This paper is a discussion around the issues faced by distributed systems when dealing with time, ordering and failures. This is particularly relevant with the work I am doing at the moment designing a very large distributed system that will need to scale massively.
It is remarkably readable and illustrates key issues and limitations in the field. It has already got me researching other areas mentioned in the paper.
Links
It is a classic way people start a conversation. Some people meet, they exchange a few pleasantries and then they ask “and, so what do you do?” And each person then takes turns, hopefully if someone doesn’t just keep talking, explaining what their job is.
This was an excellent podcast on the ethics behind some of the decisions being made about technology today. The whole talk is excellent and thought provoking however for me one part (about 27 minutes in) stood out:
A lot of coding jobs can involve fairly “normal” tasks - create a page, write to and read from a database, add a new field to this screen. These tasks have been done numerous times by numerous developers around the world in different flavours.