Those annoying adages...
Noam D. Elkies
A stitch in time saves nine.
Actions speak louder than words.
April showers bring May flowers.
As you make your bed, so must you lie in it.
Beggars can't be choosers.
Better late than never.
Dead men tell no tales.
Every dog has his day.
Fools rush in where angels fear to tread.
=
Absence makes the heart grow fonder.
All that glitters is not gold.
As you sow, so shall you reap.
Better be safe than sorry.
Blood is thicker than water.
Business is business.
Charity begins at home.
Dead men have no friends.
Divide and rule.
Don't go near the water until you learn how to swim.
Every man has his price.
---------------------------------------------------
[Anagram confirmed by http://www.anagramgenius.com/checker.html;
proverbs as seen on pages.prodigy.net/jmiller.cb/pr11.html
with one correction (it said "brings May flowers"). Several other
proverbs there are versions I'm not familiar with -- e.g. I thought
it was "As ye sow, so shall ye reap", "Better safe than sorry"
[no "be"], and "Divide and conquer" (rather than "rule") --
but I don't know where to find a similarly-sized collection
of the standard versions.]
NDE
Rick Rothstein
Rick
"Noam D. Elkies" <elkie...@h.harvard.edu> wrote in
message news:abchdf$up8$1...@news.fas.harvard.edu...
Hans-Peter Reich
Is this generated with the same program as in the country-name thread??
Nice tool.
And here is another annoying one:
Tastes differ ~ if feed starts.
And there is no doubt that
Red fits feast.
Cheers
--
---
Hans-Peter
ha...@x-mail.net
"Noam D. Elkies" <elkie...@h.harvard.edu> schrieb im Newsbeitrag
news:abchdf$up8$1...@news.fas.harvard.edu...
Adrian H
Noam D. Elkies wrote
> A bird in the hand is worth two in the bush.
>
> A stitch in time saves nine.
[SNIP]
David Bourke
db
Mey K.
>
> A stitch in time saves nine.
>
> Actions speak louder than words.
>
> April showers bring May flowers.
>
> As you make your bed, so must you lie in it.
>
> Beggars can't be choosers.
>
> Better late than never.
>
> Dead men tell no tales.
>
> Every dog has his day.
>
> Fools rush in where angels fear to tread.
>
> =
>
> Absence makes the heart grow fonder.
>
> All that glitters is not gold.
>
> As you sow, so shall you reap.
>
> Better be safe than sorry.
>
> Blood is thicker than water.
>
> Business is business.
>
> Charity begins at home.
>
> Dead men have no friends.
>
> Divide and rule.
>
> Don't go near the water until you learn how to swim.
>
> Every man has his price.
Unacceptable, Noam. Where are the cheats??? ;)
Unbelievable and a huge Special Nom (you didn't even get to ones later
in the Alphabet! I'm scared to think what would happen if you did).
Later!,
Mey K.
Mick Tully
David Bourke <david_b...@hotmail.com> wrote in message
news:3cda62c5$0$8513$cc9e...@news.dial.pipex.com...
> One of the cleverest posts I've seen.
Yes, exceptional!
Mick
Graham Medlock
"Mick Tully" <mick....@ntlworld.com> wrote in message
news:zJwC8.6606$xb4.1...@news6-win.server.ntlworld.com...
Absolutely brilliant
Graham
Zoran Radisavlevic
Hi.
Everybody says "brilliant","perfect","supercalifragilistic",
but before I gave my opinion I must know is that work done by
hand or it was computer program (same as MALTA+IRAQ = QATAR+MALI)
Zoran
Phil Carmody
> Hi.
>
> Everybody says "brilliant","perfect","supercalifragilistic",
> but before I gave my opinion I must know is that work done by
> hand or it was computer program (same as MALTA+IRAQ = QATAR+MALI)
Indeed. I hate to piss on the fire, but while it's a great find, the birthday paradox tells us that with enough messages (subsets of adages) you will find two that hash to the same value (have the same number of each letter), and the number required is only the square root of the number required to find an exact match to a prespecified hash, i.e be an anagram of something external. The fact that a solution was found in the first 5 letters of the alphabet indicates that there are dozens more.
When I saw it, the _first_ thought that went through my mind was not "that's superb", but "someone's written the program I've been planning to write for months".
Great coding Noam.
Bugger, and I was going to ask a question about the Schoof-Elkies-Atkin algorithm....
Phil
William Tunstall-Pedoe
<firstnam...@nokia.com> writes
>Zoran Radisavlevic wrote:
>> Hi.
