News

Soundbyte 72 – Vive la rebellion

2 december 2012

Last year Bert and I were discussing how to ever top last year’s conference tour with talks at major Java conferences such as Devoxx and JavaOne. Well, with sixteen talks at eight conferences in a year. That’s how. And that’s only my talks, so the actual number of Luminis talks is even higher. One of the last conferences I spoke was Øredev in Sweden. The theme of the conference was “Vive la rebellion”; Software is changing the world and we are responsible for it. Innovation is not just happening to us, but we are making it happen. One of the keynote speakers, David Rowan from WIRED magazine, named education as one of the main areas where software will make revolution happen in the near future. This obviously fits our ideas with PulseOn (the new brand name for Leren-op-Maat) perfectly, which only strengthens the feeling that we’re on the right track. This also shows in numbers. While this year is still a pilot phase for PulseOn, we’re getting massive attention from all directions. For example there was a workshop last week where over 150 people had a hands-on introduction with PulseOn.

On the technology side it’s always interesting to be at conferences to just get an idea what technology might be interesting to take a closer look at. Modularity is definitely one of the big themes. Besides our own talks there were several other modularity related talks, and all of them had full rooms. Pushing our modularity vision with Amdatu seems to be right on time. Besides a talk, Bert, Marcel and myself did a hands-on lab with OSGi and Amdatu at Devoxx. The room was packed, and people were surprised how easy it is to work with.
Another big theme is the browser. Under the banner of HTML5 a lot of things are happening really fast. One of the things that this means for us is that we have to take client-side development a lot more serious. A considerable amount of our code base is moving to JavaScript when using client-side user interfaces. So far this has been the domain of interaction designers/developers mostly, but I think this is no longer enough. When working with a large JavaScript code base we will need similar software constructs for this code as with other parts of our code base. If not, you will end up with an unmaintainable mess. Automated testing, design patterns, code standards, we will need it all when moving more and more client-side. One of the interesting things happening right now is the rise of MVC JavaScript frameworks. The one that I find most interesting is Angular.js from Google. Angular has a focus on declarative templating that offers a much cleaner separation between the view and model/controller. We recently decided to standardise on Angular in PulseOn as well. Easily integrating new technology in a code base is another benefit of modularity by the way. In a modular code base it’s easy to start using new technology (Angular in this case) without having to re-write the whole code base. Another UI technology that is very interesting is Dart. Dart is a new programming language that very much “feels” like Java. It is designed to run in the browser however. With Google Dartium (a next-gen build of Chrome) Dart runs natively in the browser. For other browsers it compiles to JavaScript. Dart is still under heavy development and it’s a bit early to start using it in production, but it looks very promising. It might give us a platform to develop in the browser with a language that is more fit for large scale development, for example because of the more powerful tooling possibilities.

Another technology trend is the rise of new stacks for web development such as Node.js and Vert.x. My take on this is that these frameworks are the new PHP. It’s easy to get started with, and easy to do most web development tasks. Easier than with Java you might say, all tough I’m not even so sure about that. The real problems that make software development difficult are not addressed however. What about modularity for example? How do you use this technology at large development scale? This kind of innovation is very useful and we should experiment with it and integrate ideas in our own stack, but I don’t believe any of these technologies is the next big thing.

A more interesting trend is functional languages. At Øredev I had a very interesting conversation with Ted Ewald, one of the core developers of Datomic and a big name in the Clojure world. He has been doing development in pretty much every big technology stack, including Java and .NET and believes that we have reached a point where better frameworks and tooling are not going to help us much any more. Instead we need a paradigm shift, and functional languages could be that shift. I don’t fully agree with him (yet), but there is certainly a point in what he is saying. I do believe that there are a lot of problems that are more efficiently solved using languages such as Clojure however, and we should teach ourselves to do so and integrate them in the stack that we use.

All the videos from Øredev are available online, and all Devoxx videos will be on Parleys in the next few weeks.

The Luminis Technologies team is growing as well. Tran and Maurice joined us recently and this Monday Jago de Vreede will start working with us. We have some more news in this area in the next few months as well… In the Technologies team we apply the concept of apprentices by assigning a (more experienced) coach for each person. This is not something just for the first few months, but an ongoing process to make sure the new guys get up to speed quickly, but also to make sure we keep improving. The reason I start talking about this is to introduce the video for this week’s Soundbyte. While browsing for some music I found this video of BB King and John Mayer. I’m a big fan of BB and it’s fun to see them together on stage. Although John Mayer is one of the big names right now, it must be a great experience to play together with one of the truly great.


