Loading ...
Sorry, an error occurred while loading the content.

Text-only exquisite corpse / eat poop you cat

Expand Messages
  • Sai Emrys
    First, if you don t know the game: http://www.alexandrafinger.com/epyc/ tl;dr: Party game. First player secretly writes a sentence. Next player gets the paper,
    Message 1 of 18 , Feb 27, 2010
    • 0 Attachment
      First, if you don't know the game: http://www.alexandrafinger.com/epyc/

      tl;dr: Party game. First player secretly writes a sentence. Next
      player gets the paper, sees that sentence, secretly draws a picture
      thereof. Next player gets the paper, sees ONLY the picture, writes a
      caption thereof. Repeat - each time you only see the previous move -
      until returning to the first player, at which point you read out the
      results dramatically to much laughter.

      It's fun and silly and has the ability to diverge quite quickly & dramatically.

      For speed, this is usually done in parallel, with one paper per person
      participating (maybe -1 to ensure it's an odd total and you get
      textual beginnings & ends). This also lets everyone present the
      results of a set at the end of each round.


      My question: Can a game with the essence of EC/EPYC be played entirely
      with text?

      The game has some features that I think are critical to its enjoyment:
      * lossy and therefore ambiguous output, especially text -> picture
      (not as much during picture -> text)
      * openness to interpretation
      * fully player-generated content (vs. e.g. madlibs)
      * oscillating turns that provide a somewhat different experience each turn

      Naïvely, text alone would be bad; it'd degenerate into e.g. mere
      synonym games or the like.

      Textual telephone probably requires translation to generate the noise,
      which is fun for what it is (eg Conlang Relay), but is a pretty heavy
      constraint on who can play.


      One thing that Alex came up with is to have player 1 write a 6
      sentence story (each sentence approx. equal length). The next player
      gets this story with the odd sentences hidden, and fills them in. The
      next gets only those new sentences and fills in the evens, etc. I
      think this is a pretty good idea which will have all of the needed
      properties.

      What other methods would work?

      - Sai
    • David Peterson
      ... My friends and I played a game that only involved drawing, where a paper was folded into sections, and each person got one section. For every participant
      Message 2 of 18 , Feb 27, 2010
      • 0 Attachment
        On Feb 27, 2010, at 4◊18 AM, Sai Emrys wrote:

        > <big snip>
        >
        > What other methods would work?


        My friends and I played a game that only involved drawing,
        where a paper was folded into sections, and each person got
        one section. For every participant but the first, what they
        receive is a blank section with a few trailing lines. These are the
        connecting points to the previous picture. When the picture is
        done and unfolded, you get a picture that connects, but is totally
        random.

        I tried to take this idea and apply it to writing. I created an
        accordion book, and played a similar game with my wife and
        her siblings. I used the first page to write the first part of a
        story. The next participant read my page and continued the
        story. The participant after that, though, only was able to read
        the preceding page, and so on. By the end, my cantankerous
        badger named Spitzy had become a space-faring (sentient)
        haggis.

        So, based on this, I had two ideas:

        (1) Each person writes one sentence of a continuous story. The
        only sentence they're allowed to see is the previous sentence.
        By the end you have a long story that connects but probably
        makes little sense (much like the picture).

        (2) With conlangs, I came up with a crazy idea that might not
        work at all. This is how it goes:

        Person A: Writes an English sentence.
        Person B: Translates the English sentence into their conlang.
        Person C: *WITHOUT* a lexicon or grammar, C translates B's
        sentence into English.
        Person D: Translates the English sentence into their conlang.

        And so on. This means that there will be two very different
        types of turns. How this will differ from a relay, though, is
        that the decoder will simply have to use their wits to guess
        what the sentence is--perhaps creating something wholecloth
        which, nevertheless, could plausibly be associated with the
        sentence they receive (e.g. they could prove that if x were a
        tense suffix, and y were a noun, etc.).

        Another possibility is giving the decoder a set of English words
        that occur in the sentence without correspondence. So, for
        example:

        Sentence:
        A fulele ei i hava eu kavi i mala oi'i.

