Which choice? [poll]{First choice}{Second choice}{Third choice}
How much do you agree? [poll: Agree.....Disagree]
How likely? [poll: probability]
How many? [poll: number]
The first is a conventional multiple-choice poll, with the options in curly braces. The second gives radio buttons along a scale, with labels (such as Agree/Disagree) on the left and right. After you’ve submitted a ballot, it will show you the vote breakdown. There are also probability and number polls, for which you can get the mean and median. There is also a ‘raw data’ link, which provides a CSV file with all of the votes broken down by question and user (either their username, if they unchecked the ‘vote anonymously’ box, or a number), so people can do fancy analysis on the results.
The main areas that still need work are the templates (the results look a little ugly), error handling, and bug-testing. If all goes well, it should be ready to deploy on Less Wrong in another week or two.
Mostly finished, but lost motivation due to lack of feedback—I heard that a test was set up on an internal Benton house server, but never saw it or heard what the results were, and other things took precedence. I plan to come back to this, but need a better test environment or the same thing will happen again.
My polling code for Less Wrong is now (mostly) complete. Check it out with ‘git clone http://github.com/jimrandomh/lesswrong.git’. The syntax for creating a poll in a comment is:
The first is a conventional multiple-choice poll, with the options in curly braces. The second gives radio buttons along a scale, with labels (such as Agree/Disagree) on the left and right. After you’ve submitted a ballot, it will show you the vote breakdown. There are also probability and number polls, for which you can get the mean and median. There is also a ‘raw data’ link, which provides a CSV file with all of the votes broken down by question and user (either their username, if they unchecked the ‘vote anonymously’ box, or a number), so people can do fancy analysis on the results.
The main areas that still need work are the templates (the results look a little ugly), error handling, and bug-testing. If all goes well, it should be ready to deploy on Less Wrong in another week or two.
What is the status of this attempt to add support for polls to Less Wrong please?
Mostly finished, but lost motivation due to lack of feedback—I heard that a test was set up on an internal Benton house server, but never saw it or heard what the results were, and other things took precedence. I plan to come back to this, but need a better test environment or the same thing will happen again.
I was going through the top-rated comments list, and found the several-greats grandparent—did this ever make any more progress?
Unfortunately not. The code from where I left off is still up on github, if anyone wants to pick it up. (The URL seems to have changed—it’s now https://github.com/jimrandomh/lesswrong instead of https://github.com/jimrandomh/lesswrong.git).
:) Thanks! I am excited.