26

The wise other editors at Wikipedia have declined my request to move the Wikipedia article on the Rabin–Karp algorithm to what I think it should be called, the Karp–Rabin algorithm, on the basis that the Rabin–Karp name is used more often (false, if one goes by Google scholar's numbers), or that it sounds better out loud (really?). The original publication name order was Karp and Rabin, alphabetically as it usually is for theory papers, which is why I requested the move.

One major proponent of the Rabin–Karp name ordering is the Cormen–Leiserson–Rivest–Stein Introduction to Algorithms textbook. The Wikipedia outcome is unlikely to change barring significant new evidence one way or another, and it seems unlikely that either Rabin or Karp cares about $\epsilon$ more credit, but now I'm curious: Do any readers remember the history of this and have any explanation why CLRS (or anyone else) chose the Rabin–Karp name order?

David Eppstein
  • 50,990
  • 3
  • 170
  • 278
  • You have convinced me. I can try and help with the move. – Tyson Williams May 14 '14 at 10:32
  • 2
    Too late. The move request is closed. One reason I didn't post here while it was still open is that outside canvassing on things like that is discouraged. – David Eppstein May 14 '14 at 15:12
  • 5
    Crochmore & Rytter (2003): Jewels of Stringology — Text Algorithms, a popular text book on string algorithms, uses the version "Karp–Rabin". The book does not cite any paper with authors "Rabin & Karp" in this order. The index does not mention "Rabin–Karp". – Jukka Suomela May 14 '14 at 17:15

1 Answers1

25

I've asked Dick personally out of curiosity a few years back. He said that as far as he knows Rabin-Karp was a random switch many years after the paper was first published. He also indicated that it is his understanding that Michael would say the same thing if asked, since at some point they had talked about it.

Alex Lopez-Ortiz
  • 266
  • 3
  • 3