>>
>> Everybody says "brilliant","perfect","supercalifragilistic",
>> but before I gave my opinion I must know is that work done by
>> hand or it was computer program (same as MALTA+IRAQ = QATAR+MALI)
>
>Indeed. I hate to piss on the fire, but while it's a great find, the birthday
>paradox tells us that with enough messages (subsets of adages) you will find two
>that hash to the same value (have the same number of each letter), and the
>number required is only the square root of the number required to find an exact
>match to a prespecified hash, i.e be an anagram of something external.
This is true but the size, range and probability distribution of the
possible hash values is a fundamental part of this. The hash, as you
call it, is a vector of 26 integers (giving a count of each letter). It
isn't instantly obvious until you try (at least to me) that the
unimaginably large numbers generated by the birthday paradox maths
outweigh the unimaginably large number of possible vectors. Having said
that I can see how counting combinations of items and then counting
pairs of those combinations can get to some pretty spectacular
numbers...
Anyway, isn't this kind of maths the entire basis for this group? What
makes one go "wow" at a brilliant anagram is the difficulty the brain
has in understanding how a perfect, grammatical anagram can be so
relevant and yet not be the result of some force other than coincidence.
The explanation is always rational but the anagrams can still be amazing
nonetheless!
>The fact
>that a solution was found in the first 5 letters of the alphabet indicates that
>there are dozens more.
>
Possibly millions more.
Does anyone remember Mike Keith's "doubly true" anagrams that were
created with the names of elements where both sides of the equation were
not only anagrams of each other but where their atomic weights summed to
the same total too?
There were also the arithmetic equations by Richard G similar to the
classic "Eleven plus two" -> "Twelve plus one" anagram.
Perhaps we need some additional constrains on this problem as well. Any
ideas?
[...]
William
---
Author of Anagram Genius the software and book
http://www.genius2000.com/ag.html
Noam D. Elkies
investigative, to my post here last week. I was particularly glad
to see Adrian Hickford's comment, since it's his Steven Wright anagram
(the Special Anagrammy winner in February) that' the closest predecessor.
Zoran wrote:
> but before I gave my opinion I must know is that work done by
> hand or it was computer program (same as MALTA+IRAQ = QATAR+MALI)
I thought that computer assistance was accepted practice here --
you didn't really think I could have found my anagram purely by hand?
However, the country-pair anagrams are a red herring; see below.
Phil Carmody:
>Indeed. I hate to piss on the fire, but while it's a great find,
>the birthday paradox tells us that with enough messages
>(subsets of adages) you will find two that hash to the same value
>(have the same number of each letter), and the number required
>is only the square root of the number required to find an exact
>match to a prespecified hash, i.e. be an anagram of something external.
Now that Phil is relieved ;-) -- yes, one subtext of my post
was to suggest this kind of analysis. However, more needs to be said,
since there are some 10^35 possible "hashes" for a set of about 250
English letters. This is way more than the square of the number of
adage subsets of size 10 or 11 taken from the 80 or so with initial
letters A through F, of which there are about ten trillion.
William Tunstall-Pedoe comes a bit closer with:
>This is true but the size, range and probability distribution of the
>possible hash values is a fundamental part of this. The hash, as you
>call it, is a vector of 26 integers (giving a count of each letter).
>It isn't instantly obvious until you try (at least to me) that the
>unimaginably large numbers generated by the birthday paradox maths
>outweigh the unimaginably large number of possible vectors. Having said
>that I can see how counting combinations of items and then counting
>pairs of those combinations can get to some pretty spectacular numbers...
See above. The probability distribution clearly must play a part here
-- the distribution of "hashes" is very far from random -- and the tough
question is how to quantify its effect.
The other subtext is the computational one of how one actually finds
one of the anagram that are known or surmised to exist "out there".
The suggestion of using the merge-and-sort trick that produced
GABON + ITALY = LIBYA + TONGA is wildly impractical: my computer
can easily accommodate and sort the twenty thousand country pairs,
but I don't have nearly enough RAM to store, let alone sort, ten trillion
potential anagrams! While there are ways of doing this computation
that are not as memory-intensive, just generating the 10^13 combinations
would take months on a gigahertz machine. Fortunately there's a better way...
There is also the artistic question of what list of phrases or quotes
to use for such a task. It should be a reasonably large collection,
and the phrases should be familiar enough that the anagram is striking:
a URL for the phrases should be needed only for confirmation that I
didn't "cheat" by adjusting the grammar to fit the anagram. I couldn't
think of many possibilities besides the "annoying adages" -- and the
Steven Wright lines that Adrian already used. Ogden Nash poems would
probably work but I couldn't find more than ~20 of them in one place online.
Shakespeare quotes, perhaps?
At any rate, even if in the final analysis my post was something of
a troll, it should at least be a more interesting troll than the recent
rash of pot- and potty-themed off-topic offal :-)
NDE
Adrian H
> Thanks for the several replies, enthusiastic as well as (ahem)
> investigative, to my post here last week. I was particularly glad
> to see Adrian Hickford's comment, since it's his Steven Wright anagram
> (the Special Anagrammy winner in February) that' the closest predecessor.
