Francesco Casella

@thorade, I would suggest that you make a proper release with the previous version using MSL 3.2.3, so people can still use it if needed. Then, I'd also make one...

Sounds good. For the old release (which I recommend to make) you should add ``` annotation(version = "1.0.0"); ``` to the top package, so that Modelica package managers are aware...

Good argument. I guess we should sort this out explicitly in the spec, but that is a separate issue. So the MSL 4 version should have ``` annotation(uses(Modelica(version="4.0.0")), version="2.0.0", conversion(noneFromVersion="1.0.0"));`

I'm not 100% sure either, but I believe the interfaces of Modelica.Media (which is all HelmholtzMedia needs, besides SI units) haven't really changed at all, so I think this should...

> Typically, this consideration is the domain of package managers (like pip, conda, apt, etc), as you are probably aware, and one tool that is used is SAT solvers. Sure....

@thorade, we're now testing the conversion of HelmholtzMedia to MSL 4.0.0, using OpenModelica. As reported [here](https://github.com/OpenModelica/OpenModelica/issues/7683#issuecomment-945647556), we are getting a large number of failures because you are using `Modelica.Media.Examples.Tests.Components.PartialTestModel` that...

@dietmarw this criterion (thou shalt not extend from examples) is good for me, but are we declaring it explicitly somewhere?

@mahge, do you think we can exploit such a fine-grained parallelism? I'm afraid the overhead could kill any potential speedup.

> I can not say much without looking at it further. However, the design and implementation is intended to be used for fine-grained parallelism, i.e., at equation level instead of...

Thank you Matthis! We plan to use your model for a library on which we'll be going to publish extensively. Can you point me out the correct citations for your...