[ExI] standard form for creating a test, was: RE: humanities plus schmooze
spike66 at att.net
Sun Dec 9 17:13:21 UTC 2012
>... On Behalf Of Mike Dougherty
Subject: Re: [ExI] standard form for creating a test, was: RE: humanities
> On 12/8/12, spike wrote:
>> ****Question code hipsters: is there a standard product or a
>> software framework that allows me to insert or paste in my test
> ... "when your only tool is a hammer, you make all your problems look like
of questions... Mike
[approximate comment by someone here, possibly Mike]: If you need more than
65k rows, Excel is not the right tool...
Agreed that I am the poster child for Excel abuse. What I was doing was a
brute force Scrabble player. I found a list of English words, eliminated
all which have more than 15 letters, and found that with those gone, there
are 354938 words left. I got the code to work, but it was still a practical
failure at Scrabble because that list contained too many words that would
piss off one's opponent if you played them. For instance, starting at the
The first word is a single letter, useless for Scrabble, the second word is
worth 2 points, and then I am all the way down to word 14, aardvark, before
I get to anything I would consider a legitimate Scrabble word. At the other
end of the alphabet, we have:
so the term zygous is the last one on that list I would consider a
legitimate Scrabble word, the biology term which I think is the adjective
form referring to a zygote. Any word I have never heard of in my tragically
many years of info-grazing, I would consider illegitimate for Scrabble.
Indeed, Microsloth Word marks all but 48 of these 58 words as misspellings,
but oddly enough, accepts zyzzyva and zyzzyvas as legitimate, which are also
useless in Scrabble anyway because they contain 3 Zs.
Regarding Scrabble, I had an AHA insight. I took my list of words, wrote a
script which took each word and arranged them in alphabetical order, then
made a table with the words listed in alphabetical order by their
alphabetical order. Then I take my Scrabble tiles and arrange them in
alphabetical order, do a table lookup and find all the possible words that
match my tiles, which usually aren't very many. So the code works, but the
list is too long, and includes about three quarters of the list as words you
or I wouldn't use in Scrabble. I am not about to go through 355000 words
and eliminate those I don't know. Nor will I attempt to create a list that
I do know.
But I agree this is a classic case of Excel abuse.
More information about the extropy-chat