It's an interesting question - does the method used to achieve
anagrams of this kind this make any difference to the quality of the
'gram? Mick recently quipped that on his wish-list for a future
version of AG, he wanted "AG to produce, automatically, interesting
subject matter at the push of an extra button, produce fantastic
anagrams, possibly using the additional assistance of William's very
clever 'Crossword Maestro' software."
I think Noam has started the ball rolling on this one, Mick!
If you don't yet know about him, a quick Internet search on the
Professor will be quite an eye-opener.
Number theoretician = Be arithmetic, neuron!
In tribute, cheer, "Noam!"
---
Adrian
Lardy Girl
> Thanks for the several replies, enthusiastic as well as (ahem)
> investigative, to my post here last week. I was particularly glad
> to see Adrian Hickford's comment, since it's his Steven Wright anagram
> (the Special Anagrammy winner in February) that' the closest predecessor.
While I do think there is less art in anagrams where items can be shuffled
between the two sides (so to speak) until something works out, I wouldn't
want to dismiss the craft involved in writing good programs which can do
the bulk of such work.
Having said that,
> I thought that computer assistance was accepted practice here --
> you didn't really think I could have found my anagram purely by hand?
most people here do use some form of computer assistance for anagramming,
so it would be hard to draw a line re: when a person vs a program
should be given credit for an anagram's discovery.
While I think we are a long way from, say, AG finding subject matter,
submitting its findings automatically to the archives, NOMming them, etc.,
a darkish grey position has been established on the scale now (no, Noam
may not don a black hat just yet). I must admit that I derived some added
pleasure from not relying on computer assistance at all with my sima recipe
/May Day anagram (When I do medium to long 'grams, I generally use a
program (an) to see what words can be made from the last 10 or so letters,
though I do the rest by hand). Yet being lily white in this case won't
stop me from acknowledging that there are all types of anagrams and that
programs such as Noam's aren't bastard offspring of 'proper anagramming'.
Lardy,
doing random nattering to avoid work
--
A n n a S h e f l -- Abandon 'safety' to reply
"Never eat more than you can lift." -- Miss Piggy
-=#=-
Anna's News Clippings archive and more: http://theanna.org/
William Tunstall-Pedoe
<elkie...@h.harvard.edu> writes
>Thanks for the several replies, enthusiastic as well as (ahem)
>investigative, to my post here last week.
[...]
>Zoran wrote:
>
>> but before I gave my opinion I must know is that work done by
>> hand or it was computer program (same as MALTA+IRAQ = QATAR+MALI)
>
>I thought that computer assistance was accepted practice here --
It is.
However, for anagrams like this, it is reasonable for people to want to
know how it was done so they can determine whether their admiration is
for your computer programming skill or sheer bloody-minded persistence
messing around with the letters! Although it was pretty clear to me that
you used a computer, there are certainly people in this group who would
attempt such a task by hand!
[...]
>
>Phil Carmody:
>
>>Indeed. I hate to piss on the fire, but while it's a great find,
>>the birthday paradox tells us
[...]
> yes, one subtext of my post
>was to suggest this kind of analysis. However, more needs to be said,
>since there are some 10^35 possible "hashes" for a set of about 250
>English letters. This is way more than the square of the number of
>adage subsets of size 10 or 11 taken from the 80 or so with initial
>letters A through F, of which there are about ten trillion.
>
>William Tunstall-Pedoe comes a bit closer with:
>
>>This is true but the size, range and probability distribution of the
>>possible hash values is a fundamental part of this.
[...]
>See above. The probability distribution clearly must play a part here
>-- the distribution of "hashes" is very far from random -- and the tough
>question is how to quantify its effect.
>
This effect is also related to anagramming generally.
As everyone who has tried creating anagrams will know, subject texts
vary enormously in how "good" they are and this is not just a case of
having rare letters. Some names with only common letters are virtually
impossible simply because they have an unusual distribution. This is
also why the anagramming of non-English subject texts (into English) is
often so difficult: the distribution of letters is more likely to be
atypical.
With a bit of thought it should be clear that the skewness of the
distribution also partly explains why so many good anagrams exist at
all. If the frequency of letters in natural English was more evenly
distributed through the alphabet far fewer gems would be there to be
found.
This thought has made me start thinking about how suitable English is
for anagramming compared with other languages and whether some kind of
mathematical analysis could be done to get a handle on this. For a
start, it would be interesting to compare letter frequencies in
different languages and see how accurate the skewness of the
distribution is in predicting (some measure of) their anagrammability.