Last year Bert and I were discussing how to ever top last year’s conference tour with talks at major Java conferences such as Devoxx and JavaOne. Well, with sixteen talks at eight conferences in a year. That’s how. And that’s only my talks, so the actual number of Luminis talks is even higher. One of the last conferences I spoke was Øredev in Sweden. The theme of the conference was “Vive la rebellion”; Software is changing the world and we are responsible for it. Innovation is not just happening to us, but we are making it happen. One of the keynote speakers, David Rowan from WIRED magazine, named education as one of the main areas where software will make revolution happen in the near future. This obviously fits our ideas with PulseOn (the new brand name for Leren-op-Maat) perfectly, which only strengthens the feeling that we’re on the right track. This also shows in numbers. While this year is still a pilot phase for PulseOn, we’re getting massive attention from all directions. For example there was a workshop last week where over 150 people had a hands-on introduction with PulseOn.

On the technology side it’s always interesting to be at conferences to just get an idea what technology might be interesting to take a closer look at. Modularity is definitely one of the big themes. Besides our own talks there were several other modularity related talks, and all of them had full rooms. Pushing our modularity vision with Amdatu seems to be right on time. Besides a talk, Bert, Marcel and myself did a hands-on lab with OSGi and Amdatu at Devoxx. The room was packed, and people were surprised how easy it is to work with.
Another big theme is the browser. Under the banner of HTML5 a lot of things are happening really fast. One of the things that this means for us is that we have to take client-side development a lot more serious. A considerable amount of our code base is moving to JavaScript when using client-side user interfaces. So far this has been the domain of interaction designers/developers mostly, but I think this is no longer enough. When working with a large JavaScript code base we will need similar software constructs for this code as with other parts of our code base. If not, you will end up with an unmaintainable mess. Automated testing, design patterns, code standards, we will need it all when moving more and more client-side. One of the interesting things happening right now is the rise of MVC JavaScript frameworks. The one that I find most interesting is Angular.js from Google. Angular has a focus on declarative templating that offers a much cleaner separation between the view and model/controller. We recently decided to standardise on Angular in PulseOn as well. Easily integrating new technology in a code base is another benefit of modularity by the way. In a modular code base it’s easy to start using new technology (Angular in this case) without having to re-write the whole code base. Another UI technology that is very interesting is Dart. Dart is a new programming language that very much “feels” like Java. It is designed to run in the browser however. With Google Dartium (a next-gen build of Chrome) Dart runs natively in the browser. For other browsers it compiles to JavaScript. Dart is still under heavy development and it’s a bit early to start using it in production, but it looks very promising. It might give us a platform to develop in the browser with a language that is more fit for large scale development, for example because of the more powerful tooling possibilities.

Another technology trend is the rise of new stacks for web development such as Node.js and Vert.x. My take on this is that these frameworks are the new PHP. It’s easy to get started with, and easy to do most web development tasks. Easier than with Java you might say, all tough I’m not even so sure about that. The real problems that make software development difficult are not addressed however. What about modularity for example? How do you use this technology at large development scale? This kind of innovation is very useful and we should experiment with it and integrate ideas in our own stack, but I don’t believe any of these technologies is the next big thing.

A more interesting trend is functional languages. At Øredev I had a very interesting conversation with Ted Ewald, one of the core developers of Datomic and a big name in the Clojure world. He has been doing development in pretty much every big technology stack, including Java and .NET and believes that we have reached a point where better frameworks and tooling are not going to help us much any more. Instead we need a paradigm shift, and functional languages could be that shift. I don’t fully agree with him (yet), but there is certainly a point in what he is saying. I do believe that there are a lot of problems that are more efficiently solved using languages such as Clojure however, and we should teach ourselves to do so and integrate them in the stack that we use.

All the videos from Øredev are available online, and all Devoxx videos will be on Parleys in the next few weeks.

The Luminis Technologies team is growing as well. Tran and Maurice joined us recently and this Monday Jago de Vreede will start working with us. We have some more news in this area in the next few months as well… In the Technologies team we apply the concept of apprentices by assigning a (more experienced) coach for each person. This is not something just for the first few months, but an ongoing process to make sure the new guys get up to speed quickly, but also to make sure we keep improving. The reason I start talking about this is to introduce the video for this week’s Soundbyte. While browsing for some music I found this video of BB King and John Mayer. I’m a big fan of BB and it’s fun to see them together on stage. Although John Mayer is one of the big names right now, it must be a great experience to play together with one of the truly great.


Geef een reactie

Het e-mailadres wordt niet gepubliceerd. Vereiste velden zijn gemarkeerd met *