        Words (in alpha order):

        big
        clam
        eat
        for
        I
        mother
        my
        want

        There could be some constraints on this. For example, let's say
        you can only use the citation form (that would knock out "my"),
        and you couldn't say how many times a word is used. Now, you
        can probably figure out what the sentence is based solely on the
        words used, but if you imagine that the game has gone on for
        awhile, the sentence may make absolutely no sense by that point,
        so the words will be no help.

        Anyway, those are my ideas. Whatever you come up with, I want
        to play!

        -David
        *******************************************************************
        "A male love inevivi i'ala'i oku i ue pokulu'ume o heki a."
        "No eternal reward will forgive us now for wasting the dawn."

        -Jim Morrison

        http://dedalvs.com/

        LCS Member Since 2007
        http://conlang.org/
      • Claude Almansi
        A variant I played with students of French as a foreign language, divided in groups of 5-6: - participants get a longish slip of paper, then - write a name -
        Message 3 of 18 , Feb 27, 2010
        • 0 Attachment
          A variant I played with students of French as a foreign language,
          divided in groups of 5-6:

          - participants get a longish slip of paper, then
          - write a name
          - pass on their slip of paper to person on the right
          - write a transitive verb, then pass on etc.
          - write a noun, then pass on etc.
          - write a location, then pass on etc.
          - write a time indication
          - in turn, participants read the results aloud

          When you play this with pre-teens and teenagers, it is advisable to
          have a rule excluding living people and/or vulgarity: otherwise the
          results become offensive and boringly repetitive.

          Best

          Claude


          On Sat, Feb 27, 2010 at 1:18 PM, Sai Emrys <sai@...> wrote:
          > First, if you don't know the game: http://www.alexandrafinger.com/epyc/
          >
          > tl;dr: Party game. First player secretly writes a sentence. Next
          > player gets the paper, sees that sentence, secretly draws a picture
          > thereof. Next player gets the paper, sees ONLY the picture, writes a
          > caption thereof. Repeat - each time you only see the previous move -
          > until returning to the first player, at which point you read out the
          > results dramatically to much laughter.
          >
          > It's fun and silly and has the ability to diverge quite quickly & dramatically.
          >
          > For speed, this is usually done in parallel, with one paper per person
          > participating (maybe -1 to ensure it's an odd total and you get
          > textual beginnings & ends). This also lets everyone present the
          > results of a set at the end of each round.
          >
          >
          > My question: Can a game with the essence of EC/EPYC be played entirely
          > with text?
          >
          > The game has some features that I think are critical to its enjoyment:
          > * lossy and therefore ambiguous output, especially text -> picture
          > (not as much during picture -> text)
          > * openness to interpretation
          > * fully player-generated content (vs. e.g. madlibs)
          > * oscillating turns that provide a somewhat different experience each turn
          >
          > Naïvely, text alone would be bad; it'd degenerate into e.g. mere
          > synonym games or the like.
          >
          > Textual telephone probably requires translation to generate the noise,
          > which is fun for what it is (eg Conlang Relay), but is a pretty heavy
          > constraint on who can play.
          >
          >
          > One thing that Alex came up with is to have player 1 write a 6
          > sentence story (each sentence approx. equal length). The next player
          > gets this story with the odd sentences hidden, and fills them in. The
          > next gets only those new sentences and fills in the evens, etc. I
          > think this is a pretty good idea which will have all of the needed
          > properties.
          >
          > What other methods would work?
          >
          > - Sai
          >



          --
          The Education Technology and Change journal
          http://etcjournal.wordpress.com/ now tweets its posts and comments in
          real time at http://twitter.com/etcjournal .
        • John Vertical
          ... A comic strip example of basically this concept: http://www.tailsteak.com/archive.php?num=366 John Vertical
          Message 4 of 18 , Feb 27, 2010
          • 0 Attachment
            >One thing that Alex came up with is to have player 1 write a 6
            >sentence story (each sentence approx. equal length). The next player
            >gets this story with the odd sentences hidden, and fills them in. The
            >next gets only those new sentences and fills in the evens, etc. I
            >think this is a pretty good idea which will have all of the needed
            >properties.