There must be a number of other factors relating to how good a language
is for anagramming including the grammatical flexibility of the language
and the efficiency by which meaning can be expressed with a given number
of letters. This must also include the number of words in the language
(the more the better) and the lengths of those words (the shorter the
better). All these points suggest that English *may* be very suited to
anagrams in comparison with other languages, the latter generally having
longer word lengths, more restrictive grammars and fewer words.
Can any multi-lingual readers of this group offer any opinions about how
easy their languages are for anagramming compared with English?
>The other subtext is the computational one of how one actually finds
>one of the anagram that are known or surmised to exist "out there".
>The suggestion of using the merge-and-sort trick that produced
>GABON + ITALY = LIBYA + TONGA is wildly impractical: my computer
>can easily accommodate and sort the twenty thousand country pairs,
>but I don't have nearly enough RAM to store, let alone sort, ten trillion
>potential anagrams! While there are ways of doing this computation
>that are not as memory-intensive, just generating the 10^13 combinations
>would take months on a gigahertz machine. Fortunately there's a better way...
>
Care to share? Presumably you are doing some kind of non-exhaustive,
goal-directed search?
Computer search problems, especially ones involving words, are a
particular interest of mine.
>There is also the artistic question of what list of phrases or quotes
>to use for such a task. It should be a reasonably large collection,
>and the phrases should be familiar enough that the anagram is striking:
>a URL for the phrases should be needed only for confirmation that I
>didn't "cheat" by adjusting the grammar to fit the anagram.
These points are very important not just from an artistic point-of-view.
Anagramming is about making the reader fall off their chair in amazement
that such a thing could be both true and yet also be nothing more than
the result of coincidence. Amazement at the latter is far less likely if
the material appears contrived in any way.
>I couldn't
>think of many possibilities besides the "annoying adages" -- and the
>Steven Wright lines that Adrian already used. Ogden Nash poems would
>probably work but I couldn't find more than ~20 of them in one place online.
>Shakespeare quotes, perhaps?
>
What about song titles of a particularly prolific artist? A little bit
of web-searching should find some sources. e.g.
http://www.yimpan.com/Songsite/Lyric/index.asp?aid=133&page=1&pagesize=2
0&order=title
http://www.musicroom.com/se/ID_No/0016971/details.ihtml
Mick Tully
Adrian H <a.j.hi...@soton.ac.uk> wrote in message
news:922ebf00.02051...@posting.google.com...
> Noam D. Elkies wrote
> > Thanks for the several replies, enthusiastic as well as (ahem)
> > investigative, to my post here last week. I was particularly glad
> > to see Adrian Hickford's comment, since it's his Steven Wright anagram
> > (the Special Anagrammy winner in February) that' the closest
predecessor.
>
> It's an interesting question - does the method used to achieve
> anagrams of this kind this make any difference to the quality of the
> 'gram? Mick recently quipped that on his wish-list for a future
> version of AG, he wanted "AG to produce, automatically, interesting
> subject matter at the push of an extra button, produce fantastic
> anagrams, possibly using the additional assistance of William's very
> clever 'Crossword Maestro' software."
I sure did, but as you say I was only joking. When the software does the
job all on its own, I always like to give it the credit. On the other hand
it is just as much a tool for the letter-worker as is a power-saw to a
carpenter, taking out the drudgery and finishing the job quicker. However,
I pay tribute to wordsmiths of the past who produced such wonderful classics
without benefit of such help.
>
> I think Noam has started the ball rolling on this one, Mick!
Indeed! I'm amazed.
>
> If you don't yet know about him, a quick Internet search on the
> Professor will be quite an eye-opener.
You are quite right, Adrian. Thanks for the tip.
>
> Number theoretician = Be arithmetic, neuron!
> In tribute, cheer, "Noam!"
I agree with that sentiment absolutely. Very nice 'gram.
Zoran Radisavlevic
are starting with "A" I realized that I was right about my suspicions.
While most of you think it's normal to click on "enter" and get some
anagrammatic sentences where both sides are equal(with counting of
subjects)I think it doesn't have anything with art of anagramming.Do
you
remember,anagrams = ars magna ?
I approve partially using of "Anagram genius" and "Anagram artist",
especially for long anagrams.I don't believe that many of you
have find some new and brilliant short anagrams only with
computer;in most of cases you have an idea and computer
give some help but not 100%.But this Noam's work is totally
without sense and it's decadency of anagramming.
Works like that are just examples of computer's possibilities.
There's no skillness of anagrammer but skillness of programmer.
And,what's the purpose ?I am expecting from anagram some nice
explanation of phrase or sentence on left side.I repeat,anagrams
with counting on both sides done by computer are total decadency.
Zoran
Noam D. Elkies <elkie...@h.harvard.edu> wrote in message news:<abs0tr$43n$1...@news.fas.harvard.edu>...