Messages

Hello Team!
It's me back again each semester 

This time I would suggest the use of the current filter values as default values when the user presses the "Add" button for constraints.

For instance, I noticed the importance of this when I was inserting teachers time constraints (SCREEN "Constraints teacher not available times"). Let me explain...

As I had many constraints already set inherited from the previous semester, I was firstly filtering by name and modifying the preferences previously saved. At this moment, when I was filtering by a given name and that name had NO constraints yet, I was forced to "Add" a new constraint to that teacher and several times I included the constraint with the first name on the list, having so to step back and remove the wrong constraint.

I think this mistake happens because our brain thinks: "I have already chosen the name", but it does NOT take into account that it was on the previous screen. So, how about we give a hand to the brain and inherit the values from filters to the next "Add" screen.

[Could SOMEONE else shed some light on this?]

In my opinion that values will not represent the real/fair situation. That values will just fool other teachers and also yourself.

Example:
Teacher A: 1 subject, 4 activities
Teacher B: 1 subject, 4 activities
Teacher C: 2 subjects, 4 activities

So is it fair? If you see this values you will think Teacher A and B got similar fair data, while teacher C is double bad.

But if you will look into the advanced statistics you maybe will see this:
Teacher A: Physics in Group 5a, 5b, 5c, 5d
Teacher B: Physics in Group 7a, 8a, 9a, 10a
Teacher C: Sport in Group 5a, 5b, 6a, 6b

So Teacher A need to prepare in fact only 1 "subject" (build up experiments,... ), since he will just repeat it 4 times.
But Teacher B must prepare much more: He need to prepare another 3 experiments. So much more work.
And Teacher C: I don't think he prepare very much. I guess less work even if you compare with Teacher A.

So in my opinion i can't see a reason why to add those numbers into statistics. Those numbers will only delude yourself. You can see that only in the advanced statistics. I think that one should be used.
Let me know your opinion.

If I correctly understood your comment you mean that 5a, 5b, 5c and 5d are the same year, but different classes. On the other hand, 7a, 8a, 9a and 10a are different years, say, the seventh, eighth, ninth and tenth years respectively. And so, the same subject in different years, in fact, means completly different matters.

In this point of view you are completly right, Volker. And, this way, in fact, my suggestion can fool my coleagues.
By the way, we could think about a conditional flag to make the FET user choose whether to show it or not.

Could SOMEONE else shed some light on this?

Liviu and Volker,
After testing the new feature I figured out about another situation that occurs in my institution.
Here we have a rule that recomends coordinators not to assign more than three different subjects to one teacher. Sometimes it occurs, but it's not the ideal scenario.
So, I thought about include two more collumns in the statistics file:
* "Total of Activities"
* "Differente Subjects" (counting different string description of the subjects assigned to the teacher)
Do you think I am asking so much? 
Thanks for your attention.

Dear Liviu and Volker,
I've just tested the implementation and it is exactly like I thought! The goal here is to clearly show the other teachers that "I" have done the work fairly, as Volker mentioned before in this post.
Additionaly, the collumns order seems very good to me too, Benahmed Abdelkrim.
THANK YOU VERY MUCH!

Dear Volker,

How about including in the "teachers_statistics.html" file the information shown on screen FET -> statistics -> Teachers statistics?
Sometimes I would like to send it to my coleagues, but I can't, because that concise data is only shown on screen.
Thanks,

Fernando A F Lordão

Ok, ok, man! I give up! I can see your attention trying to show me that, but I also see that I need to study CSS in order to use it properly. 
So, given that I can't deal with CSS yet, let me give another suggestion: Is it difficult to automatically generate images like those colored views in "Timetable > View (students)" and "Timetable > View (teachers)"?

[old]

It's me again, Volker!  I'm sorry!

When I tried "display:none;" only titles of the days (MON, TUE, WED etc) were hidden, but I want to hide the lines whose values are allways "-x-". If you can, look at the appearence of the files I attached at this post to understand the effect I want (looking at first and second subgroups is enough).

If you could, I would ask you to make these changes into schedule_subgroups_days_horizontal_old.css in order to achieve similar appearence between the HTML attached files.