            >- Sai

            A comic strip example of basically this concept:
            http://www.tailsteak.com/archive.php?num=366

            John Vertical
          • Sai Emrys
            ... That s the standard version of http://en.wikipedia.org/wiki/Exquisite_corpse ... This is somewhat similar to the Cheddar Gorge / collaborative storytelling
            Message 5 of 18 , Feb 27, 2010
            • 0 Attachment
              On Sat, Feb 27, 2010 at 4:43 AM, David Peterson <dedalvs@...> wrote:
              > My friends and I played a game that only involved drawing,

              That's the standard version of http://en.wikipedia.org/wiki/Exquisite_corpse

              > I used the first page to write the first part of a
              > story. The next participant read my page and continued the
              > story. The participant after that, though, only was able to read
              > the preceding page, and so on.

              This is somewhat similar to the Cheddar Gorge / collaborative
              storytelling type of ?game, though the hidden information makes it
              more interesting.

              It seems like it'd require a lot of reiteration of story elements. And
              it feels like a qualitatively heavier game than the 6-line-story one -
              something I think most people would balk at.

              > Person C: *WITHOUT* a lexicon or grammar, C translates B's
              > sentence into English.

              Wouldn't this be pure noise modulo [C's L1 family]-langs and C's
              knowledge of B's conlang?

              Especially without enough corpus to effectively *derive* a
              lexicon/grammar, which is essentially what they'd have to do...


              On Sat, Feb 27, 2010 at 6:26 AM, Claude Almansi
              <claude.almansi@...> wrote:
              > When you play this with pre-teens and teenagers, it is advisable to
              > have a rule excluding living people and/or vulgarity: otherwise the
              > results become offensive and boringly repetitive.

              Having just come from a game of EPYC, I can say that people appear to
              have a strong tendency to have penile pareidolia.

              I don't find it offensive, but I think that's rather a matter of culture.

              On Sat, Feb 27, 2010 at 6:43 AM, John Vertical <johnvertical@...> wrote:
              > A comic strip example of basically this concept:
              > http://www.tailsteak.com/archive.php?num=366

              I forgot about that one; that's another good method. Tailsteak is quite amusing.

              - Sai
            • Gary Shannon
              ... How about this idea? Person 1 writes a question. Person 2 reads the question and writes and answer. Person 3 sees only the answer and writes an appropriate
              Message 6 of 18 , Feb 27, 2010
              • 0 Attachment
                On Sat, Feb 27, 2010 at 4:18 AM, Sai Emrys <sai@...> wrote:

                >
                > My question: Can a game with the essence of EC/EPYC be played entirely
                > with text?
                >

                How about this idea?

                Person 1 writes a question.
                Person 2 reads the question and writes and answer.
                Person 3 sees only the answer and writes an appropriate question.
                Person 4 reads that question and writes and answer.
                ...etc...

                This could end either with a question or an answer since they are both
                text and should both provide opportunities for jocularity.

                Answer should not repeat the question but should not be one-word
                answers like "yes", or "blue". Questions should encourage short essay
                answers.

                --gary
              • John Lategan
                ... I played a similar version too, like so: subj; verb; adv; article+adj; obj; conjunction... or whatever the languages syntax is. I enjoyed it.
                Message 7 of 18 , Feb 27, 2010
                • 0 Attachment
                  > divided in groups of 5-6:
                  >
                  > - participants get a longish slip of paper, then
                  > - write a name
                  > - pass on their slip of paper to person on the right
                  > - write a transitive verb, then pass on etc.
                  > - write a noun, then pass on etc.
                  > - write a location, then pass on etc.
                  > - write a time indication
                  > - in turn, participants read the results aloud

                  I played a similar version too, like so: subj; verb; adv; article+adj;
                  obj; conjunction...

                  or whatever the languages' syntax is. I enjoyed it.
                • Jim Henry
                  ... I suspect something somewhere in between the two forms David mentioned would be ideal -- a whole page may be too much; a single sentence may be too little.
                  Message 8 of 18 , Feb 27, 2010
                  • 0 Attachment
                    On Sat, Feb 27, 2010 at 7:43 AM, David Peterson <dedalvs@...> wrote:
                    > I tried to take this idea and apply it to writing. I created an
                    > accordion book, and played a similar game with my wife and
                    > her siblings. I used the first page to write the first part of a
                    > story. The next participant read my page and continued the
                    > story. The participant after that, though, only was able to read
                    > the preceding page, and so on. By the end, my cantankerous

                    > So, based on this, I had two ideas:

                    > (1) Each person writes one sentence of a continuous story. The
                    > only sentence they're allowed to see is the previous sentence.
                    > By the end you have a long story that connects but probably
                    > makes little sense (much like the picture).


