Coding Questions/Documentation

Documenting some of my thoughts, and also opening up questions about priorities for the rest of the project.

First, now that I am beginning to code some of the community manager authored posts, I am not so sure it is a good idea. Particularly the ones about meeting the edgeryders and describing events like Open Village: they end up mostly summarising stories that people have told elsewhere, and I am worried that coding them will create some false associations (it will look on the graph like people are discussing Open Insulin, but really it is just an announcement from the community manager). I don’t want to skew the graph in this way. I will still go through and make sure actual content isn’t there, but otherwise I think it is better to leave them. I’ve created a little code to mark the post as read but not coded (“nc”) and will delete that code once I’m done.

If anyone has any objections or counterpoints, do let me know.

Secondly, I want to have a discussion about priorities from here on out. At this point, there are a few different things I could prioritise.

  1. Coding the remaining uncoded stories. These all have 0, 1 or 2 comments (max), or they are authored by community managers.

  2. Doing a second pass of existing codes and cleaning them up. This is very important, but will be very slow going since the technology is not there to support it.

  3. Establishing clear hierarchies among codes. Same issue as pont 2 (important, but tech doesn’t support doing so easily).

  4. Authoring papers about ethnographic findings. At this point, there are a lot of different strands well worth writing about (not to mention the project as a whole).

  5. Keeping up with new posts. (This, I think, is vital and isn’t too time consuming).

  6. Engaging in other kinds of ethnographic work (like commenting on posts to elicit further responses, making posts that elicit feedback on the ethnographic data, joining community calls to take notes, and other more participant-observation oriented work).