I will list here some ideas from the timetabling community. The aim is to bring ideas from different projects together and see where we could increase interoperability and create standards. This project is community driven, so please provide your feedback. I think that there is no perfect way of doing something, i.e. I am open to new ideas -
your ideas!
At least we should create
one consistent object model for the following 'projects':
Do you want to be included? Please
contact me. There is no deadline yet.
- The mailing list archive is here and you could easily subscribe to the gstpl-standardslist if you want (preferred method for communication).
- The related forum with public access is here.
- Check out the current sources via:
svn checkout https://gstpl.svn.sourceforge.net/svnroot/gstpl/branches/ttstandards ttstandards
- The class diagram is ready for downloading: ttstandards-data.jpg.
- Here you will find the latest news. (Linked with itself; for now)
Idea
|
|
Author
|
1. Hire more members, i.e. projects to be adapted. Then we could create a more general model.
2. Create the object model. Then we can form the file format and the database-table layout
3. If 2. does not fail, then plan the framework, decide which language ...
We should think about pro's and con's of Java, C++ and C# (Even Others?)
|
|
pk
|
|
|
|
Should we use UML and create code from it?
We should use XML to store the objects, right?
Look at the initial class diagram:
ttstandards-data.jpg or check out the current sources via:
svn checkout https://gstpl.svn.sourceforge.net/svnroot/gstpl/branches/ttstandards ttstandards
I collected some
links of currently available open source 'timetabling frameworks' written in C++, C# and Java.
Language
|
|
Pro's
|
|
Contra's
|
|
|
|
|
|
Java
|
|
|
|
|
C#
|
|
|
|
|
C++
|
|
|
|
|
There were at least three attempts to create a timetabling language:
- Standard TimeTabling Language (STTL)
- TimeTabling Mark-up Language (TTML)
- UniLang
Taken from
here there they said:
Perhaps, the main reason these languages have not been adopted as standard is that
they offer no advantages to the user over any traditional programming language.
These idealistic languages do not simplify the modelling process, and can even be
restrictive in that they do not have all the features of a modern programming
language, are overly complicated or appear cumbersome.