                    On Sat, Feb 27, 2010 at 9:52 AM, Sai Emrys <saizai@...> wrote:

                    > It seems like it'd require a lot of reiteration of story elements. And
                    > it feels like a qualitatively heavier game than the 6-line-story one -
                    > something I think most people would balk at.

                    I suspect something somewhere in between the two forms David mentioned
                    would be ideal -- a whole page may be too much; a single sentence may
                    be too little. You want a small enough overlap that people don't get
                    bored or distracted waiting for their turn, but large enough that the
                    story doesn't degenerate into total nonsense too quickly. Maybe two
                    or three sentences?

                    When we figure out which of these variations we want to try, I'd like
                    to participate in such a play-by-email game. I've been too busy for
                    the current translation relay or the round-robin conlang, but this
                    looks manageable and fun.

                    --
                    Jim Henry
                    http://www.pobox.com/~jimhenry/
                  • John Lategan
                    I also thought of doing an email one, but couldnt think how we would hide the prev entries that arent meant to be seen? unless its all done offline (not to
                    Message 9 of 18 , Feb 27, 2010
                    • 0 Attachment
                      I also thought of doing an email one, but couldnt think how we would
                      'hide' the prev entries that arent meant to be seen?

                      unless its all done offline (not to entire List); where one sends your
                      part on and to a non-participating "collector" who puts the final
                      story together.

                      ...or am I missing an obvious way in which it would have been done?
                      either way, count me in!

                      John

                      2010/2/27, Jim Henry <jimhenry1973@...>:
                      > On Sat, Feb 27, 2010 at 7:43 AM, David Peterson <dedalvs@...> wrote:
                      >> I tried to take this idea and apply it to writing. I created an
                      >> accordion book, and played a similar game with my wife and
                      >> her siblings. I used the first page to write the first part of a
                      >> story. The next participant read my page and continued the
                      >> story. The participant after that, though, only was able to read
                      >> the preceding page, and so on. By the end, my cantankerous
                      >
                      >> So, based on this, I had two ideas:
                      >
                      >> (1) Each person writes one sentence of a continuous story. The
                      >> only sentence they're allowed to see is the previous sentence.
                      >> By the end you have a long story that connects but probably
                      >> makes little sense (much like the picture).
                      >
                      >
                      > On Sat, Feb 27, 2010 at 9:52 AM, Sai Emrys <saizai@...> wrote:
                      >
                      >> It seems like it'd require a lot of reiteration of story elements. And
                      >> it feels like a qualitatively heavier game than the 6-line-story one -
                      >> something I think most people would balk at.
                      >
                      > I suspect something somewhere in between the two forms David mentioned
                      > would be ideal -- a whole page may be too much; a single sentence may
                      > be too little. You want a small enough overlap that people don't get
                      > bored or distracted waiting for their turn, but large enough that the
                      > story doesn't degenerate into total nonsense too quickly. Maybe two
                      > or three sentences?
                      >
                      > When we figure out which of these variations we want to try, I'd like
                      > to participate in such a play-by-email game. I've been too busy for
                      > the current translation relay or the round-robin conlang, but this
                      > looks manageable and fun.
                      >
                      > --
                      > Jim Henry
                      > http://www.pobox.com/~jimhenry/
                      >
                    • Claude Almansi
                      ... Maybe with a Google Docs online spreadsheet, which you can connect to a question form corresponding to the answer columns? The game organizer would e-mail
                      Message 10 of 18 , Feb 27, 2010
                      • 0 Attachment
                        On Sat, Feb 27, 2010 at 9:53 PM, John Lategan <jlateg@...> wrote:
                        > I also thought of doing an email one, but couldnt think how we would
                        > 'hide' the prev entries that arent meant to be seen?
                        >
                        > unless its all done offline (not to entire List); where one sends your
                        > part on and to a non-participating "collector" who puts the final
                        > story together.
                        >
                        > ...or am I missing an obvious way in which it would have been done?
                        > either way, count me in!
                        >
                        > John

                        Maybe with a Google Docs online spreadsheet, which you can connect to
                        a question form corresponding to the answer columns? The game
                        organizer would e-mail the forms in different order each time, and
                        keep the spreadsheet private until all the questions were answered.
                        Problem: s/he'd have to view the spreadsheet to know when someone has
                        answered before mailing the form to the next person.

