• Provide Wall-ism on demand.
• Provide Haiku on demand
• Answer any Perl FAQ, PerlMonks FAQ, or near-FAQ with a reference
• Not speak up if there is nothing useful to say (random haiku or sayings every now and then when quiet is fine).
• As a fun option, it can perhaps be told to "chatter" or "converse" and it can enter Eliza-like (or whatever), and will cease when told to "Shut-Up", "Be Quiet" or "Be Silent"

• They should use the XML interface vroom supplied at
  • http://perlmonks.org/index.pl?node=chatterbox+xml+ticker displays all current public messages
  • http://perlmonks.org/index.pl?node=private+messages+xml+ticker
  • http://perlmonks.org/index.pl?node=other+users+xml+ticker
• They should not be in the chatterbox for most testing (they can get info from the above slots, and "speak" to STDOUT for most testing.
• If and when "real" testing is necessary, the bots should either be usernames with "bot" at end, or speak "(bot)" in each post.

How will entries be submitted?
On a specified date, all Monkbots will be run for a 24 hour period. THe next day the bots code can be posted, and judged over the next 24 hours.

How will the entries be scored?
Judging should be based on achievement of listed goals, how pleasant the bot is to be around (everyone hates a loudmouth or inaccurate bot), and the elegance of the code. (and of course, adherance to good manners during testing)

Who will score the entries?
Judging will be done by any and all Perlmonks who wish to vote.

When/will submissions be viewable by everyone?
Following the Demonstration date.

I like this idea a lot, I have one small criticism. Not all monks have access to a machine connected 24/7 to the internet, of those that do, how many are allowed to run background processes for 24 hours straight.

Alright I admit it, this is a post about me, I don't have access to these. :-(

Nuance

I think helpful fellow monks with constant connections might be glad to host another's bot. In particular, if you were to write a bot, Nuance, I'd be glad to leave it running for you on either of my machines for the duration of the contest. I don't know how many monks I could do this for but probably lots of help could be enlisted if there was need.

My concern for the contest is the number of submissions. Maybe an addendum (sp?) could be made to only accept 16 (or some reasonable 2^n || 2^n-1 threshold) entries so that the chatterbox isn't oversaturated and worthlessly cluttered/unusable for the duration of the contest. If somebody asks some FAQ question which clearly is the point of the contest, all the bots are gonna respond and that can be a lot if lots of bots are spouting. Sounds like fun though. TTFN & Shalom.

-PipTigger

A very good point, but one to which I have no good answer. Selecting any period of time hurts people on some section of the globe. Certainly running for all 24 hours wouldn't be REQUIRED, but it'd hurt your chances not to be seen by everyone. Anyone want to donate a server connection?

I wanted to add something...we could also pass around a list of variables that have been creatived...good time to use descriptive names :)

How will entries be submitted?
They will be emailed to a specific email address, with a _start_ and an _end_ block around the script.

How will the entries be scored?
They will be played against eachother in a script that plays the subroutines against eachother some number of times (let's say 10? maybe another). The winning script at the end of each match, gets 1 point. The full winner wil have the most points, in the case of a tie, the tied scripts will play in a best 5 out of 9 play off.

<STRON

heh. cut me off a little. to continue:

Who will score?
I dunno. someone with linux installed, and a relatively good machine, preferably vroom, or someother high up. he doesn't have to really judge, just start a script to compete the subroutines.

when will submissions be viewed
the contest will last some predetermined time, such as 1 week, 2, or a month(it may be difficult). as soon as it is done, no more entries willbe accepted, and the entrants will be posted. the scoring will take place the next day or over the course of a few days (depending on the amount of scripts, and the length each takes).

the winning subroutine gets mega XP!!!!!

I like this idea. Even more than my own. I feel that it would be good practice for people since they could shorten the code in many different ways. I also like it because the scoring system is not based on the personal opinion of the moderators, rather it is based on the fact of how many lines you made it to be. Now, if you could elaborate on your idea, like how the code will be managed and when it will occur, I think we have a winner, at least in my book. Heh.

-- zdog (Zenon Zabinski)
   Go Bells!!

