Quick introduction of the conference:
- 250/300 people
- In some big hotel in Verona
- A lot of locals, not a lot of English speakers around, but all the talks were in English.
- 50/50 between international and Italian speakers.
- The wifi was terrible.
- They were mentioning the level of the talk (beginner, intermediate, advanced). I think it is pretty cool to have that because you can have an idea of what to expect.
Here is some details about some of the talks I’ve attended. Be careful, everything is based on sh** notes, on some random sheet of paper, with a random licking pen, so not necessarly extra-accurate.
He mentioned several times how Haskell has also an influence for him, and his implementation choice. At the end, he showed another example which was creating a UI (small progress bar), mixing Metascript and React.js. It seemed to be really clean. Worth checking if it gets bigger, as it seems to have a really small community for now.
The tone was set with this technical keynote, maybe too technical for a keynote :)
####NSA.js - Danni Friedland - Slides####
####NoFlo.js - Henri Bergius - Website####
The speaker gave a short introduction about flow-base programming, which has its origin in the 70’s. He explained how each component of an application is independent, and acts as a black box for the rest of the system. It is just passing messages/streams around. With Noflo, your application is just a set of boxes, that you connect to create a graph. The UI was pretty nice, with a zoomable interface ala Prezi, showing you more details about the different components of your system. The talk was a bit too short to really understand how it was working. Nice to see a Kickstarter campaign ending in a real product.
####The spirit of testing. - Marco Cedaro - Slides####
The talk was technical, the demos were working well, the speaker explained really well some concepts, with relevant examples.
Extra-point for this speaker also as he was wearing a bow-tie. Jakob tried to raise awareness and inspire about finding a solution for automatizing CSS testing. He said that CSS was nice, but we were usually adding CSS, never removing it because we don’t know what it could break. There is no overview on how the different styles in your application interact. He talked about code quality and TDD, and applied it to CSS and design. TDD relies on having a failing test, implementing, checking that the test passes and refactor. For implementation of a design, it should be getting a design, implementing it, having it right and iterate, but for this you need to be able to test correctly the visual appearance of your implementation. He mentioned four approaches, that we can resume as:
- checking validity of CSS (csslint)
- styleguide and code review
- screenshots (Huxley)
- comparing style in the dom (Hardy)
The solution would be a mix of those 4 points. Nobody did it yet. So my conclusion is: it is still a myth.
- Reactive programming is trendy. It resolves some important issues also. Rx.js is mentioned everywhere, and seems to solve long term issues in a nice and effective way.
- I was surprised by the nsa.js talk. I think it was really nice and impressive. I can see some real possibities of using this for automatic acceptance tests.
- I missed some talks about managing big codebase. Nobody talked about that, except maybe the browserify + npm talk where he mentioned a way of managing your codebase by features.