SCCharts Twelve Years Later: A Reflection on Sequential Constructiveness and(…)
summary
Claims that sequential constructiveness lets the compiler infer safe in-tick scheduling (allowing if(!x) x:=true) and that text-first modeling with KIELER + ELK removes manual diagram layout. Demonstrations include a C-based priority backend using computed-goto C macros and bit-scan-reverse dispatch, and a dataflow/hardware backend. Real projects: 10,000 SC nodes expand to ~650k LOC, compile ~2 min, reaction ~2 ms; used for railway interlocking and adopted by Blech/SCEst, ELK JS shows ~2,100 stars.
AI-written summary · about
chapters (23)
- 0:00Welcome
- 2:35Twelve years of SCCharts
- 5:03SCCharts origins
- 11:26Classical sync limits
- 13:09Sequential constructiveness
- 16:42Confluent updates
- 20:12IUR protocol
- 21:14Causality cycles
- 24:29Formal definition
- 26:51Core SCCharts constructs
- 28:59SCG: Sequential Constructive Graph
- 32:50SCCharts compilation
- 36:22Priority compilation
- 42:06C code with gotos
- 45:38C preprocessor tricks
- 46:43Industry adoption
- 48:29Text-first modeling
- 51:38Model-View-Controller paradigm in SCCharts
- 53:02Modeling pragmatics
- 55:10Diagram layout
- 58:47Future directions
- 1:00:12Closing remarks
- 1:00:57Q&A
related talks
01h 01mArchitecture Decisions: A Case Study -
32mDesign of Software Representation Languages: a Historical Perspective
26mTowards Efficient Model Comparison Using Automated Program Rewriting
40mLessons from two decades of writing bad code
40mAn Interdisciplinary View on Prototypes - Towards Executable Models That Can Be...
19mJens, the Ultimate PL Renaissance man!
34mClassic programming techniques in frontend
01h 02mFishbowl @ Functional Conf