Well then... I guess the best way to tackle this is, someone who enters the contest gets a script to work on -- I'm not sure how we should do this, to each a different script or all the same, ideas? -- and when done, mails the new code some authorative figure (whom we shall name The CareTaker -- The Manager sounds too corporate ;). I guess this could be Vroom. When reviewed, the code should be posted up on a node for all to comment on.

I like that idea. It would be a good training for newbies finding out shortcuts that are typical Perl. Of course the script shouldn't be shortened just by removing all unnessecary white-space :-)

H: Either by following up to a contest node, or via email to a maintainer.

H: Elegance: 30%, usefulness: 70%

W: A team of Monks with XP over $somenum selected by PM staff

W: At time of submission (followup) or at time of final decision (email)

How will entries be submitted?
Entries should be submitted by email because some may be quite long.

How will entries be scored?
Scoring will be based on the depth of the tutorial and how easy it is to understand.

Who will score the entries?
All registered users will be able to vote on which tutorial is the best, in their opinion.

When will submissions be viewable by everyone?
There will be a deadline for submitting entries. After this deadline, those received will be viewable for voting.

How will entries be submitted?
Entries can all be emailed to one person who will not participate (I'll volunteer if we need one, I can't obfuscate well). That person will post each submission for judging when entry time has expired (Vroom will need to ensure actual author's get due credit and XP) Each entry should be accompanied by the authors PM username, and three sample runs, one of which must be without command-line options. Alternatively, Vroom can supply us with a private section to post code in that will become public at a certain time.

How will the entries be scored?
Community voting after posting, with a request (but not demand) that each person vote no more than 3 times. Scores should be based on how obfuscated the final code is, and how elegant the original script is. Voting period lasts for 1 week from initial posting. Votes can be acquired after that, but do not count for the "contest".

Who will score the entries?
Someone (who can see all the reps) merely needs to note the top-ranked persons. Vroom can award XP and titles to those people as appropriate.

When will submissions be viewable by everyone?
At beginning of voting date.
If you want hard dates, I propose that all entries be submitted by 12 noon GMT -0000 June 19. Voting begins the next day and lasts 1 week.

• Elegance/Flair/Obfuscation
• Speed
• Accuracy

A weekly or biweekly game, to be played by the submitted programs. These games could be suggested and voted upon by any interested party.

How will entries be submitted?

Entries can be mailed to a single person, or an entry handling script, or can be posted to a non-publically viewable node.

How will the entries be scored?

Entrants get a number of XP based on the number of opponents defeated, with bonus XP for making it to the semi-finals, finals, or winning.

Who will score the entries?

Since this is a fairly straightforward method of calculating a winner, it could be automated with a game-running script.

When/will submissions be viewable by everyone?

The submissions will be viewable by everyone the day after the deadline for entry, allowing competitors to discuss various strategies, and predict the winners before the resolution of the game.

This sounds like a great constest!

Forgive the non-Q&A structure of this post. All of the "required five" are in here, I promise! :)

The contest: solving a particular problem using perl. The problems should be fairly simple, as we are focusing on more of a "perl" rather than a "programming" skill contest (like the POTM). Each week (hopefully, actual time may end up being a month, knowing how everyone else (even vroom) has other things to do in life) a new contest will be announced by the Contest Master, who is the monk who won the previous contest. It is their responsibility to run the contest. They also have to come up with the contest, but can solicit (private) ideas from others. I've got dozens in mind! :)

(Whew!) About the contest: as I mentioned in a previous node, there will be 4 different categories:

• Most Elegant (easiest to read, nicely formatted, "cleanest" use of perl, etc.)
• Most Obfuscated.
• Most Terse (i.e. fewest characters/lines)
• Most Efficient (which one does the task at hand the quickest)

The first two are subjective, the bottom two are objective. The CM shall be responsible for measuring the last two (wc and Benchmark, natch), and the first two shall be voted upon by the Monk masses. How? Well, since obfuscation and elegance should be extremely easy to distinguish, there is no need to discriminate - just have each person get two votes, which they use to mark what they think are the Most Elegant and the Most Obfuscated. Or they could use both votes for two elegance answers, etc. I don't think that matters too much.

Once the voting period is over, the CM tallies the votes (and she/he should be the *only* one allowed to see the counts, unlike today's voting system). The highest number of votes in each category wins. Tiebreakers of the first three categories are determined by a efficiency runoff, and Efficiency ties are determined by terseness. Final authority on all ties shall lie with the CM.

