Talk:Semordnilap: Difference between revisions
m
→Semordnilap Phrases
(add comment regarding dead link to input file) |
|||
| (4 intermediate revisions by 3 users not shown) | |||
Line 28:
The file to be used for input appears to be missing or blank. I attempted to use a file of the same name I found in someone's github repository, but I am getting different output from the other examples and I suspect it may be a different file. Any advice on how to proceed? --[[User:Chunes|Chunes]] ([[User talk:Chunes|talk]]) 13:37, 3 November 2018 (UTC)
:After some poking around the website, I found this working link: http://wiki.puzzlers.org/pub/wordlists/unixdict.txt. I'll give this one a shot. --[[User:Chunes|Chunes]] ([[User talk:Chunes|talk]]) 13:49, 3 November 2018 (UTC)
::The issue was that my code was incorrect. Nevertheless, I've updated the link in the task description to the working one. --[[User:Chunes|Chunes]] ([[User talk:Chunes|talk]]) 14:36, 3 November 2018 (UTC)
== Semordnilap Phrases ==
It's perhaps worth noting that Semordnilap phrases pose something of an open-ended problem (unless a specific phrase length limit is imposed). This would be complicated if there were also a requirement that the phrases represent valid english grammar. Perhaps the task should be clarified to eliminate phrases from consideration? I didn't notice anyone implementing support for them... --[[User:Rdm|Rdm]] ([[User talk:Rdm|talk]]) 02:09, 30 January 2020 (UTC)
: Semordnilap phrases is not just an "open-ended problem"; the problem is not even well-defined, and thus open for interpretation.
: Your interpretation, I assume, is that a Semordnilap phrase ''P'' is a concatenation of some words ''x1...xm'' contained in the dictionary (possibly delimited by punctuation such as spaces, commas, etc.), such that its reversal ''rev(P) = rev(xm)...rev(x1)'' is also a concatenation of some words in the dictionary, viz ''rev(P) = y1...yn'' for some words ''y1...yn'' in the dictionary. In other words, if you remove all punctuation from a Semordnilap phrase and reverse it, then you can divide the resulting word into one or more words of the dictionary. It is obvious that there are infinitely many such Semordnilap phrase pairs, so as you mentioned, a phrase length limit should be imposed.
: My initial interpretation was that a Semordnilap phrase is a phrase contained in the dictionary (eg. a phrase forming one line of the dictionary file). Then there are finitely many such phrases (assuming a finite dictionary), and phrase X is a Semordnilap of phrase Y iff X stripped of punctuation equals the reversal of Y stripped of punctuation.
: Your interpretation is, of course, more meaningful; but besides the problem of infinite solution space, there is the problem of grammatical correctness of both phrases, which would be rather difficult to deal with. That considered, I suggest removing the mentions of "phrases" from the task description altogether, and perhaps clarifying that dictionary entries are simple words (not containing any punctuation). [[User:Dick de Bill|Dick de Bill]] ([[User talk:Dick de Bill|talk]]) 13:52, 3 February 2020 (UTC)
:: Exactly -- grammatical ambiguities are out of scope. --[[User:Rdm|Rdm]] ([[User talk:Rdm|talk]]) 16:28, 3 February 2020 (UTC)
| |||