Clojure transducers from the ground up: using them in practice

Re:Clojure transducers from the ground up: using them in pra... by exikyut on Sunday October 29, 2017 @07:29AM (#15578916)

This is not related to electrical transducers, which (without any additional scoping context) was what I thought this article was about.

Relabeling it "Clojure transducers" would help the title in every way. But it's been up for 7 hours now, so we'll see if it changes.

[ Parent | on HN ]

Re:Clojure transducers from the ground up: using them in pra... by ulber on Sunday October 29, 2017 @07:42AM (#15578943)

Transducer is the standard term inside CS for this model of computation [1]. Often there are some disambiguating terms, such as finite-state transducer, but this article does not limit itself to any particular kind. Of course Hacker News has content from both fields so the overloading of the term is unfortunate.

[1] https://en.m.wikipedia.org/wiki/Finite-state_transducer

[ Parent | on HN ]

Re:Clojure transducers from the ground up: using them in pra... by moomin on Sunday October 29, 2017 @10:00AM (#15579324)

Transducers don’t really directly map to any existing CS construct. In fact, Clojure transducers occupy a strange twilight world between “fold endomorphism” and “stream transformer”. Most transducers are both. Some are just one or the other.

[ Parent | on HN ]