ok guys, so I found what seems to be a bug/glitch in the system. steps to reproduce: create a simple survey like the following: 

1. question with a few radio list options. set 'other' to yes
2. array question with array filtering based on question 1 (so to only display the selections chosen there). Include a subquestion 'other' with field value {question1_other} (ie {G01Q01_other} for instance.
3. array question

 this works fine wrt question, and if you enter an 'other' option in the first question, it displays nicely in the question 2 array. however, if you add a condition in the condition designer to only display question 3 on a particular selection in question 1, that breaks the logic between questions 1 and 2, so that if you enter an 'other' option in q1, the field will display in q2, but not have the title. once you select the option in q1 that will show q3, it shows. if that is unselected it does not. 
please advise. 
Checked on limesurvey 4, as well as latest 5.1.13.

Hi,
please add a lss export of this.

What happens if you use "relevance equation" instead of the "outdated" condition designer?

Joffm

Here’s an example lss that should give the same issue I described above

I was not aware the condition designer had been replaced..

I just checked in 5.1.13 : issue seems different.
Condition is OK, but text is updated only of SQ01 is checked.
 
The update of HTML is wrapped in a ficntion checking this.
The LEMval('613131X49X896SQ001.NAOK') == "Y" is an expression manager issue.

Best seems to report the issue community.limesurvey.org/bug-tracker/

The issue didn't happen with 3.X version (attached lss)
 

 

this is the same issue, I think.

what's notable is that this happens only while there is a condition for question three on question one subquestion one. if you remove that condition you will probably see the issue is not there. would you confirm this?

I was not aware the condition designer had been replaced..

Well, the condition designer hasn't really been "replaced", it is still there, for "historical" and "sentimental" reasons, I guess...

But I don't think that many power users are using the condition designer since Expression Manager was implemented. EM is just a lot more flexible, quicker and better over all.

So bugs with conditions will most probably take a long time to be detected, because hardly very few will use it, I guess.

It's not only with condition designer : i replace by same condition in expression : same issue.
Again (and again) : please report the issue.

Ok, I reported it as bug. bugs.limesurvey.org/view.php?id=17627