Talk:Look-and-say sequence

From Rosetta Code

Start of Sequence

In example output might be best shown as '1' rather than the '11' of some examples? --Paddy3118 06:01, 17 May 2009 (UTC)

You mean, every example should show the "generator" '1', instead of starting with '11'? --ShinTakezou 11:35, 17 May 2009 (UTC)
Yes, the WP article starts from 1 and it just seems more natural to me. It wouldn't be a big fix to those language examples that give sample output. --Paddy3118 11:42, 17 May 2009 (UTC)

AutoHotkey / Untested Examples

Should we allow untested examples? There is always the stupid error that may creep in which makes an example that hasn't even been run too prone to error.
Unless the comment means "untested, but I am about to test it now", then maybe contributors should refrain from adding code that even they are so unsure of?
It would be courteous to run code for a task of this complexity, (although the task is not complex), and check it gives some of the results required. --Paddy3118 06:13, 27 November 2009 (UTC)

All these untested examples of mine are from computers that can't run the program. Problem is, I generally forget. I do add {{untested|language}}, though. I'm not "so unsure" of it, I think it'll work, but I haven't tried it. When I get home, I'll run though all the untested AHK examples, when I have time, I'll run through all the untested examples I can run.
Thanks BR, it is appreciated. --Paddy3118 05:23, 28 January 2010 (UTC)

Run-length Encoding

This task is actually identical to the Run-length_encoding task, applied iteratively to strings of digits. In fact, the easiest/best solution is simply to re-apply the RLE task here. I think this should be mentioned in the task description. As it is, some of the code here is far more complicated than it needs to be, because the identity of the two tasks is not pointed out.

I do think it's OK to keep this as a separate task, though, since the Conway sequence is definitely an interesting application of RLE in its own right. --Snoman 20:20, 11 July 2010 (UTC)

Since Paddy3118 added a note about RLE, I went ahead and expanded it a bit. Hope that's OK --Snoman 04:23, 12 July 2010 (UTC)
Thanks :-)
--Paddy3118 04:59, 12 July 2010 (UTC)

PicoLisp

For the Picolisp solution, what happens if input sequence has an element repeated 10 or more times? --Ledrug 03:47, 13 October 2011 (UTC)

Morris Number sequence

(this is moved from a duplicate/to be deleted page, since this thread should be here) eMBee (talk) 03:51, 17 January 2020 (UTC)

Look-and-say is also known as Morris Number sequence. should that be added somewhere to make it easier to find? eMBee (talk) 16:20, 16 January 2020 (UTC)

Yes,   I added a sentence to the Rosetta Code task (mentioned below in another talk section) about the   Morris Number sequence,   and another reference as well, including a book that it was referenced in.     -- Gerard Schildberger (talk) 02:18, 17 January 2020 (UTC)
I also added the phrase   look and say   (without the hyphens)   so that Rosetta Code task can be found when performing a search without the punctuation between the keywords.     -- Gerard Schildberger (talk) 02:23, 17 January 2020 (UTC)


look-and-say versus run-length-encoding

I generalized the   string of digits   to a   string of characters   (for the REXX language example).   There is nothing magical about decimal digits,   any characters will do   (but should be limited to displayable characters for obvious reasons, and a   blank   character becomes problematic).   Also, the starting/initial number   (or character string)   can be specified.   See the 2nd REXX output entry for a non-numeric example.     -- Gerard Schildberger (talk) 21:29, 25 April 2020 (UTC)