I am living and working in Zurich. Visit my side-project rss-proxy.
Announcing RSS Proxy 1.0 - Automated Feed Generator
tl;dr: rss-proxy transforms almost any website into a feed, just with one click. RSS is the glue code for an open and free internet. It helps you build your own network of sources. Unfortunately some websites don’t support it at all or not in a way to fill your need. This is where rss-proxy comes to play. It is a simple middleware that transforms HTML to a feed. The user will choose a suggested feed from a collection of feeds that exist in a website of interest....
End to End Testing - Before You Get Started
A lot of articles have been written on the topic end-to-end testing. I won’t cover the basics here. In this post I want to draw your attention to some questions, you should answer before you start even looking for a framework. And.. start slow! Here is my story: In my last job I inherited the task to write end-to-end tests, so it was somehow easy to get started. We were using protractor, angular was still hot in these days....
Kontor CLI (2017 - Present)
If you are like me and try to keep looking for interesting ideas and concepts, you will at some point reach a point of information and cognitive overload. The internet is full of distractions and its difficult to keep the focus for some time, even harder to stay focused on a topic for a longer period like weeks. I was looking for an automated way to keep following interesting people/sites archive them and while avoiding all the distractions Hence, I am implementing a little cli tool that handles web sources in a general way, which in most cases are pure RSS feeds, in the complex cases JSON or HTML corpora, that I have to map to a proper format....
Automate the UI-test Automation
Test automation especially ui test automation is like a huge jenga tower, everything is always about to collapse. How can we rely on them? Only run stable tests (stability) Keep the execution time low (performance) Only run the important tests for first feedback (performance) Loosly coupled with the code under test (maintenance) A test should have a maintainer that is required to fix it (maintenance) A failing test should express what functionality has been broken, ideally in a way even non-tech people can understand (maintenance) ...
Principles of Luhmann's Zettelkasten
Write (small) notes, using note sequences to develop an idea or topic Where possible or necessary (i.e., if an idea can be considered as “start” of a new note sequence, but does fit into the current linear argumentation), branch off a new idea-path Check if similar topic or “story line” has been written down elsewhere; if yes, add manual links / references Identify relevant notes that function as “starting” point and tag these notes (from Daniel Lüdecke [2] ) https://vimeo....
How to learn efficiently
State-dependent memory Train your brain to get into the mental state of studying: sit at your study area turn on your study lamp after 25 min (the typical time you can learn) get up and rest for 5 min avoid distractions (avoid the living room, kitchen) The more active you are in your learning, the effective Concepts are more important than facts Study groups are great Rote learning Pomodoro technique Source: https://www....
When the mind follows the body
Trick your mind into different mental states by changing your behavior first. Fake it till you make it. Used in cognitive behavioral therapy. https://vimeo.com/68572000...
Perceptual Diff Testing
One goal of successful testing is decoupling the tests from the code under test, with the goal to reduce the need for maintenance of the test code, when changing the actual code. Perceptual Differences (pdiff) is one approach to achieve that goal. Enriched with protractor, its possible to compare screenshots within a flow. "...
Biases in Machine Learning
ML is often considered to be fair and objective. Unfortunately the opposite is true. https://media.ccc.de/v/33c3-8026-a_story_of_discrimination_and_unfairness...
Developer Dashboards
( I already had some thoughts on that topic ). Goal Get a better overview of future/pending/past work and keep it. “If you want a developer to do something, e.g. establish a high test coverage, make it easy with sane reasons” What? Pre-Deployment Feature Branches, consisting of PRs Build (Test) Reports Static Code Analysis (Report?) Build Status green/yellow/red Human Code Analysis (Code Reviews) Post-Deployment Monitoring Active (Alerts) Passive Configuration Feature Flags?...