I am creating a new Elan project (not my first one) and thought it could be wise to distinguish consultants and researchers by assigning a different type to their utterances.
Thus I created the types tx for consultants and txx for researchers (both aligned by symbolic association to a parent type ref). Tiers of both categories of participants/types are nonetheless called tx@xy regardless of their type being tx or txx.
Now when I want to use the transcription mode, I can only display one of the two types, either tx or txx. Is there any way to resolve this or is there a different workflow intended to distinguish researchers and consultants?
I’m not sure if I fully understand. If I create a tier and type setup based on your description, when switching to transcription mode, I can e.g. configure a 3 column layout with the ref type in the first column and tx and txx types in columns 2 and 3, or create a 2 column layout with only the tx and txx types.
I’m not sure if I get what it means that tiers of both categories are called tx@xy. Does this indicate that each parent tier of type ref either has a tx dependent tier or a txx dependent tier? (Even then, if the ref type is in the first column both tx and txx should be selectable in a 3-column layout.)
if I select the 3 column layout with ref as the first type, I can select both tx and txx for the second and third column.
I have another type called nt (also aligned to ref by symbolic association): if I choose this one for the first column, I can select tx and txx for the second and third one respectively
However, if I select either tx or txx for the first column, I can only select nt for the second and if I want to add a third column, I get a message that no further types are available.
As regards your question: the ref tiers always have child tiers called tx@something, but these tx tiers can have both types tx or txx. I thought it might have to do with the names of the tiers not (always) matching the names of the types (although this seems somehow weird to me).
In any case, if the setup works on your computer, it might be a strange problem only I have. I could possibly do with having the nt type being displayed as first column and then tx and txx as second and third respectively.
PS: I have been trying with having ref or nt displayed as first column in transcription mode now. The types are shown nicely, but the interplay between tx and txx still does not work: txx works fine, but clicking on a tx tier (or pressing “enter”) frequently leads the programme to hang up.
The names of tiers and types don’t have to match, that shouldn’t be the problem. Concerning your point 3: it might be that you cannot select tx (or txx) for the first column and txx (or tx) for second or third column if there is no tier hierarchy containing both a tier of the one type and a tier of the other type. And if indeed each ref tier has a child tier called tx@something which either has type tx or txx, these two types will not be in the same hierarchy and then it doesn’t make too much sense to have just these two types in the transcription mode’s table.
But maybe I’m wrong and I have still not a clear idea of the tier hierarchies. then it might be useful if you could send a small eaf file illustrating the problem to me (han.sloetjes AT mpi.nl). That would allow me too to try to reproduce the hang ups you mention too.
