Mailing List Archive: 49091 messages
  • Home
  • Script library
  • AltME Archive
  • Mailing list
  • Articles Index
  • Site search
 

RRGib - Random Rebol Gibberish

 [1/18] from: arolls:bigpond:au at: 31-May-2001 15:44


Hello, I am writing a program to generate random pieces of rebol code, (rebol gibberish,) and post them automatically to this list for appraisal. The name for this program is RRGib. Here is the core of its functionality: forever [print random/only first system/words] The ultimate aim is to create self- sustaining rebol programs - rebol life. I've got it to the stage where the program can almost generate itself, possibly. Just a few more words and it's done. But I need your help in this effort. All I need you to do is to point out for me where all the mistakes are, so that any given random program is more likely to work. If anyone is interested in joining my project, please email me publically. Future programs that might be generated by this process include: - a full-featured paint-program - psycho-analyst AI-bot thing - random gibberish runtime debugger (randometer) - and many more possibilities! Hope to hear from possibly most of you all soon! Regards, Anton.

 [2/18] from: carl:cybercraft at: 31-May-2001 18:18


On 31-May-01, Anton wrote:
> Hello, > I am writing a program to generate
<<quoted lines omitted: 10>>
> possibly. > Just a few more words and it's done.
One of the words you obviously need is 'unrandom... USAGE: UNRANDOM value /different DESCRIPTION: Returns a non-random value of the same datatype. UNRANDOM is an inaction value. ARGUMENTS: value -- The non-random value you want returned. (Type: any - of course) REFINEMENTS: /different -- Returns a non-random value of a different datatype.
> But I need your help in this effort.
See above.
> All I need you to do is to point out > for me where all the mistakes are, so
<<quoted lines omitted: 10>>
> Hope to hear from possibly most of you > all soon!
Count me in, Anton!
> Regards, > Anton.
-- Carl Read [carl--cybercraft--co--nz]

 [3/18] from: brett:codeconscious at: 31-May-2001 17:15


> I am writing a program to generate > random pieces of rebol code, (rebol > gibberish,) and post them automatically > to this list for appraisal.
Hey that's my job!
> The name for this program is RRGib. > > Here is the core of its functionality: > > forever [print random/only first system/words] >
Damn. I'm about to be replaced by one line of Rebol code. How depressing. :-/ Brett.

 [4/18] from: carl:rebol at: 31-May-2001 1:05


You will want to be careful, because this is the same method that Microsoft used to create the Windows operating system. Except that they used thousands of programmers randomly banging on keyboards for several years... but then poof, there it was. Perhaps with your program REBOL will produce Windows 2002 in much less time. Don't tell Bill. -Carl

 [5/18] from: steve:shireman:semaxwireless at: 31-May-2001 5:06


Anton wrote:
> Hello, > I am writing a program to generate
<<quoted lines omitted: 15>>
> that any given random program is more > likely to work.
If you design automatic evaluation (selection) of your random scripts, mate the better random programs with each other, then you have essentially done the Genetic Programming method (Koza 1990) This could be much faster than human evalution of the random scripts on this list. ;-) Rebol is uniquely suited for Genetic Programming, because it almost never crashes bad with wayward code, (and can be tested with try, anyway) Essentially Genetic Programming is a search for a solution, and uses inductive logic to reach a final design, which runs counter-intuitive to most programmer types who are trained to try to use deductive logic. Anywho, thanks for the one liner. Mutate fast, Steve Shireman Your brain may no longer be the boss Firesign Theatre

 [6/18] from: gjones05:mail:orion at: 31-May-2001 6:24


Hi, Anton, Like Brett, I'm sorry to see that my functionality to the list is being supplanted by code! I believe that a few more features are needed in order to perfect the program in order to allow maximum list un-productivity! The program output suffers from poor formatting, but, hey, this ain't Python! ;;;;;;;;;;;;;;;;;;;;;; random-system-word: does [random/only first system/words] new-random-word: does [ new-word: make string! 20 loop add random 2 1 [ append new-word copy join random-system-word "-" ] append remove back tail new-word ": " ; return new-word ] new-random-line: does [ new-line: make string! 30 loop add random 4 2 [ append new-line copy join random-system-word " " ] return new-line ] new-random-program: func [len] [ new-program: make string! 300 loop len [ append new-program do random/only [new-random-word new-random-line newline] ] ] ;example usage ;print new-random-program 100 random-help-request: does [ random/only [ {Hi, List,^/I've been struggling with this program for hours, and I can't figure out where I'm going wrong. All feedback is welcome.^/^/} {Hi, List,^/Here is a nifty program that is bound to be a killer app. Check it out, but copyrights do apply! There are still one or two bugs to work out.^/^/} {Hi, List,^/Should I submit this to UHURU/pro, or has it not been sufficiently refactored?^/^/} ] ] ;example submission minus the 'send email command ;-) print [random-help-request new-random-program 100] ;;;;;;;;;;;;;;;;;;;;;; Now, we just need a program that will automatically respond to the list and we won't even need to attend to the list at all!!! --Scott Jones

