Open Ethnographer Feature Request Sandbox and Wishlist

IMPORTANT NOTE: Please now document all bugs and function requests on the GitHub. It’s easy to use and there’s no need for us to duplicate that effort here.

If you need a sandbox to discuss these before you put them on the GitHub, you can use the forum below to do so and @amelia, @alberto and @matthias can help turn your thoughts into a clear feature request. Otherwise, read through the list on the Git Hub, and add anything missing :slight_smile: We need to prioritise these in terms of importance (what you need now to make your work doable vs what would enhance your work but you don’t need in a pressing fashion), so you can also use this space for such discussions.

Please use this wiki to document wishlist items, which we can convert into formal requests if they are doable within the scope of our project/we can experiment with them in a Masters of Networks.

Open Ethnographer Bug and Feature Sandbox

Are there any bugs in the current system that you need resolving? What functions do you need Open Ethnographer to have in the short-term?
(Once added to Git Hub, please delete from here!)

  • The sort by UPDATED doesn’t seem to be working as it should. For example I change the code’s name, add a translation (thus updating it, I think) and it is not listed under updated. (@Wojt is this added to Git Hub? Just checking with you before I remove it here).

  • Some newly assigned automatically merge with other codes of the same name assigned by other ethnographers, while others don’t (the default, which is ideal, is for them not to automatically merge). (@Wojt same question as above)

Open Ethnographer Wishlist (in perceived order of difficulty)

Finding posts to code

  1. List of posts that have not been coded.

Coding for SSNA

  1. Toggle on/off in Open Ethnographer backend re appearance in SSNA
  2. Designate codes as “pattern” rather than semantic codes in backend (related to 1, but allows for a more granular categorisation of “off” codes).

Project-Level Coding

Being able to see everyone else’s codes as suggestions on a project is great in theory. In practice, though, it means that when you select someone else’s code, the annotation automatically falls under their code, and it becomes difficult to check that you meant the same thing as them.

  • Long-term solution : create colour-coded differentiation between ethnographers in the same project, so you can see whose code you are selecting in the drop-down.

Airtable Medium-Term Request

@amelia requests an import of our Sheets codebooks into the Open Ethnographer backend (code + code descriptions), and then a regular import of OE data into Airtable, which we will use as our codebook from now on (@hugi has set this up for BBU). This way we can work with the data as we create it, rather than using parallel processes which don’t sync. Until we have a codebook in OE, this seems like the best medium-term solution. @danohu @matthias @hugi, can we do this as soon as possible/have a conversation about it? @alberto happy to update you as well.

Codebook Integration

  1. Easily add and edit code definitions (the fewer clicks the better, ideally doable on main page with code list + definitions).
  2. Drag and drop hierarchical relationships (on main page).
  3. Ability to comment on each others’ codes.

Annotation View

  1. Generate a list of annotations where the two concepts co-occur. Ability to generate a co-occurrence annotation list of all the annotations where two codes co-occur and/or where either of two codes occurs, but in the same post (so both intersection and union). This requires fixing two things: 1. not being able to assign multiple codes to a single highlighted text in one go. 2. Creating a way of generating this list.

List of Active Ethnographers:

NGI: @CCS @Leonie @katejsim
POPREBEL: @Richard @Jan @Wojt @SZdenek @Jirka_Kocian
Plato: @MariaEuler
BBU: @hugi @jakobskote
Global: @amelia @alberto

(This will be a subsection of 📗 Open Ethnographer Manual along with the Ethnographic Coding Wiki)

6 Likes

Ethnographers, during your meetings, get in the habit of live documenting both wishlist and issue log items in this wiki!

1 Like

I’m not sure if this is necessarily about making coding easier but what would be useful for me (as a sociologist) is to have some form of online Post-It on individual participants, where I could keep notes about any self-references to their nationality, age, gender, etc.

Related to this, it would also help if we could identify individual participants in, say, a conversation about Polish populism who, for example, turn out to be Belgian and whose personal experiences of populism do thus not relate specifically to Poland. Can we add and remove individual participants to bring the data into line with the needs of specific sub-projects?

Thanks for getting this going, @amelia! Especially agree with your points on the codebook. Ongoing commenting/communication between coders on their code choices and definitions is a key point.

@katejsim and I are converting our Codebooks from Google Docs to Sheets. Will update Wishlist as we go!

Here is a teaser (already much easier to work with :slight_smile: )

2 Likes

Also, don’t know where to put this on the Wishlist, but can we agree on/is there already an existing Style Guide for codebooks, so everything is consistent and easier to read/cross-reference between teams?

1 Like

The wiki, but please feel free to add further style points to it! the more standard the better.

I’m about to add this to the wiki, but instead of using a separate “comments” cell just use the comment function on google sheets to add your own comments to your codes, so that a conversation can happen in the comment bubble thread and notification occur.

1 Like

Hi everybody!
@alberto @matthias @Jirka_Kocian @SZdenek @Jan @Richard

  1. So both Jan and I had this problem with changing languages in codes. Not a big issue, but still. Does it have to be the case that once we assign a language to a given code we cannot change it (even if we change the code itself?)
  2. Second thing. When we looked for the code car this is what we saw. Has it always been the case that the system returns all codewords with these three letters in this particular order?
  3. I think I already wrote about it somewhere, but I may put it here as well. So Jan and I have a code “Catholic Church”, not only did it automagically merge with already existing codes (we didn’t do it, maybe someone did it, maybe the system’s arcane algorithms) but it also divided them between alberto and johncoate. In Jan’s panel in the backend and in our codebook which links to it (as presented in the second printscreen) we have my and his codes grouped together (as should be the case after merging), but when we use the search codes option in the codes tab it shows several churches and our codes are assigned to/merged with alberto’s and johncoate’s (as shown in the first printscreen).
    To me it happens at random, several of my codes, despite having an already exisiting name, remain separate. Some merge without even asking me for permission…
    It’s also strange because Jan changed the name of the code to capital letters after the merger (catholic church -> Catholic Church) and my codes still show here as spelled lowercase.

