Talk:Sorting algorithms/Gnome sort

From Rosetta Code

C problem?[edit]

The second j := j + 1 from the example seems to be missing? --Paddy3118 17:04, 27 April 2009 (UTC)

The code works for sure (I've used it in Huffman codes... anyway I've written also a test code and did some testing and it seems to work...); I can't remember, maybe I've "reinterpreted" it or optimized it... I will study if it is still Gnome, or if I've changed it so that it can be called Gnome sort no more. --ShinTakezou 18:28, 27 April 2009 (UTC)
How I can understand it now: the previous C version was simply the original Gnome sort as explained in the Dick Grune citation on Wikipedia (the garden gnome moves always one step forward or backward); the pseudocode is an optimized version, since we don't need to pass through every pot to reach the same position we left when started to go backward; so we can restart from j instead of 1. --ShinTakezou 20:37, 27 April 2009 (UTC)
Third thought (too many distraction sorry): it's true it's an optimization, but not the rest. When the gnome hits the beginning of the list (i==0), elements 0 and 1 are in the proper order, so that if we put i=1, the next iteration will execute the "true part" of the if, i.e. i=j and j++; ... doing it directly in the "then part" of the if, will avoid an extra comparing we know will be true. This optimization apart, it was really the same. --ShinTakezou 20:54, 27 April 2009 (UTC)

Wikipedia markup reference problem[edit]

The Wikipedia markup produces a link to new page when it should link to wp:Gnome_sort. Not sure how to fix the markup. --Dgamey 14:34, 12 September 2010 (UTC)