The SSNA stack: help is on the way!

Good news! The deal recently secured by @nadia on the RECODE Democracy project in Sweden, we have decided to hire the mighty @luca_mearelli to help us strengthen what I’m calling the “SSNA Stack”. This means Open Ethnographer and Graphryder.

“Strengthen” in this context means above all design and usability. The data model and engine behind SSNA, I think, are quite solid already. If you have admin server access and can do Ruby on Rails and SQL, you can do practically anything. But of course anthropologists and ethnographers tend not to have admin server access, and they do not program in Ruby or write SQL queries, so we need to do some work on the interfaces.

So: calling to people who are using these tools now or might be interested to do in the future. This is where you say what you would like. If we can, we’ll implement it.

I am talking to you, @amelia, @noemi, @federico_monaco, @akmunk, @meta, @louromainwatson, @Nermine. You are invited to take part in this discussion. Here we choose what Luca will do, then he’ll go out and do it.

@matthias and @daniel: don’t worry, I am not counting on you for this. I guess Luca will set up a staging server somewhere; when he is done, he’ll merge onto Github, and we will ask you to deploy (for open ethno; for Graphryder it will be even easier, though it makes sense for the dashboard to be on our server).

4 Likes

Hi everyone :slight_smile: Great news! Personally, my problem with the usability was twofold. Firstly, I could see the codes that I made and the number of posts under each code, but I could not know what the original posts within each code were. For instance; I would like to see all the original posts under the tag “Egypt” for example. Secondly, I couldn’t assign several codes to the same post, only one code is saved, no overlays. I don’t know if that makes any sense, or if there’s a way that this could be overcome, cause it would really be a great help.

1 Like

We have several threads on OE requests… I will see if I can find them so as not to duplicate efforts here.

But my highest priorities would be (echoing @Nermine as well):

  1. ability to assign multiple codes to one block of text (rather than having to hit enter, re-highlight, reassign)
  2. ability to assign codes to a post as a whole (without having to highlight the whole thing)
  3. ability to display all the text associated with a particular code
  4. no case sensitivity on the suggestive codes (so, when I type “Open” I should still see “open source” suggested)

Thanks. Only @amelia’s number 3 has made it to GitHub: Implement "annotation lists per code" in Open Ethnographer · Issue #28 · edgeryders/discourse-annotator · GitHub. This also takes care of @Nermine’s point

(Though I think she means an ethno code, not a tag)

Your number 1, requested by both of you, is an issue as of now (#126).

Pending issues on Open Ethno are here.

Hi @alberto, thanks for involving me again.
On my side i have had some experience especially on Tulip to visualize the process of some given tags distributing and combining during opencare timeline and Guy @melancon has supported me a lot in this. The goal for me would be to compare what i could “see like” in real life in Milan and what is represented by SSNA.

in this video an example with the tag MIR
I still have many doubts (for instance: is the date considered to pop the tag out about the post or comment, or about the tagging?) but i think it is worth to research on this stream as processes can tell stories and visualising the growth and changes of the graphs could bring to further insight and understanding.
So, i wonder if it ever be possible to render and visualise the opencare networks as process given a tag or more tags, and timelines.

1 Like