 [7/18] from: joel:neely:fedex at: 31-May-2001 6:57


hi , [gjones05--mail--orion--org] , {have a nice} either 12:00 < now/time ["afternoon"] ["morning"] {there's something relevant to your question in} random/only [ "the mail list archive" "the Docs folder at the rebol.com RebSite" "REBOL: The Official Guide" "REBOL for Dummies" "Jeff's Zen of REBOL book, to be publised in 2005" "an REP on Ladislav's web site" "my hard drive" ] [gjones05--mail--orion--org] wrote:
{[snip]}
> Now, we just need a program that > will automatically respond to the > list and we won't even need to > attend to the list at all!!! >
random/only [ ";-)" "Hope this helps!" "Every newbie asks about that one" "Use THROW-ON-ERROR" "Put COPY in front of the series" "That's in the next release" "That's a known bug" "Carl could do it with fewer lines" "Volker could do it with more lines" "Joel would write 257 pages about it" ] {Generated on behalf of} {-jn-} {by rebolmailbot} self/ver ------------------------------------------------------------ Human beings: slow, messy, high maintenance ... Pick any two! rebolmailbot 0.2.19

 [8/18] from: gjones05:mail:orion at: 31-May-2001 7:27


From: "Joel Neely"
> hi , [gjones05--mail--orion--org] , > {have a nice}
<<quoted lines omitted: 31>>
> ] > {Generated on behalf of} {-jn-} {by rebolmailbot} self/ver
Hi, Joel, again, you made me laugh out loud! --Scott Jones

 [9/18] from: gjones05:mail:orion at: 31-May-2001 7:32


From: "Joel Neely" ...
> "Joel would write 257 pages about it"
... Should be: Joel would refactor it, and then write 257 pages about it ;-) --Scott Jones

 [10/18] from: allenk:powerup:au at: 31-May-2001 23:16


Anton, I could try it with RNILL maybe. Maybe he could write scripts after reading the demo library. I did the similar thing with RNILL and Shakespeares sonnets. But I'd make sure security level was strict before I unleashed any of its new creations. :-) Cheers, Allen K

 [11/18] from: ryanc:iesco-dms at: 31-May-2001 8:20


As Steve just mentioned, mating would be great! If we follow our own makeup, this would lead to a system that contained Dominant and Recessive qualities. This would be like having a database with two sets of ranked words (dominant, recessive). The interpeted version would be a combination of the two. During mating a single ranked word set is created by the father and passed on to the mother program who combines it with a single ranked word set created by herself. An interpetable program is then created by her and spawned. Pretty basic DNA stuff really. Of course the programs, mother and father, should be very similiar, and maybe even should a bit more fualt tolerent dialect that relates better to dominant and recessive qualities, and also oriented toward the type of being that your creating. Not necessary though. To get things started you just write an Adam and Eve, the rest takes care of itself. I suppose if you have the patience, you can wait for them to appear out of RRGib. I do have to say, that one has to be EXTREMELY careful to contain their creations, as life loves to spread. There is a fine line (if any) between these good natured creations and nasty computer viruses. --Ryan Steve Shireman wrote:
> Anton wrote: > >
<<quoted lines omitted: 41>>
> [rebol-request--rebol--com] with "unsubscribe" in the > subject, without the quotes.
-- Ryan Cole Programmer Analyst www.iesco-dms.com 707-468-5400 I am enough of an artist to draw freely upon my imagination. Imagination is more important than knowledge. Knowledge is limited. Imagination encircles the world. -Einstein

 [12/18] from: depotcity:telus at: 31-May-2001 9:53


I think the output would suffer from the second law of thermodynamics and entropy. Although there are millions of potential good programmes that could/would be generated, there are an INFINITE number of useless ones. This is entropy. What you need is a programme that can decide what is good, and what is infinite. T Brownell