  4. My sort by name doesn’t seem to be working.
    EDIT: It works fine as long as I have English set in User Settings. As soon as I change it to Polish this is what happens.
  5. We also noticed that when you change your language in user preferences, the system will stop showing you results of search codes in languages different than the one set there. Unless the language is English, that is.
    I had my language set to Polish and coded for a name of a radical-right organisation called “all-polish youth” (I tend to code in English first for some reason). The system had no problems finding it. But when I added the Polish translation (“młodzież wszechpolska”) it returned no results for “all-polish” youth and was able to find the code only when I put “mlodziez”. It all works in reverse when I set my language to English - I can search for English names of codes, but the system returns nothing for Polish translations. Do you guys have the same issue? Can we ever make it so that a given code can be searched for its English and native language name?

Hi Wojt! Thanks for this. It’s a wiki, so instead of adding this as a comment, just go ahead and press the “Edit” button at the bottom of the main post, and add these into the section “Open Ethnographer Issue Log”

Let me know if you have any problems :slight_smile:

For the first one, there’s an easy workaround, though the system needs fixing (I’ve encountered this too) – you just have to move the Polish codename to a new box and select the Polish language in the new dropdown, and write the English code in the box that was defined as “English” rather than trying to select a new language via dropdown. basically, the system for some reason doesn’t like it when you try to assign a new language via the dropdown, but it doesn’t mind if you just add a new one, in Polish, and rename the original code to the English translation. In short, don’t reassign via dropdown at all, just retype the code in the language of the original dropdown and assign the translation via adding a new box. It’s one more step but until we sort this out it’s not too tedious.

2 Likes

@amelia and @Wojt Many thanks Amelia. Wojtek and I are wondering if you could meet with us tomorrow (Monday) at my 10:00 am?
J

Can we do 11 your time 4 mine? Have a call that looks like it will be running over.

1 Like

Even better for me. 11:00 (here) it is. Hope you are well! J

@Wojt Hope it is OK for you as well? I have my next engagement (a seminar on doing ethnographic fieldwork in difficult settings (mostly Russia) and in the midsts of a pandemic at 1:00. It may be of interest to @amelia, if you have time. Just in case:
Here is the direct registration invitation that you can share with your students/non ASEEES members.

You are invited to a Zoom webinar.
When: Jul 20, 2020 01:00 PM Eastern Time (US and Canada)

Topic: “Fieldwork interruptus: research in the Time of COVID”

Register in advance for this webinar:

After registering, you will receive a confirmation email containing information about joining the webinar.

1 Like

I’ve already registered! Currently working on polishing our codebook, but will be there!
Aren’t we having a meeting in a quarter?

1 Like

Ok, my bad. In 75 minutes :slight_smile:
We can talk about it then.
[distinct polishing sounds]

1 Like

Great. Do you have a zoom link we can use?

Yup (below), but give me please 10 minutes. I am going over the list of our questions one more time
Join Zoom Meeting

Join by SIP
99169454795@zoomcrc.com

Meeting ID: 991 6945 4795
Password: 548897
One tap mobile
+13126266799,99169454795# US (Chicago)
+16465588656,99169454795# US (New York)

Join By Phone
+1 312 626 6799 US (Chicago)
+1 646 558 8656 US (New York)
+1 301 715 8592 US (Germantown)
+1 346 248 7799 US (Houston)
+1 669 900 9128 US (San Jose)
+1 253 215 8782 US (Tacoma)
Meeting ID: 991 6945 4795
Find your local number: https://rutgers.zoom.us/u/abZUmaXDMm

Join by Skype for Business

If you have any questions, please contact the Office of Information Technology Help Desk

You are welcome to post them so I have them in visual form too.

  1. Issue number 1: let’s go over the relative pluses and minuses of two “systems” “mother-child” versus “venn diagram intersections”

  2. Second key issue: actual versus proximate intersection.

  3. For example: a three level construction: (1) Holidays, cars, houses, mobiles as children of (2) status competition, as a child of (3) status.

  4. Need code STATUS, we have STATUS MARKERS under socioeconomic topics

  5. OR: Collective decision making: democratic versus authoritarian?

  6. OR: Polskość a Katolicyzm

  7. What to do about past/now comparisons? (see lines 91 and 131, 167): intersection “issue” and “time”

  8. Basic political : extreme left (e.g. totalitarian communists), far left (socialists), left (socdem), liberals (centre), right (conservatives), far right, extreme right (fascists)

  9. How to restart/rekindle conversations that are “dead”?

  10. Relative deprivation

  11. Can an invivo code be used as a mother code?

  12. Feminine/masculine forms in Polish, what about them?

  13. Foreigners/expats on Polish forum, what to do? Line 46 (decent income by Noemi)

  14. Will “positive emotion/feeling” be a category?

  15. People speak of resignation, maybe merge with disappointment/despondency

  16. Maybe we should introduce intermediate category for Morals and values, e.g. political values, economic values (see Polish codebook line 76)

  17. Line 96, what about

  18. Dimensions of polarisation

  19. Pensions/low pensions, what about it?

  20. We need the code DIGNITY

  21. Catholic Church, when spelled with lowercase shows johncoate’s code with one Jan’s annotation and alberto’s with some codes of wojt, Jan’s code shows only when

  22. Four codes: migrants, immigrants, emigrants, refugees

@amelia The first two are key, in addition to @Wojt long-ish earlier post. We will explain.

I am ready at that link.