![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
One of tirinian's solutions for a problem in Oath was described, somewhat mockingly as this. But more and more often, I've been coming up with examples. Maybe not ethical problems so much, but at least messy social dynamic problems.
Take trying to figure out what your sixteen players want to do. We used to send email saying "tell us whether you think X or Y." Some people would send email saying X or email saying Y, and some would send email saying "Hmm, they both sound good," or, "Y has some benefits, but X might be better in the long run", or, heaven forbid, "My character thinks X, but my player thinks Y." Then we'd have to look at all of them and try and announce to the run "Lo! It is X" and nobody is really sure how much of a decision it was. Now, we have web pages with radio buttons, that send email to a perl script, that passes the output to gnuplot and convert and puts it on a web page. No complicated human language parsing required, and no questions about why we never go to Highguard. :)
Or, in another system I was describing earlier today, I'm writing a game and people keep being lame and not writing their assigned character sheets, but nobody wants to actually scold one's fellow GMs for something that's a volunteer fun activity. Write a perl script that tracks the number of sheets written versus time left and tells you how behind (or not) you are, and who's done stuff recently, and count on natural competitiveness of enough of the GMs to not want to be the Officially Lamest GM that they get stuff done. (This may not work on all people, but it works on me...)
In a slightly different sort of technical/social engineering, mjp and I have a system of composing email to players in Oath, where one of us writes the email, and it goes in the outgoing-drafts mailbox, and then the other one runs the "agree" program, and it says if there's mail waiting to go out, and lets the person read it or diff it from the previous version, or edit it, or send it. But if you edit it, you don't get to send it, it goes back in the box for the other person to agree to. So 90% of the stuff doesn't need to be talked about before writing, just someone needs to have an idea about the answer, and then the other person has to like it. And calling the program "agree" helps enforce the psychology of it, more than calling it "sendstuff" or "vet-stupid-other-gm-email" would.
Take trying to figure out what your sixteen players want to do. We used to send email saying "tell us whether you think X or Y." Some people would send email saying X or email saying Y, and some would send email saying "Hmm, they both sound good," or, "Y has some benefits, but X might be better in the long run", or, heaven forbid, "My character thinks X, but my player thinks Y." Then we'd have to look at all of them and try and announce to the run "Lo! It is X" and nobody is really sure how much of a decision it was. Now, we have web pages with radio buttons, that send email to a perl script, that passes the output to gnuplot and convert and puts it on a web page. No complicated human language parsing required, and no questions about why we never go to Highguard. :)
Or, in another system I was describing earlier today, I'm writing a game and people keep being lame and not writing their assigned character sheets, but nobody wants to actually scold one's fellow GMs for something that's a volunteer fun activity. Write a perl script that tracks the number of sheets written versus time left and tells you how behind (or not) you are, and who's done stuff recently, and count on natural competitiveness of enough of the GMs to not want to be the Officially Lamest GM that they get stuff done. (This may not work on all people, but it works on me...)
In a slightly different sort of technical/social engineering, mjp and I have a system of composing email to players in Oath, where one of us writes the email, and it goes in the outgoing-drafts mailbox, and then the other one runs the "agree" program, and it says if there's mail waiting to go out, and lets the person read it or diff it from the previous version, or edit it, or send it. But if you edit it, you don't get to send it, it goes back in the box for the other person to agree to. So 90% of the stuff doesn't need to be talked about before writing, just someone needs to have an idea about the answer, and then the other person has to like it. And calling the program "agree" helps enforce the psychology of it, more than calling it "sendstuff" or "vet-stupid-other-gm-email" would.
