefore makes very heavy use of it, all based on jQuery. In this case its fine, as you need a system capable of rendering the Quake-browser-plugin anyway.

4) What is your favourite jQuery plugin and why?

I just implemented FancyBox as a lightbox implementation on a project – works well by default and is really beautiful.

The jQuery validation plugin is the one I recommend most of the time: It’s applicable for nearly every app, its well engineered and tested, already used by a lot of interesting sites, like newsweek.com or soundcloud.com.

5) What web/client tools do you use to manage your development lifecycles?

Firebug.

Also Eclipse with Aptana, Maven, and Subversive, XPlanner for Issue Tracking and Project Planning, Hudson as Integration Server, Maven for build and dependency management.

6) Have you been impressed by the recent releases of new browsers: Chrome, Safari 4, IE8, FF? Where do you see that world going?

New browsers are not that interesting when they are unable to push old browsers off the table. Safari 3 got mostly rid of Safari 1 and 2, which is great. But IE6 is still here to stay for another one or two years at least.

7) How difficult do you find having to support the older browsers, specifically IE 6? Would you take the decision to not support IE6 like some have recently (Instapaper for example)?

Not supporting IE6 very much depends on the type of application. If you decide to not support it anymore, assume that someone coming to your site with IE6 has no idea there even are other browsers or newer versions, and provide the necessary help for upgrading. Otherwise you are just letting people run against a wall – I can’t even convince my mother to use Firefox 3 instead of IE6.

8 ) What emerging web technologies are exciting you?

Bespin has some very interesting ideas, I especially like the idea of contributing to an open-source project without any effort on getting the project to checkout and compile on my local machine, instead just opening the editor on their site, testing my change on their integration server, and committing the result as a patch to a branch or ticket system.

9) Has the recent success of jQuery surprised you, or has it been a steady rise to fame?

After the release of jQuery 1.1, the jQuery core was pretty much done from my point of view. John (Resig – jQuery Author) managed to come up with all kinds of ideas for two great releases (1.2, 1.3). The 1.4 roadmap contains no real surprises, instead a bunch of solid improvements. On the other hand, the roadmap for jQuery UI contains over 50 components, thats a lot to tackle.

Overall, jQuery’s success seems to be result of very steady work from the team: producing code, and building a community by helping and evangelizing.

10) If you could give one tip to new jQuery developers, what would it be?

Even with jQuery making a lot of tasks trivial, get to learn some JavaScript basics. jQuery wraps the DOM API really well, but you can’t avoid working with the JavaScript language. I’m happy to help with the validation plugin, but I don’t want to repeat explaining why rules: { first-name: “required” } throws a syntax error (you have to quote “first-name”).

Tools to help you learn…

Learning jQuery 1.3	Amazon Kindle 2	jQuery UI 1.6