Depending on the number of entries, first, second, and third place will be awarded in each category. Entries that place in more than one category (i.e. elegance *and* efficiency) gain bonus XP for their creator. The overall winner is determined by adding up the results of all their entries as follows:

• First place: 3 foobars
• Second place: 2 foobars
• Third place: 1 foobar
• Same program placing in 2 or more categories: 2 foobars

The one with the most foobars wins. Ties are solved by the method explained above (efficiency or terseness)

Once the final judging is done, the page is revealed, with all entries shown, as well as the top three marked. Showing total votes is probably not a good idea: why embarrass anyone who gets 1 vote compared to 120 for the third place? The page becomes a "normal" page at this point, as people reply (and discuss, and argue) about the entries.

Posting answers elsewhere on the site results in a -100 XP penalty and an immediate disqualification. AM's get their IP banned from the site (how else can you punish them?) Perhaps have the contest only visible to logged in users? Hard to say if these measures will be needed until we try, but I'm being thorough.

The only question is, should each person be allowed four entries? Two? Only one? Each will have to be a separate post, assuming that all of this will be done by using the existing perlmonks engine. This could be written as an external cgi script somewhere, but then there is no access to the perlmonk user data, so perlmonks code it is! :)

Question/comments welcome.

Those of us who thinks best during day and live on the other side of the globe, might suffer from a contest starting US time. Maybe we could push a button and have the problem mailed to us. The time taken will then be calculated from the button-push to the posting time.
... and monks must promise not to forward the mail to someone else. But that would be quite unmonkish, wouldn't it ...

/brother t0mas

I completely understand your concern. My thought was that multiple contest could take place all at different times of the day, for the different users.
I do like your idea about having the push button that calculates time, but then when people submit their solution they will have to be emailed or something.
We can't have the other monks having a sneak peak at the code before they hit the button.
-Rage

Back With Another of Those Block Rockin' Beats

• Make a new feature or improve an existing part of the monastery. (vroom, would you allow certain monks to have a look at the code behind certain nodes?)
• Entries should be posted publicly, or sent directly to vroom if he wants to keep certain things private for the time being.
• Entries will be scored on usefulness, code completeness and elegance, and value to fellow monks.
• Voting will be open to anyone.
• Submissions will be available -- at least by description, if not actual code -- when the code is completed.
.

How will entries be submitted and when will they be viewable to everyone?
A person will volunteer to run the contest. The scripts will be e-mailed to that person by a certain date. The next day, the person will post all the scripts on a specified page.

How will the entries be scored and by whom?
On the page where the scripts are posted, Monks that did not participate(otherwise people would only vote for their own) will vote on the scripts. They will vote for a first place, a second place, and a third place. The one voted first will get 3 votes, second - 2 votes, and third - one vote. The votes should be based on program correctness, speed, how well the problem is solved, extras that elaborate on the problem, etc.

-- zdog (Zenon Zabinski)
   Go Bells!!

• Come up with the best scheme for categorizing all of the information in the monastery. How can it be presented to make sense to new monks and still allow existing monks access to what they need?
• Post entries as replies to the contest.
• Entries will be judged on the basis of workabililty and completeness.
• High-level monks (the top five?) will score the entries.
• Submissions will be viewable as soon as they are posted. No secrets here.

• What is the contest?
  Solve an AI problem with Perl as fast as possible. Also as efficiently as possible.
• How will entries be submitted?
  Entries submitted to each and everyones homenodes.
• How will the entries be scored?
  Scoring Performance and Functionality(like errorhandling).
• Who will score the entries?
  A program.
• When/will submissions be viewable by everyone?
  After the contest.

