I got wind of a new-ish public word list for crossword constructors, the spread the word(list). So I grabbed a copy and tossed it into the big pile of data that feeds the "Phraser" phrase and word lists.

I've got a few crossword lists in that big pile of data now. I want to give them lest "oomph" in how much they boost things. Crossword constructors don't care about spaces between words: they happily lay BRITISHCOLUMBIA into a grid; but in other contexts you'd expect a space between British and Columbia. Because I pull in data from crossword word lists and from sources that use spaces, my Phraser lists now have both "british columbia" and "britishcolumbia". That's not terrible: space-less britishcolumbia is pretty far down in the list amongst obscure words like "brisco" and "bunglers".

But I notice that the missing-spaces "howardtheduck" is now higher in my list than "howard the duck". As Howard would say, "Waak." Trapped in a wordlist he never made. I just now changed the Phraser program so that it won't boost long space-less "words" it finds in crossword lists as much. I won't jump in and re-generate the Phraser lists right away with the new program; but the next time I do, space-less howardtheduck should be much lower.