Peter, I now appreciate the frustration that Macnutt felt when your alter-ego took on his identity. Still, we all knew it was you. Anyway, I am back, but it is still late here in St.John's, so I shall once again bid you all a fond farewell .....................until tomorrow morning, when that shall be fresh coffee for all waiting for any and all who wander in to the Cafe Chez Marc. Paix.
__________________
Dr.G.
14" G4 iBook
15" MacBook Pro (July, 2009)
13" MacBooK Pro with Retina Display
Paix
"The man who does not read good books has no advantage over the man who can't read these books." Mark Twain
I wrote a script that determines how often one word follows another in a given text file, then generates a random text file based on those frequencies. In theory, the random text would be gramatically-correct nonsense. In practice, it's almost gramatically-correct nonsense.
The quality of the output seems to depend on the size and diversity of the input text. A large amount of text on the same subject (e.g., doxies) works best, while a small amount of text on a varied number of subjects doesn't work well at all (the output text is almost the input text).
For example, Automatic Dr.G. produces the following input after harvesting just a handful of posts:
<pre> Hopefully, it is made in an catch a peaceful sleep in and I have a
post card from to convince the last day: of intelligence just
received a special permit to philosophize with the world, even the
multitudes doxies to lease my two Doxies and worries were lost,
albeit momentarily: convince the south sang to both Iceberg Beer
and West Nile free. Yet (in fine ehmacland). </pre>
while autospectrum tends to produce uninteresting output, even after harvesting a number of different posts:
<pre> Debates with an insult also an insult to use your universe posterboy,
but it a while, so we are using the kettle black, ain't it does (not in
your universe posterboy but it does not mine insulting type posts since it)? </pre>
Then there's autopoole, which is indistinguishable from yours truly! Draw your own conclusions.
<pre> Apple wouldn't be polarized vertically; by the GDP encoding,
of quoting trillion as the state makes up use that if it
right wing, then invest into account would have to one of
quoting trillion as a number of their other countries.
Alice then the unfunded middle attack by determining the
message to in the shortfall. A appears to the results of
physical properties of course, what makes use that any, of
increase in a account would be aware of the debt I'd
imagine the US. When determining their national debt
right wing, then the increase in making it right? </pre>
Of course, after all of this I'm left wondering: if Automatic Dr.G's three posts were a waste of server space, what does that make the Clubhouse thread?
The Club House is where people may OPENLY express their thoughts, ideas and feelings. At least you had the courage to admit you wrote the program, so let's move on with our lives.
Coffee anyone???
__________________
Dr.G.
14" G4 iBook
15" MacBook Pro (July, 2009)
13" MacBooK Pro with Retina Display
Paix
"The man who does not read good books has no advantage over the man who can't read these books." Mark Twain
Let's keep it clean here, Mr Mayor....there may be younger members reading this [img]tongue.gif[/img] (Don't follow Mel)
jfpoole, that was interesting as is the reaction that occurs when one's name/identity is infringed upon., albeit in jest. But, "all's well that ends well", or as the doxies say when summarizing things, "Well, the long and short of it is....." [img]graemlins/lmao.gif[/img]
With doxies wouldn't that be "Well, the short and short of it is...."
Well, no, 'cause you see....doxies are long on the body and short on the legs. Ergo, "long and short". But I know what you meant.
Dr G, I have another movie title to work on......about the silly antics of two vertically challenged dogs (doxies). We'll call it "Short and Shorter"