                        Or maybe Harvard's H2O Rotisserie (1) could be tweaked so that people
                        wouldn't see the answer they are called to comment upon: it's free
                        software.

                        Best

                        Claude

                        (1) About the Rotisserie, see
                        <http://h2oproject.law.harvard.edu/rotisserie.html>. I participated in
                        their "Development and the Internet"
                        <http://h2o.law.harvard.edu/ViewProject.do?projectID=64> course back
                        in 2003: the "semi asynchronous" rotisserie part where we were
                        assigned someone else's answer to comment upon worked really well.
                        That was way before the Web 2.0 hype....
                      • Eric Christopherson
                        What is the etymology of EPYC? Also, I remember there being a webcomic that was dedicated to this idea. I can t think what it was called (it wasn t tailsteak.)
                        Message 11 of 18 , Feb 27, 2010
                        • 0 Attachment
                          What is the etymology of EPYC?

                          Also, I remember there being a webcomic that was dedicated to this idea. I can't think what it was called (it wasn't tailsteak.)
                        • Sai Emrys
                          ... I don t think boredom is an issue (except to the extent that some people are slower than others), because you can just parallelize several rounds. To me
                          Message 12 of 18 , Feb 27, 2010
                          • 0 Attachment
                            On Sat, Feb 27, 2010 at 10:47 AM, Jim Henry <jimhenry1973@...> wrote:
                            > You want a small enough overlap that people don't get
                            > bored or distracted waiting for their turn, but large enough that the
                            > story doesn't degenerate into total nonsense too quickly.

                            I don't think boredom is an issue (except to the extent that some
                            people are slower than others), because you can just parallelize
                            several rounds. To me the issue w/ larger prompts is just that people
                            will perceive it as hard to do, where more bite-sized chunks are
                            easier.

                            Otherwise I fully agree; it's just a difference in reasoning, not import.

                            On Sat, Feb 27, 2010 at 12:53 PM, John Lategan <jlateg@...> wrote:
                            > I also thought of doing an email one, but couldnt think how we would
                            > 'hide' the prev entries that arent meant to be seen?

                            Same way as one does for Relays - send the correct info only to the
                            next person, the full info to a moderator, nothing publicly.

                            FWIW, I should mention: my ultimate intent for this is to make it a
                            fully automated game module in a webapp I'm working on. (One intended
                            for The Masses™ and heavily chat-based, with various
                            server-administered ~chat-based games.) But I think there are
                            conlanger-specific versions that could also be fun for us, given that
                            we *can* make higher requirements of players' skillsets.

                            I'd be willing to run an email version of any of these, once we figure
                            out what good rules might be. I'll set up a form for signing up after
                            it's settled out.

                            (Mind of course that that need not be any singular set; I imagine
                            there should be multiple good methods.)

                            On Sat, Feb 27, 2010 at 4:10 PM, Claude Almansi
                            <claude.almansi@...> wrote:
                            > Problem: s/he'd have to view the spreadsheet to know when someone has
                            > answered before mailing the form to the next person.

                            Not really. You can get update notifications by email.

                            More relevant is that Google Docs forms have no way to make the
                            prompts depend on the spreadsheet data.

                            - Sai
                          • Jim Henry
                            ... I was thinking in terms of the live party-game version. Games where one person s turn takes a long time + a significant number of players == games where
                            Message 13 of 18 , Feb 28, 2010
                            • 0 Attachment
                              On Sat, Feb 27, 2010 at 10:11 PM, Sai Emrys <saizai@...> wrote:
                              > On Sat, Feb 27, 2010 at 10:47 AM, Jim Henry <jimhenry1973@...> wrote:
                              >> You want a small enough overlap that people don't get
                              >> bored or distracted waiting for their turn, but large enough that the
                              >> story doesn't degenerate into total nonsense too quickly.
                              >
                              > I don't think boredom is an issue (except to the extent that some
                              > people are slower than others),

                              I was thinking in terms of the live party-game version. Games where
                              one person's turn takes a long time + a significant number of players
                              == games where a long time passes between any given person's turns ==
                              tendency to boredom or distraction. In an email game, that's not an
                              issue.