 [13/18] from: jseq:mediaone at: 31-May-2001 13:34


I think Anton could benefit from some prior art research into Genetic Algorithms I believe what Terry is describing is a fitness test, in GA-speak. There is a thriving community of people attempting to do what you've described, and you'll learn alot by going to some GA websites. (Wish I could recommend some, but my interest in the field is dated) Good luck - GA is cool stuff. js http://www.pobox.com/~johnseq 5/31/01 12:53:26 PM, "Terry Brownell" <[depotcity--telus--net]> wrote:

 [14/18] from: lmecir:mbox:vol:cz at: 31-May-2001 21:18


Hi, ----- Original Message ----- From: Joel Neely <[joel--neely--fedex--com]> To: <[rebol-list--rebol--com]> Sent: Thursday, May 31, 2001 1:57 PM Subject: [REBOL] Re: RRGib - Random Rebol Gibberish
> hi , [gjones05--mail--orion--org] , > {have a nice}
<<quoted lines omitted: 30>>
> "Joel would write 257 pages about it" > ]
Joel should elaborate on this and submit it to UHURU as the first useful script! I suggest to add one more point : "The predictability of your script will improve significantly, if it doesn't redefine system words or mezzanines while using them!"

 [15/18] from: jeff::rebol::net at: 31-May-2001 14:15


Howdy, Joel:
> random/only [ > ";-)" "Hope this helps!" "Every newbie asks about that
<<quoted lines omitted: 5>>
> ] > {Generated on behalf of} {-jn-} {by rebolmailbot} self/ver
Always felt list should have a bot or two looming around. One trick that the pundit bot in the old rebol.org chat room used is to take a bunch of quotes from someone, break the sentences into separate clauses and index the words (throwing out the first few thousand most common english words). Then when you say something it collects "relevant clauses" and strings together a certain number of clauses to produce a "sentence" ha ha. Seems like a mailing list bot could use similar approach, sort of like: [{Hello, are you asking:} x "?" newline "If so, then:" y ] I have a bot-generator script that basically does that process, if you feed it a bunch of someone's quotes. I made a ben franklin bot among others. Might be fun to create a Carl-bot and unleash it on the list.. just mumbling aloud.. (-: -jeff

 [16/18] from: lmecir:mbox:vol:cz at: 31-May-2001 23:23


> I think the output would suffer from the second law of thermodynamics and > entropy. Although there are millions of potential good programmes that > could/would be generated, there are an INFINITE number of useless ones.
This
> is entropy. > What you need is a programme that can decide what is good, and what is > infinite. > > T Brownell >
THAT would be a program that would surely deserve a place in UHURU! (-:

 [17/18] from: agem:crosswinds at: 31-May-2001 22:57


>>>>>>>>>>>>>>>>>> Ursprüngliche Nachricht <<<<<<<<<<<<<<<<<<
Am 31.05.01, 17:53:26, schrieb "Terry Brownell" <[depotcity--telus--net]> zum Thema [REBOL] Re: RRGib - Random Rebol Gibberish:
> I think the output would suffer from the second law of thermodynamics
and
> entropy. Although there are millions of potential good programmes
that
> could/would be generated, there are an INFINITE number of useless
ones. This
> is entropy. > What you need is a programme that can decide what is good, and what is > infinite.
That means there are an infinite number of good programms? Any percentage of infinite is by itself infinite, regardless of how small. Infinite * 1e-10000000 are again infinite.. hey, lots of programms! (pst! tell its all by you ;-)
> T Brownell
-Volker

 [18/18] from: depotcity:telus at: 31-May-2001 21:56


Did I say infinite? I meant to say REALLY REALLY LOTS! TBrownell ----- Original Message ----- From: "Volker Nitsch" <[agem--crosswinds--net]> To: <[rebol-list--rebol--com]> Sent: Thursday, May 31, 2001 3:57 PM Subject: [REBOL] Re: RRGib - Random Rebol Gibberish
>>>>>>>>>>>>>>>>>> Ursprüngliche Nachricht <<<<<<<<<<<<<<<<<<
Am 31.05.01, 17:53:26, schrieb "Terry Brownell" <[depotcity--telus--net]> zum Thema [REBOL] Re: RRGib - Random Rebol Gibberish:
> I think the output would suffer from the second law of thermodynamics
and
> entropy. Although there are millions of potential good programmes
that
> could/would be generated, there are an INFINITE number of useless
ones. This
> is entropy. > What you need is a programme that can decide what is good, and what is > infinite.
That means there are an infinite number of good programms? Any percentage of infinite is by itself infinite, regardless of how small. Infinite * 1e-10000000 are again infinite.. hey, lots of programms! (pst! tell its all by you ;-)
> T Brownell
-Volker

Notes
  • Quoted lines have been omitted from some messages.
    View the message alone to see the lines that have been omitted