A collection of things that are Turing-complete by accident (2013)

Missing from the list is Dwarf Fortress: http://dwarffortresswiki.org/index.php/DF2014:Computing Great when you want to try and create a computer while trying to fend off goblin attacks and zombies. Technically, it's not Turing Complete, as the "tape" is not unlimited, but other Turing Complete games have the same issue.

Also x86 mov instruction is Turing complete! [1] There is a guy who run Doom with only mov instructions, but it is of course incredibly slow, one frame every 7 hours :) [2] [1]: https://news.ycombinator.com/item?id=9751312 [2]: https://news.ycombinator.com/item?id=16218872

"Mean time to turing completeness" is a mental metric that I use to decide whether making a domain-specific language is warranted or it would be better to just use a general purpose language. If you can really limit the scope of the task to something non-turing complete, then a DSL might be a good idea. If you don't think you can keep the thing from becoming turing complete, you're going to fall prey to Greenspun's Tenth Rule [1] and you…

Too many things are Turing Complete. I would like to propose more research into things which are "Tetris Complete". https://everything2.com/title/Tetris+complete Tetris Completion extends Turing Completion by adding the minimum requirements for a functional implementation of Tetris. From the link above: * a display with at least 20 rows and 10 columns, and enough read/write random access memory to store the current contents of the display, * a way of repainting one row or column of the display without affecting the…