Over the years I've spent a lot of time supporting applications. I would like to think I'm pretty effective at it. So I thought it was the right time to share my approaches. The key is a systematic, logical approach to identify the cause - or causes.
Over the years I've done a lot of development on a variety of platforms. Error management is something developers either bake in from the start, add in later, or never get round to! It seems a good time to review my experiences and my philosophy.
People discussing Domino application development have been using a new term since early this year - "mid code". This has become necessary because of the evolution of Domino development since Domino V10.
Recently I've been diving back into running Java outside of the Domino HTTP stack, picking up some work I did quite a few years ago playing with Vert.x and Domino based on Stephan Wissel's blog series on Vert.x and Domino. Quite a few things have happened since I was last working on the project, not least the laptop I had at the time got rebuilt, I have got a new laptop, several version of Eclipse have been released and XPages SDK has been deployed to the Eclipse Marketplace (thanks Jesse Gallagher).
One of the themes which crops up from time to time in Domino application development and beyond is the theme of "performance". It's a topic which makes me grit my teeth because of the basic premise. Most articles start from what, for me, is a narrow terms of reference: performance optimisation being about how quick a specific amount of data can be served to a specific environment. My interpretation of the term "performance" is much wider and that is the reason for my scepticism and concern of how people may interpret or use such articles. There are important points raised, but there are also caveats that need to be borne in mind.