Marcel Stimberg

Hi. Unfortunately Brian does not currently have mechanism for continuous interactions between neurons with a delay (it would need an implementation of #467 first). There are some workarounds, e.g. if...

Yes, unfortunately numpy does not always handle subclasses of `ndarray` (such as our `Quantity` class) correctly. We have some documentation on what functions are safe to use and which are...

Thanks for the feedback :) What do you think about the question of `EquationTemplate` vs. `Equations` (and same for expressions)? I.e., we could also decide against introducing new classes and...

I worked a bit more on this, and I think I am now quite happy with the general approach (remember that there's still a lot of work to do to...

> @mstimberg Could you please give us an example of how to resolve a name clash? For example, if `iklt` in Rothman&Manis model has standard `m` and `h` instead of...

I've worked on this a bit more and the basic implementation is now mostly finished. It will now look at the dependencies between substitutions and replace things in the correct...

> Thinking about two questions above, it _seems_ that `eqs = Equations('dx/dt = ({x}/mV-x)/tau : 1', x='v')` should be OK, just extremely confusing. Brian can(must) flag it with warning, but...

Thanks for the feedback @thesamovar , all very reasonable suggestions. Let me start with the easiest point > I think `dx/dt=-x/tau_{x}` should probably lead to an error. What I had...

About the questions regarding the summing and the "primary" variable. I also see the danger of hiding away details here, but at the same time we want to be concise...

I see your point, it would indeed be a possibility to make local variables the default when inserting an equation into another equation. I feel that this would limit things...