Q: Is this Software still under development?
A: No.

Q: Why not?
A: This code was a by-product of my diploma thesis. As I have completed my thesis, I no longer need it.
   I published the code for other scientists to use and/or enhance it. However, as long as you comply with
   the license you can pretty much do whatever you want with it...

Q: But I want/need feature X.
A: You will have to add it yourself, the sources are open.

Q: Where are the sources?
A: They should have come with the program.
   Should the distributor (that is the person whom you got the program from) deny you the sources
   and/or you didn't recieve a copy of the license (GPL v3), please report this incident to [3].

Q: The sources are incomplete, some testfunction code is missing. Where can I obtain the missing pieces?
A: I wrote a wrapper to some testfunctions (implemented by Huang Ling for CEC07 and CEC09 MOEA contest).
   Due to license issues (those files do not come with any type of license information) I cannot distribute
   them. You can however download those files yourself from [4].
   Be aware, that the implementation is pretty crappy and not very efficient. You might be better off
   implementing the missing testfunctions yourself.

Q: Am I allowed to use the code in a commercial product?
A: Yes, under the terms of the GPL (version 3 or at your option any later version). If you cannot
   comply with all the requirements arising from the license, you MUST NOT distribute any derivate
   works at all.

Q: I am not comfortable with the GPL, is any other license model available?
A: No.

Q: I have questions about the algorithms.
A: An excellent place to look for information about them is my diploma thesis. It is available free of
   charge from [2] and [3].

Q: Is this code thread-safe?
A: No. There are a couple of issues with the code referring to that. I didn't write it thread-safe because
   that would have required extensive copying of information instead of just copying some pointers to that
   information. As a result the program would have been much slower on uniprocessors. But if you just want
   to do runs on testfunctions, you need to run the program a couple of times anyway, so you can achieve an
   even better performance boost by running N instances of the program at once on a machine with N processors.
   So ultimately there is no need for the program beeing thread-safe at all.

Q: I have found a bug in your code.
A: Oops, thanks for reporting it to [3]. Please note, that I will not fix any security-related issues, as
   the program was never intended to be run inside client or server software connected to an untrusted network.

Q: I still have a question.
A: You can try mailing me [3], however depending on my current workload it can take a while before I am able to
   answer. Please note that I do not provide support for the program at all, so please keep your question
   short.

Q: I want to use the hvset and div+ indicator, is sourcecode for those available as well?
A: Yes, look into utils sub directory. Compile the code (make). the "diversity" binary can calculate both the 
   div+ and div* indicator. The hvset binary computes the hvset value for msdtlz2, mszdt1 and mszdt2. If you
   want to create similar test problems, you have to extend the hvset program to support your mswhatever.


References:
[1] Chair of algorithm engineering, TU-Dortmund: http://ls11-www.cs.uni-dortmund.de
[2] My homepage: http://www.creatience.de
[3] My mailadress*: matthias.jung@tu-dortmund.de
[4] Website of P. N. Suganthan: http://www3.ntu.edu.sg/home/EPNSugan/

* I still use a mailaddress from TU Dortmund but I am no longer affilated with the
university in any way. I only graduated there.
