1. Contest #3 - Winner
The winner of contest #3 is C. K. Lester
Congratulations C.K., you've earned $50 in MicroEconomy money.
I couldn't award the $5 bonus, since there was a glitch when
running under Linux. Something about \r's not
being filtered out from text input on Linux.
I tested each program on a set of 5 input texts.
The score for each test was the percentage of words correct.
The overall score was the average of the 5 scores.
Here are the scores for the 6 programs that were entered:
1. C.K. Lester, seekword.ex 77% overall average
16/19 84%
60/60 100%
0/16 0% - timed out
20/20 100%
16/16 100%
2. jcb.ex (not his initials) 70% overall
18/19 95%
46/60 77%
6/16 38%
15/20 75%
10/16 63%
3. decipher.exw 61% overall
16/19 84%
60/60 100%
0/16 0% -- timed out
18/20 90%
5/16 31%
4. crypto.exw 58% overall
17/19 89%
60/60 100%
0/16 0%
18/20 90%
2/16 13%
5. decrypt.ex 53% overall
14/19 74%
60/60 100%
0/16 0%
18/20 90%
0/16 0%
6. decipher.ex 42% overall
16/19 84%
59/60 98%
0/16 0%
6/20 30%
0/16 0%
Here are the 5 inputs:
Oxtjxs mej ozzswazy zj dsxsvozs voxrjw xkwpsvy pt rszsvwfxfyzfq wsoxy fy,
jc qjkvys, iflfxd fx o yzozs jc yfx.
Ot tkkz ot es tuonusr lnkwnoyyjzw, es ikczr uk kcn tcnlnjts uxou ju eotz'u
ot sotg uk wsu lnkwnoyt njwxu ot es xor uxkcwxu. Rspcwwjzw xor uk ps
rjtqkdsnsr. J qoz nsysypsn uxs sfoqu jztuozu exsz J nsovjhsr uxou o
vonws lonu ki yg vjis inky uxsz kz eot wkjzw uk ps tlszu jz
ijzrjzw yjtuobst jz yg kez lnkwnoyt.
Pmyoxm sn pbtj ch aum opsvm qsrm.
C uovm shfk ixsvmr ca qsxxmqa, hsa axcmr ca.
Orxja fdnxh usgibadx jurdhud nhc nff aqd aqdsxk.
Hdza cdvdfsi n ixspxnggrhp jakfd.
Aqdh osxpda nff aqna nhc mbja qnue.
Ixsmxnggqxj tnfnhuq aeqdx cndfk gqhb ngshm aeq rsbx gnosx rssc mxsbij:
unrrqdhq, jbmnx, mxqnjq, nhc jnfa.
Here are the quotations that I used:
1.
Anyone who attempts to generate random numbers by deterministic means is,
of course, living in a state of sin.
-- John von Neumann
2.
As soon as we started programming, we found to our surprise that it wasn't
as easy to get programs right as we had thought. Debugging had to be
discovered. I can remember the exact instant when I realized that a
large part of my life from then on was going to be spent in
finding mistakes in my own programs.
-- Maurice Wilkes discovers debugging, 1949
3.
Beware of bugs in the above code.
I have only proved it correct, not tried it.
-- Donald Knuth
4.
First learn computer science and all the theory.
Next develop a programming style.
Then forget all that and just hack.
-- George Carrette
5.
Programmers balance their daily menu among the four major food groups:
caffeine, sugar, grease, and salt.
-- From John Walker's The Hacker's Diet
I'll post C.K.'s program later today.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
2. Re: Contest #3 - Winner
On Thu, 4 Apr 2002 18:29:42 -0500, Robert Craig
<rds at RapidEuphoria.com> wrote:
>2. jcb.ex (not his initials) 70% overall
That be me
>18/19 95%
>46/60 77%
>6/16 38%
>15/20 75%
>10/16 63%
I'm completely lost though. I'm happy with 2nd place but cannot see
how you arrived at those figures. The output of my program, interlaced
with cut & pasted excepts from your post is:
(Each entry is marked up with <your score>?<my belief>):
1) 18/19? 19/19
Anyone who attempts to generate random numbers by deterministic means
Anyone who attempts to generate random numbers by deterministic means
is, of course, living in a state of sin.
is, of course, living in a state of sin.