Dear Volker,

I appreciate very much your considerations and tested both tips you gave about CSS. Now, I want to comment as follows:

1. Idle rows were not omitted with CSS changes you suggested. See that there are data in those rows, each cell is filled with "-x-", so they have to be shown. Because of that I deleted the corresponding source code in order to achieve the desired result;

2. About columns width, changing CSS my first column became larger than I want;

3. Regarding width percentage beyond 100%, I did NOT find any rendering problems with Chrome, Firefox, IE or Edge. In fact I had already thought about these potential problem and I was going to suggest a individual percentage to the first column, but considering the satisfactory result applying width attribute only to days columns, then I thought it wouldn't worth the additional work.

P.S.: I did my changes with two regex find and replace operations.

I have started to work with FET recently, but my colleagues have been working with it for 2 semesters.

To present scheduled time, they usually capture snapshots from the views generated through the menus "Timetable > View (students)" and "Timetable > View (teachers)". They agree with me that this procedure is a repetitive and boring task, but they do it this way because they think HTML files automatically generated by FET is difficult to appreciate due to the lack of alignment and width equalization between columns.

To trick this factor I tried to do a simple modification in HTML generated files in order to try a better width equalization between columns and they thought that it is better now (see old and new files in attachments).

The modification consists in inserting "width" attribute in "th" tags of day labels. It was assigned width value equals to (100%/NUMBER_OF_DAYS). Thus, as we used only weekdays, the attribute value was 20% and we modified HTML source code as follows:

* ORIGINAL CODE

Code Select Expand


        <tr>
          <!-- span -->
          <th class="xAxis">MON</th>
          <th class="xAxis">TUE</th>
          <th class="xAxis">WED</th>
          <th class="xAxis">THU</th>
          <th class="xAxis">FRI</th>
        </tr>


* MODIFIED CODE

Code Select Expand


        <tr>
          <!-- span -->
          <th class="xAxis" width="20%">MON</th>
          <th class="xAxis" width="20%">TUE</th>
          <th class="xAxis" width="20%">WED</th>
          <th class="xAxis" width="20%">THU</th>
          <th class="xAxis" width="20%">FRI</th>
        </tr>


Additionaly we also modified HTML files to ommit what we say "idle rows" like those ones produced by the following source code:

Code Select Expand


        <tr>
          <th class="yAxis">07:00 - 07:50</th>
          <td>-x-</td>
          <td>-x-</td>
          <td>-x-</td>
          <td>-x-</td>
          <td>-x-</td>
        </tr>


Oh, my god!!! How long time it could save me if I have known before.
Next time, maybe
Thanks a lot for helping me.

In fact, Liviu, it really worked in the way I expected!
Nice tip!

I'd suggest you to go to "All time constraints" dialog, filter "Contains" "Teacher not available", then "Sorted". Because the weight is always 100% and the teacher is after the weight in the constraint description, it works

If this does not exactly solve your suggestion, please tell me and I'll add your words in the TODO.

By default, teachers time constraints are permanently shown at same sequence that we add.
It would be useful if we could sort them by teachers' name in order to reduce time wasting for systematic modifications in time restrictions for many teachers. By the proposed way we could do modifications teacher-by-teacher following the name sequence without having to apply filter for each of them.
Similar sort options could be designed to other restrictions.

When importing activities from .csv text file FET 5.28.0 shows the message "Skipped line X: Fields 'Split Duration' and 'Total Duration' haven't the same value." which means that the system waits the same value in both fields and, consequently, avoids activities as the followings:

* ACTIVITY ONE
Teacher: Tony
Students: Elementary_A1
Subject: Introduction
Total duration: 2
Split duration: 1
(2 hours/day, 1 day/week)

* ACTIVITY TWO
Teacher: Tony
Students: Elementary_A1
Subject: Introduction
Total duration: 4
Split duration: 2
(2 hours/day, 2 days/week)

So, I think FET has to change this validation to the following:
1. Total duration >= Split Duration

Additionally, FET would use the following values to fill Duration in each splited day:
* If day count <= (Total Duration MOD Split Duration), then day duration will be  ((Total duration DIV Split Duration) + 1)
* If day count > (Total Duration MOD Split Duration), then day duration will be  (Total duration DIV Split Duration)

The attached file contains:
* .fet file to create an environment test
* .csv file to reproduce the importing error
* .png images containing some snapshots from importing steps to show how the screen fields were filled out.