I'd definately be more interested in competing with code :-) All the posts along these lines seem to have been a bit vague, so here are some more specific ideas:

W: Write a perl script to solve a simple 10x10 maze. The maze is generated by a perl module, and the script is only allowed to call functions defined there and the standard perl functions. Access to the data structures from the maze module are strictly forbidden! (The script may only be allowed to 'look' north, south, east or west, to see if there's a wall or not, and 'move' in a direction). The script should travel from an entrance to the center of the maze and out to an exit (different from the entrance)
H: Submitted to homenodes
S: The script that traverses a random selection of mazes in the least number of steps wins
W: The maze module
W: After the contest

And some more ideas along a similar vein:

• Solve a 5x5 sliding block puzzle, only using the four squares around the empty square. (Once again, scored by total moves over a random selection of boards)
• Play the game 'black box' against another perl script. This gives you a number of atoms in a box. The script has to work out where they are by shooting rays into the box. If a ray hits an atom dead on, the ray is absorbed. If a ray passes one square away from an atom, it is deflected 90 degrees away from the atom (and again, if it meets a new atom on its travels). Each script can see the results of the other, and the winner is the one that guesses the correct position of the atoms first, so the competition can have a knockout format ;-)

Hmm, dried up, I have. There are plenty more ideas in this area, though. Knockout competitions in particular would be fun. Prizes could be XP, and maybe an 'awards cabinet' section on the home node (because then you could add runners-up and booby prizes such as 'most obfuscated code', 'shortest correct solution', 'most blatent cheat', 'most compiler warnings' and so forth :-)

Andrew.

How will entries be submitted?
All code must be posted for peer review as a reply to the contest node. ONLY contest submissions will be allowed as first level-replies.

How will the entries be scored?
1 point for every line of executable code. Comments and whitespace won't help you! +10 bonus points for style (formatting of code/output, interesting routines, etc.)

How will the entries be scored?
Code length, of course. "Style" bonus points determined by peer review, based on either Rep. or replies.

Who will score the entries?
The community, or one individual selected by vroom.

When/will submissions be viewable by everyone?
Of course, the submissions will be viewable from start to finish.

- Ozymandias

Maybe I am missing something, but it seems to me that one could submit any long perl script, provided that it also printed Hello World a few times... I think there need to be more rules...I mean, you could write a million line program that had one variable that you manually added 1 (one) to at each line, then printed "hello world". Seems weird. --Ivory

Mmmm. Good point. Perhaps it should be something like "each line must do something towards the goal of printing the output; but that would get really difficult. Scoring would have to become a matter of popular vote, and that has too much chance of simply becoming a popularity contest.

The reason I like the idea of a "longest script" contest is that it's a challenge. We're all very used to squeezing code down to the minimum; it shows in our penchant for one-liners, our preference for haiku. Every contest I see, one of the criteria is "shortest code" or "most elegent code" - which amounts to the same thing.

That's normally a good thing. It helps us write better code. I just thought that, for once, vying to create the LONGEST script would be a good change of pace.

- Ozymandias

Ya, I agree with ivory. The truly good Perl programmers work to make their code as short, as readable, and as fast as possible. This idea of the longest Perl program to print Hello World needs to be edited a bit.

-- zdog (Zenon Zabinski)
   Go Bells!!

I had an odd post on Meditations when it struck me (and another Monk suggested) that this would make a great contest.

if ($newt =~ /running from alien queen/i) {
    RIPLEY: {
        use HEAVYLOADER;
        kill $alienqueen || undef $newt;
        open (AIRLOCK, ">>hacknedstoryidea.dat") || last RIPLEY;
    }
}
[download]

Submission: My first thought is to have everyone go to Amazon.com and enter their submission as a user review to a book about a Micro$oft product and then post the link here. Then I realized it would be easier to just post submissions here to a contest node.

Scoring:

• Do you like the story? +/- 10 points by each person voting.
• Do you understand the story? +/- 10 points by each person voting.
• Minus 5 percent of grand total for each syntax error.
• Plus 10 percent of grand total if the program compiles.
• Plus 20 percent of grand total if it actually does something useful (We might want to put a minimum program size for this one).

When will submissions be viewable: When a submission deadline passes.

Cant really see this one taking off.... Its too complicated, and most people cant be bothered to search for palindromic regex's.

How do you propose to handle the issue of people that are level 5 and above can insert pictures, and the rest can't? And are there any other formatting rules that are level sensitive?

I was thinking about that. But I have seen some cool home nodes that didn't have pictures (admitedly not many though) Maybe we could run two parallel contests for those people below level five and those above.

Is it me, or are missing the results of this quest? (Of course, I think my idea was not voted high enough, but everyone thinks that of theirs, I'm sure :)