2) 46/60? 60/60
As soon as we started programming, we found to our surprise that it
As soon as we started programming, we found to our surprise that it
wasn't as easy to get programs right as we had thought. Debugging had
wasn't as easy to get programs right as we had thought. Debugging had
to be discovered. I can remember the exact instant when I realized
to be discovered. I can remember the exact instant when I realized
that a large part of my life from then on was going to be spent in
that a large part of my life from then on was going to be spent in
finding mistakes in my own programs.
finding mistakes in my own programs.
3) 6/16? 3/16
Female ow ftjb in she afove rode. I have onyx gloved is rollers, nos
Beware of bugs in the above code. I have only proved it correct, not
slied is.
tried it.
4) 15/20? 20/20
First learn computer science and all the theory. Next develop a
First learn computer science and all the theory. Next develop a
programming style. Then forget all that and just hack.
programming style. Then forget all that and just hack.
5) 10/16? 15/16
Programmers valance their daily menu among the four major food groups:
Programmers balance their daily menu among the four major food groups:
caffeine, sugar, grease, and salt.
caffeine, sugar, grease, and salt.
What results did you get?
I'll confess to 3 being mainly rubbish but I can only see three
correct words whereas you've marked up six. Otherwise, the only thing
I can think of is I got marked down for not preserving line breaks
&/or the odd extra bit of whitespace in the output. (My program
automatically breaks the output lines at the first word break before
column 75)
Pete
3. Re: Contest #3 - Winner
This is a multi-part message in MIME format.
------=_NextPart_000_00D1_01C1DC36.DB596C60
charset="iso-8859-1"
Pete writes:
> I'm happy with 2nd place but cannot see
> how you arrived at those figures.
I ran the second one again, just now, and I get the same
result (46/60) that I reported. The second last
line of your output is all wrong for some reason.
Maybe you fixed a bug but sent the wrong version to me.
I've attached the .zip that you sent me.
Maybe someone else can check it.
I claim that the attached .zip will give:
As soon as we started programming, we found to our surprise that it wasn't
as easy to get programs right as we had thought. Debugging had to be
discovered. I can remember the exact instant when I realized that a
aumsd jump ie to aged emit pzdn in buy signs pi rd yjdnp gn
finding mistakes in my own programs.
when the input file is:
Ot tkkz ot es tuonusr lnkwnoyyjzw, es ikczr uk kcn tcnlnjts uxou ju eotz'u
ot sotg uk wsu lnkwnoyt njwxu ot es xor uxkcwxu. Rspcwwjzw xor uk ps
rjtqkdsnsr. J qoz nsysypsn uxs sfoqu jztuozu exsz J nsovjhsr uxou o
vonws lonu ki yg vjis inky uxsz kz eot wkjzw uk ps tlszu jz
ijzrjzw yjtuobst jz yg kez lnkwnoyt.
I ran it as:
ex jcb < input2.txt
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
------=_NextPart_000_00D1_01C1DC36.DB596C60
Content-Type: application/x-zip-compressed;
name="jcb.ZIP"
4. Re: Contest #3 - Winner
On Fri, 5 Apr 2002 00:14:30 -0500, Robert Craig
<rds at RapidEuphoria.com> wrote:
Ah. If you'd chucked the whole thing at it without line breaks (as my
tests did) it would have worked.
If anyone wants to play with it you might want to change the final
loop to:
object input, wholetext
wholetext={}
while 1 do
-- printf(1,"Enter Cipher ",{})
input=gets(0)
if atom(input) then exit end if
if input[length(input)]=10 then
input=input[1..length(input)-1]
end if
if length(input)=0 then exit end if
tt=time()
-- main(input)
wholetext&=input&" "
end while
main(wholetext)
Then it'd get an 82.5% rating.
Pete
5. Re: Contest #3 - Winner
Congratulations to the C. K. Lester (and all those who submitted an entry)!
