Most wanted workflows and features for working with sources, codes and codebooks

[kadewe]

I just did a usability/feature discussion in one of my research seminars with students working with OpenQDA. Some are new, some are unresolved issues (see bottom of post for references). I put a (1 star) would be nice, (2 star) needed, (3 star) non negotiable needed feature

so the context is:
all students in one project, doing interviews which are shared with the project group,
one large common codebook, specialized codebooks for students for specific research questions.

Codebook handling in CODING VIEW:

• Hide a codebook or all other codebooks. (1 star)
• Move codebooks (up/down) (1 star)

Merging and splitting codes:

• merge a code into one (in the codes menu "merge into") > all the selections become re-assigned to the other code, this code is then deleted (ATTENTION: this ripples through all sources!!! In collab mode there should be agreement for this) (3 star)
• split a code/bulk move of selections to new code: probably best in a "list of selections by code"-view (coming in 1.0.7?) you select the selections you want to move into the new code, then apply "move to new code" (3 star)

Moving codes between codebooks:

• difficult from a implementation view. Suggestion: maybe an "import codebook into codebook" could do most of the heavy lifting, if there also would be "delete selected codes" bulk deleter for codes. (2 star)

"Folders", Sets, and Cases:
As soon as you have a lot of sources, things need to be organized. Let's have a look at REFI standard:

Case | Defines a specific methodological unit of analysis.
Set | Defines an organizational grouping of project entities.

Variables (Metadata)
Case | Can hold assigned elements.
Set | Cannot hold variables or metadata.

Contents
Case | Maps to source files or specific coded data segments.
SET | Contains GUID references to grouped entities (Codes, Sources, Cases).

In 1.0.7. support for source-level variables is coming, which allows you basically to define CASES for exporting data (but only based on sources, not segments). Still, we need native UI support for sets and cases for REFI. We already have them in the database, but not in the UI.

Now we have the recurring wish to sort the sources into folders. There are no folders in REFI but SETs. Therefore a first help for people with 100 sources would be to have two VIEWS for SOURCES: "ALL" (just showing all sources as it is now) or "SETS": there you can define sets and for each set you have a list of all sources which you can select/deselect to add/remove them from a set. SETS look like folders and you can open or close them. There are no SUBSETS (right?), but you can create as many SETS as you need (3 star + 2 extra stars from Andreas )

Summary in decreasing importance:

1. SETs (5 stars)
2. MERGE a code into another one (3 stars)
3. SPLIT code/BULK MOVE of selections to another code (3 stars)
4. import codebook into codebook / bulk delete codes out of codebook (2 stars)
5. hide and move codebooks (1 star)

yours
Karsten fresh out of seminar

Change selection of text section #195
Merging codes #194
Moving Codes between Codebooks #159
Sets and Cases #14 4
Document level variables #287 < upcoming in 1.0.7!

[kadewe]

ok, and i forgot mentioning Change selection of text section #195
right now, you need to delete the selection and create a new selection. This is not optimal if you have already added a memo (which you will loose when you delete the selection). For a satisfactory solution we need to implement little selection handles

I would postpone this until we have agreed on a maintained CRDT (Conflict-free Replicated Data Type) editor as Quill (the editor we are using) is no longer maintained. There are good alternatives, though, but @jankapunkt @helge-hb need to evaluate them.
Because we only do collaborative text editing in the preparation phase it is not of the highest priority right now in comparison, but as we also want to provide a nice and "jummy" text highlighting view with excellent text rendering we should take the time to do a good selection (sustainable maintenance being the most important criteria). Anybody reading this and have an idea, please share!