This week is the Dagstuhl Event Processing “seminar” (a.k.a. workshop) in the pretty Saarland region of Germany, attended by vendors, researchers and Gartner analyst Roy Schulte. Dagstuhl is a kind of non-profit research collaboration centre that runs “seminars” on various computer science topics throughout the year – its like a kind of academic retreat, with principles like “no one can lock their bedroom door” when they leave their room (which must be a mainland European concept…).
The Dagstuhl campus is actually in an old provincial palace (think château) with modern extensions, under a hill with a ruined 13th century castle (populated by giant snails, and apparently destroyed by Napoleonic invaders – facts which may or may not be related).
The group here (40+ folks) is being managed by Opher Etzion, and we are arranged into 5 groups covering the “why, “what”, “relative IT context”, possible standards, and Grand Challenges in Event Processing. Various numbers signed up for these groups, with varying from 3 of us on “standards” to 10+ looking at the “why” and “what” groups.
Today the “Why use Event Processing” group presented their early thoughts – based on looking at benefits (like latency of response) versus costs (like acquiring additional tools, skills etc). Probably they will need to look at real comparative customer use cases to get actual data on this stuff… but they’ve made a good start.