                              > because you can just parallelize
                              > several rounds.

                              But that is a very good idea too.

                              > To me the issue w/ larger prompts is just that people
                              > will perceive it as hard to do, where more bite-sized chunks are
                              > easier.

                              True.


                              > On Sat, Feb 27, 2010 at 12:53 PM, John Lategan <jlateg@...> wrote:
                              >> I also thought of doing an email one, but couldnt think how we would
                              >> 'hide' the prev entries that arent meant to be seen?
                              >
                              > Same way as one does for Relays - send the correct info only to the
                              > next person, the full info to a moderator, nothing publicly.

                              And the moderator could also be a participant, if they have a separate
                              relay torch email address at which to receive the copies of other
                              people's turns and not look at said emails until the game is over.


                              > But I think there are
                              > conlanger-specific versions that could also be fun for us, given that
                              > we *can* make higher requirements of players' skillsets.

                              What about a Kalusa or Madjal kind of thing, where each turn consists
                              of several conlang sentences plus smooth English translations thereof,
                              the sentences forming parts of a semicoherent story and the corpus of
                              a semicoherent conlang without explicit lexicon or grammar?

                              And, in doing multiple rounds, maybe we should alter the orders of
                              people's turns on each round -- permute the list of players each round
                              -- so everyone follows and precedes everyone else sooner or later.

                              --
                              Jim Henry
                              http://www.pobox.com/~jimhenry/
                            • Sai Emrys
                              ... So was I. Standard practice IRL IME is to have it fully parallelized. Everybody starts a round simultaneously, and passes to the left. You re never waiting
                              Message 14 of 18 , Feb 28, 2010
                              • 0 Attachment
                                On Sun, Feb 28, 2010 at 6:27 AM, Jim Henry <jimhenry1973@...> wrote:
                                > I was thinking in terms of the live party-game version.  Games where
                                > one person's turn takes a long time + a significant number of players
                                > ==  games where a long time passes between any given person's turns ==
                                > tendency to boredom or distraction.

                                So was I. Standard practice IRL IME is to have it fully parallelized.
                                Everybody starts a round simultaneously, and passes to the left.
                                You're never waiting unless somebody is slower than average, in which
                                case they'll start to backlog.

                                If everyone's about the same speed, then you never have downtime.

                                > What about a Kalusa or Madjal kind of thing, where each turn consists
                                > of several conlang sentences plus smooth English translations thereof,
                                > the sentences forming parts of a semicoherent story and the corpus of
                                > a semicoherent conlang without explicit lexicon or grammar?
                                >
                                > And, in doing multiple rounds, maybe we should alter the orders of
                                > people's turns on each round -- permute the list of players each round
                                > -- so everyone follows and precedes everyone else sooner or later.

                                I take it that this proposal does *not* have the "only see one
                                previous step of history" constraint? So it's more like the general
                                'collaborative storytelling' method - i.e. full history always
                                visible.

                                - Sai
                              • Jim Henry
                                ... Well, we could do that, but I don t see how it would be interestingly different from the original Kalusa model. I d like to try it where each player only
                                Message 15 of 18 , Mar 1, 2010
                                • 0 Attachment
                                  On Sun, Feb 28, 2010 at 3:22 PM, Sai Emrys <saizai@...> wrote:
                                  > On Sun, Feb 28, 2010 at 6:27 AM, Jim Henry <jimhenry1973@...> wrote:

                                  >> What about a Kalusa or Madjal kind of thing, where each turn consists
                                  >> of several conlang sentences plus smooth English translations thereof,
                                  .......

                                  > I take it that this proposal does *not* have the "only see one
                                  > previous step of history" constraint? So it's more like the general
                                  > 'collaborative storytelling' method - i.e. full history always
                                  > visible.

                                  Well, we could do that, but I don't see how it would be interestingly
                                  different from the original Kalusa model. I'd like to try it where
                                  each player only sees part of the corpus, the part that they've
                                  contributed plus the parts that their predecessors of various rounds
                                  have contributed. Everyone's *langue* is influenced by everyone
                                  else's *parole*, directly or indirectly, but nobody has access to the
                                  entire corpus -- which is true on a larger scale of natlangs.

