Hi Stijn, I'm OK with most of what you suggest.
I think the report from Jean-Marie is in the 1.8 project, does that sound possible? In practical terms, and over 6 years of having the fill-in-the-blanks question type, I only ever heard of two people asking about variable-size input fields (including Jean-Marie), so I guess it's going to be difficult to get real input about that one... Let's hope he can help you with that. Also, and just mentioning this as an additional piece of information) I think it's worth mentioning that one of the philosophy of Dok€os (with I think is totally valid, and there are also a few studies about that around) is to present the types of questions in a more inspiring way to the teachers. For example, additionally to "Fill in the blanks" have "Listening comprehension", where you make clear to the teacher that you want him to add an audio file and that the blanks are going to be about the text the student listens to. One issue we came across recently with the negative scores (that you mentioned last) is the "propagation" of the negative results from different answers, and we decided to give the exercise a setting of propagation on/off (maybe that should be set at the question-level instead of the exercise-level): Case 1: propagation of negative results: Q1 = -1 Q2 = -1 Q3 = +2 Q4 = -1 Total: -1 Case 2: no propagation (trunking everything to 0): Q1 = -1 -> 0 Q2 = -1 -> 0 Q3 = +2 Q4 = -1 -> 0 Total: +2 The whole point in this setting is that some of the questions might have this form (A for "answer"): Q1: --- A1 = +2 A2 = -1 A3 = -1 A4 = -1 If someone checks all the answers, he shouldn't get a positive score, that's for sure (here he would get -1), but in some cases the teacher doesn't want the negative score to impact other questions, so he decides that the negative score is not propagated, so the total for the question is 0 instead of -1. I don't know how it is in Chamilo 2.0 but I take the opportunity of you being around that score to mention it, in case that would help you decide something important at the right moment. Yannick E l mar, 03-05-2011 a las 14:56 +0200, Stijn Van Hoecke escribió: > Hi fellow developers! > > I'm currently refactoring the fill in the blanks question type in order to > display the inputfield IN the text instead of in a table (which is extremely > clumsy when you have a large text). > > During this redesign I noticed a few issues on the issue tracker concerning > the features of the fill in the blanks that indicated some problems with the > current structure. > > I already implemented some minor functional problems on dev: use | to > separate answers instead of , (see http://support.chamilo.org/issues/3327) > and allow parsing of negative scores (see > http://support.chamilo.org/issues/3326). > > I'm about to add the remaining missing features that change a lot of the > structure and would appreciate your opinion on this refactoring. > > Attached to this email you find the mock-up of the > fill-in-the-questions-to-be. > The first issue I read about was the possibility to adjust the inputfield > length (proposed by Jean-Marie in some issue I can’t seem to find anymore > :-(). I took this into account when designing the new inputform. Numbers 1 > and 2 offer a clean solution to this. I split the textfield into two new > fields: sized textfield and uniform textfield. > > Sized textfields will always take the size of the individual answer much like > the ____ in the edit mode (repository), while uniform fields will all have > the same size. If the user chooses the uniform fields, some additional > settings are shown. (S)he will be able to define a fixed size (e.g. 15), use > the size of the longest answer in the text, or just use whatever size fits > best (unlimited). > > The second issue is the hardcoded default values. If a teacher wants to avoid > guessing and give a penalty, he should be able to put the default correct > score to +2, and default negative score to -1. Number 3 offers the > possibility to set these scores. > > Number 4 shows the parsing of negative scores, for example to give a penalty > for the most obvious wrong answer. > > I hope some people concerned with the fill in the blanks take their time to > review these options! > > Stijn Van Hoecke > Intern Hogent > _______________________________________________ Dev mailing list > Dev@lists.chamilo.org http://lists.chamilo.org/listinfo/dev _______________________________________________ Dev mailing list Dev@lists.chamilo.org http://lists.chamilo.org/listinfo/dev