I ran the samples through my (unsubmitted) program and verified that, indeed,
my code didn't have a chance. The genetic algorithms eventually stumbled on a
solution (of sorts), but typically only long after the allotted time was
passed.
-- David Cuny
6. Re: Contest #3 - Winner
C.K. writes:
> Rob, I show 100% for the first one. It's solution #9. 100% correct. Can
> you verify or demonstrate why you say I only got 84%?
Your program spits out 22 different answers for:
Oxtjxs mej ozzswazy zj dsxsvozs voxrjw xkwpsvy pt rszsvwfxfyzfq wsoxy fy,
jc qjkvys, iflfxd fx o yzozs jc yfx.
I decided in advance that any program that spits
out multiple answers would be judged on the
last answer appearing on my screen. I didn't look
at the other answers. I can see now that one
of your 22 answers is 100% correct, but the last
one is only 84% correct (3 words wrong).
On March 11, Rob Craig wrote:
> To make my life simpler, I'll need you to guess at *one*
> possible solution. Perhaps you could pick the most
> common word, or at least the most common letter,
> and hope for the best.
The Solutions
01: [18] "Anyone flo attempts to generate random numbers by deterministic means
is,
oh course, viking in a state oh sin.
"
02: [18] "Anyone flo attempts to generate random numbers by deterministic means
is,
ow course, hiking in a state ow sin.
"
03: [18] "Anyone flo attempts to generate random numbers by deterministic means
is,
ow course, viking in a state ow sin.
"
04: [18] "Anyone flo attempts to generate random numbers by deterministic means
is,
ox course, hiking in a state ox sin.
"
05: [18] "Anyone flo attempts to generate random numbers by deterministic means
is,
ox course, viking in a state ox sin.
"
06: [18] "Anyone flo attempts to generate random numbers by deterministic means
is,
oz course, hiking in a state oz sin.
"
07: [18] "Anyone flo attempts to generate random numbers by deterministic means
is,
oz course, viking in a state oz sin.
"
08: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
of course, liking in a state of sin.
"
09: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
of course, living in a state of sin.
"
10: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
of course, viking in a state of sin.
"
11: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
ok course, filing in a state ok sin.
"
12: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
ok course, fixing in a state ok sin.
"
13: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
ok course, living in a state ok sin.
"
14: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
ox course, filing in a state ox sin.
"
15: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
ox course, liking in a state ox sin.
"
16: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
ox course, living in a state ox sin.
"
17: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
ox course, viking in a state ox sin.
"
18: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
oz course, filing in a state oz sin.
"
19: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
oz course, fixing in a state oz sin.
"
20: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
oz course, liking in a state oz sin.
"
21: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
oz course, living in a state oz sin.
"
22: [18] "Anyone who attempts to generate random numbers by deterministic means
is,
oz course, viking in a state oz sin.
"
Total Program Time: 19 seconds
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
7. Re: Contest #3 - Winner
Rob,
I think the contest was a great idea. I would be interested in seeing a
brief post-mortem, discussing the approaches and relative merits of the best
entries, and if there were any surprising aspects to any of the entries.
I hope to see more contests in the future.
Colin
----- Original Message -----
From: "Robert Craig" <rds at RapidEuphoria.com>
To: "EUforum" <EUforum at topica.com>
Sent: Thursday, April 04, 2002 7:29 PM
Subject: Contest #3 - Winner
>
>
> The winner of contest #3 is C. K. Lester
8. Re: Contest #3 - Winner
Colin Taylor writes:
> I would be interested in seeing a
> brief post-mortem, discussing the approaches and relative merits
> of the best entries, and if there were any surprising aspects to
> any of the entries.
That would be interesting, but I'm not in a good position to do it.
I wrote a solution for #1, without any concern about speed.
I did not try to solve #2 or #3. I only thought about them long enough to
decide if they were too hard or too easy. I was concerned that #3
might be too hard, but we ended up with 6 good solutions.
I don't want to read through, and analyze all the code.
Assigning a score was enough work. 
I also do not have permission to disclose the code
for most of the non-winning solutions.
Maybe the winners or the other contestants would like
to describe their algorithms and what they learned.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