                                  --
                                  Jim Henry
                                  http://www.pobox.com/~jimhenry/
                                • Sai Emrys
                                  ... Could you be more specific? How many predecessors would one see? Would there be some a minimum cycle length (i.e. # of other people to go before you can go
                                  Message 16 of 18 , Mar 2, 2010
                                  • 0 Attachment
                                    On Mon, Mar 1, 2010 at 5:08 PM, Jim Henry <jimhenry1973@...> wrote:
                                    >  I'd like to try it where
                                    > each player only sees part of the corpus, the part that they've
                                    > contributed plus the parts that their predecessors of various rounds
                                    > have contributed.

                                    Could you be more specific? How many predecessors would one see? Would
                                    there be some a minimum cycle length (i.e. # of other people to go
                                    before you can go again)? Would you see *all* your own contributions?
                                    Would you see all contributions that you've previously seen, and so
                                    you're just seeing a subset of others' contributions?

                                    - Sai
                                  • Jim Henry
                                    ... Probably just one, unless the number of players is large, in which case it might make sense to let you see the previous two or even three players
                                    Message 17 of 18 , Mar 2, 2010
                                    • 0 Attachment
                                      On Tue, Mar 2, 2010 at 1:12 PM, Sai Emrys <saizai@...> wrote:
                                      > On Mon, Mar 1, 2010 at 5:08 PM, Jim Henry <jimhenry1973@...> wrote:
                                      >>  I'd like to try it where
                                      >> each player only sees part of the corpus, the part that they've
                                      >> contributed plus the parts that their predecessors of various rounds
                                      >> have contributed.
                                      >
                                      > Could you be more specific? How many predecessors would one see?

                                      Probably just one, unless the number of players is large, in which
                                      case it might make sense to let you see the previous two or even three
                                      players' contributions.

                                      > Would
                                      > there be some a minimum cycle length (i.e. # of other people to go
                                      > before you can go again)?

                                      I expect all other players would go between any given player's turns.

                                      > Would you see *all* your own contributions?

                                      The way we're talking about playing, online, I don't see how you could
                                      be kept ignorant of some of your own previous contributions.

                                      > Would you see all contributions that you've previously seen, and so
                                      > you're just seeing a subset of others' contributions?

                                      Yes, I expect so. Suppose there's a moderate number of players,
                                      maybe 5. Then we permute the list players as ABCDE, ABCED, ABDCE,
                                      etc; on the first round player D sees player C's contributions plus
                                      their own, then player E's and their own, then player B's and their
                                      own. And they have email archives of previous rounds, so on round 3
                                      they can look back and player C's round one contribution, etc.
                                      Everone sees two-fifths of the corpus. I figure we'd vary the number
                                      of previous players' contributions one sees with the total number of
                                      players so the fraction of the total corpus any player sees isn't so
                                      high as to prevent amusing mutation nor so low as to lead to total
                                      incoherence.

                                      Alternately, if we're playing with a website like the Kalusa site, but
                                      require logins to see part of the corpus and make additions to it,
                                      then we could be more sophisticated with not allowing a player
                                      constant access to the same subset of the corpus they've seen before
                                      -- what they see on any given day is always fifth of the corpus, say,
                                      but which particular sentences they can see varies from day to day.
                                      I'm not sure there would be any benefit to that; one could easily save
                                      offline copies of the stuff one has seen before and so build up a view
                                      of a larger subset of the corpus, thus advantaging the more
                                      computer-savvy players over the less so.

                                      --
                                      Jim Henry
                                      http://www.pobox.com/~jimhenry/
                                    • Sai Emrys
                                      I ll add one: P1 writes one paragraph. P2 summarizes that somehow in one sentence. P3 expands the summary into whole paragraph. Etc. Starting a new thread for
                                      Message 18 of 18 , Mar 7, 2010
                                      • 0 Attachment
                                        I'll add one: P1 writes one paragraph. P2 summarizes that somehow in
                                        one sentence. P3 expands the summary into whole paragraph. Etc.

                                        Starting a new thread for signups.

                                        - Sai
                                      Your message has been successfully submitted and would be delivered to recipients shortly.