Trifork Blog

Axon Framework, DDD, Microservices

My impressions at GOTO Amsterdam 2013

July 11th, 2013 by
|

A few weeks a go I attended the GOTO conference in Amsterdam. One of the reasons I like to go to conferences is to meet people I haven’t seen for a while, talk about technology and other stuff. The other reason to go to conferences is to get inspired. After the conference I like to write down what I got inspired by. But it is always hard to capture the impressions I got in a blog entry…

goto_web_header_AMS_2013_960x175

For this conference I could have written about Linda Rising and her keynote talking about incentives. Why should you use them, or why not. Than I would have talked about managers that still think you are doing nothing as a programmer when you are not working on your computer. I probably would have told about the effect of giving rewards to people. That it is usually counter productive when people get rewards. People can think they need to be given a reward because they need to do something that is not fun to do. I probably also would have told you the story about a company that paid new hires 3000 dollar to leave after their initial 2 week training program.

If you do want to know what I will write about, read on.

Another thing I could have blogged about is how Jeroen Reijn created a recommendation engine for Hippo using Couchbase and Elasticsearch. Than I would have told you that Hippo has created a very nice targeting engine for visitors of websites created with Hippo. Than I would have explained that you do not want to store this kind of information in JackRabbit. That you need a much faster solution. Therefore at hippo they started looking for a distributed cache. That they preferred solutions that were backed by a large company. That they chose Couchbase as a drop in replacement for memcached. That the query options provided by Couchbase were not sufficient and therefore they also used Elasticsearch, to make the search run faster.

Another thing I could blog about was the very inspiring talk by Erik Meijer. That guy has so much energy. He has the power to give a nice keynote at the end of the day. He talked about how bad the relational model is. I probably would have talked about some of his remarks like:
Screen Shot 2013-06-12 at 10.52.28 AMA relational database needs to know everything about the world to function well.
The world does not want to be normalized, it wants to be normal.

I would have told you more about how he explained why foreign key constraints are hard in the real world and also why it does not hurt to use a little bit of disk space to store certain values twice.

Another interesting topic to blog about is Schemaless data storage, which was presented by Martin Fowler. In that case I would have blogged about implicit schema’s. Even if you use a HashMap to load and store you data you can still wrap it in an implicit schema. I would have talked about a predicate system to check data before sending it to storage. I would have explained that a predicate system is like the xsd for xml documents. I would have made a joke that I did not understand what predicate means, but that I did understand the xml/xsd sample that he gave. In the end I would have discussed that having this mechanism enables you to support contextual validation. No validate the complete model or validate nothing. You determine what you validate by context. This is something that is very hard with a data store that has very strict rules on valid data.

Instead of the soft side of software development I could also write a blog about some tools they use at Github to be effective programmers. In that blog I would have talked about creating and hosting a website on the Github servers using Jekyll. I could also talk about creating personal pages on github, or even project description pages using the special gh pages branch. Because I am a mac fan I would also have discussed the tool that is only supported on the mac for now called Boxen. I would have explained how this tool can be used to prepare you mac in a few minutes to start working on a project.

The final thing I could have blogged about is how the architect and his architecture fit in our way of doing projects. Both Simon Brown and Dave Thomas discussed this subject. In this blog I would have discussed some of the remarks these guys made around architecture like.
Screen Shot 2013-06-12 at 10.52.28 AMSoftware architecture allows you to scale teams
Software architects should be coaches and mentors, they a duty to educate.
Big architecture is dumb, no architecture is dumber.

In the end

I am still not sure what I should have blogged about. I know I had great fun, attended a number of good presentations and met interesting people and nice people from the past. I already generated my personal page on github. I started thinking about what Linda explained related to giving rewards. I also have an idea for a new presentation based on the contest during one of the presentations where a colleague won a FirefoxOS phone. So yes, all is all I did get inspired, will be going next year and hope to see you there too …

Comments are closed.