[svn-inject] Tagging upstream source version of scowl

[deb_pkgs/scowl.git] / 6 / r / uk-freq-class / notes.txt

diff --git a/6/r/uk-freq-class/notes.txt b/6/r/uk-freq-class/notes.txt
new file mode 100644 (file)
index 0000000..bc7af0a
--- /dev/null
+++ b/6/r/uk-freq-class/notes.txt
@@ -0,0 +1,68 @@
+     UK English Wordlist With Frequency Classification
+
+This wordlist is primarily intended to be useful for
+checking spelling. Editorial policy is conservative.
+
+Principal omissions:
+
+   -   words requiring a capital letter
+   -   abbreviations
+   -   slang
+
+Colloquialisms and archaisms are generally excluded. A rare
+word similar to a common word may be excluded. Both -ise and
+-ize spellings are included.
+
+The character set is: lowercase letters, hyphen, apostrophe.
+Words which can be spelt with accents occur here in their
+plain form.
+
+If this wordlist is to be used with ispell the following
+lines may be appropriate for the affix file:
+
+   boundarychars [---]
+   boundarychars '
+   wordchars [a-z] [A-Z]
+
+The commonest words are labelled 16 and the least common 0.
+
+Coverage of common words should be good, but note the
+categories excluded.
+
+                        Brian Kelk bck22@bckelk.uklinux.net
+                        April 2002
+
+
+Here are bits of a brief conversation I had with the author:
+
+From: Brian Kelk <Brian.Kelk@cl.cam.ac.uk>
+Date: Sat, 08 Jul 2000 20:27:21 +0100
+
+> I was wondering what the copyright status of your "UK English Wordlist
+> With Frequency Classification" word list as it seems to be lacking any
+> copyright notice.  Also, how did you arrive at the "Frequency
+> Classification".
+
+There were many many sources in total, but any text marked
+"copyright" was avoided. Locally-written documentation was one
+source. An earlier version of the list resided in a filespace
+called PUBLIC on the University mainframe, because it was
+considered public domain.
+
+Briefly about frequency: rather than counting occurrences of
+a word this classification is more along the lines of counting
+the number of texts in which the word occurs. That way you
+get some noise immunity, which you very much need. It's based
+on maybe 5-10 million words of text on the Cambridge mainframe
+in the 1980s. I had in mind that it might be useful for ranking
+possible corrections ...
+
+Date: Tue, 11 Jul 2000 19:31:34 +0100
+
+> So are you saying your word list is also in the public domain?
+
+That is the intention.
+
+
+
+