View Thread > H2O Meta > 0.5 Feature Priorities > polling and custom attributes
Now that we have a list of potential features, the next step is to prioritize
which features should be implemented in the 0.5 release and in what order. Each feature has a guestimated number of days associated with it. We have about 120 developer days before a release in mid-February. Your job is to pick out which features should be included in the next release (all of your features should only take 120 days to complete) and then, *very importantly*, to prioritize when those features should be implemented. Once we set a concrete release date in February, the release date will be firm, while the number of features we complete will not, so the priority of the features will determine which features will definitely get into the release and which may not.
As our esteemed leader, Jonathan Zittrain will be ultimately responsible for putting together the final list of priorities for the release, but the feedback we get from this discussion will play a very important role in his decision.
For the list of features, see:
http://h2odev.law.harvard.edu/viewcvs/*checkout*/h2o/docs/plan/feature-priorities.csv?rev=1.6
I think the following (somewhat ranked by priority) are the most important/coolest features:
o Bulk user load (5)
o Bulk project acceptance page (5)
o Help system infrastructure - support for including "?" links... (3)
o More basic FAQs... (1)
o Project deletion (3)
o polling, incl. poll setup in rotisserie management screen... (15)
o optional automatic expulsion on round skip... (5)
o pre-defined & custom attribues (10) & (15)
o all forms of routing, but I think the most interesting are: bounce back (2), poll-based (2), attribute (5), and value (5)
o Full text rotisserie post search (15)
o simple post comment system -- I like the idea of adding a simple post comment system, but I don't think participants should be able to post until after the round (or maybe the whole rotisserie) has ended. This feature needs to be thought about carefully so that it won't undermine the whole spirit of the rotisserie (10)
o Additional questions in rotisserie rounds... (3)
o Add crank error cases - better error testing for the crank (10)
o Word count support on response page (2)
o Improve flow between view thread page and rating assignment page (3)
o Improve available projects list... (3)
that's about 120 days...
These look a good set of tasks/priorities for me. The only ones that I'm not sure about is how important extra custom attributes or specialized types of polling are. I haven't seen the beta feedback from users, so I don't have special insight here, but in making the priority decision, I would try to consider what features the instructors and the students are most going to need to have a high-quality experience with this tool.
Tony
The polling and custom attributes are important b/c they allow more interesting kinds of routing. For example, including the option to select a poll answer in addition to the standard paragraph form answer allows the system to route the posts based on those poll answers. Imagine the case in which in which law school students are asked to indicate whether they are arguing for the plaintiff or for the defendant and the system routes posts among the plaintiffs / defendants for the second round and then between the two different groups for the third round. Similarly, imagine a discussion in which all of the participants has indicated their home countries and the discussion routes posts depending on the similarity or different of the students' home countries. These kinds of interesting routing algorithms are at the heart of what makes the rotisserie powerful and are vitally important if the rotisserie is to reach anywhere near its potential.
The poll is a simple way to add graphical feedback and fodder for in-class discussion. It's quick for the instructor to set up, quick for the students to answer, and can be used to launch a rotisserie discussion or stand-alone. That's why I ranked it as a high priority. Student attributes didn't interest me much.