The logo at the top is a bit squashed.

Is it possible to colour-code in the layout at which neurons the spike detector and voltmeter are plugged in?

Is there a particular reason that I have to go through "apply" and "save" to make changes effective?

The order of the spike detector is "highest number on top". Are the voltmeters in a different order? Again, small numbers to indicate neuron # would help.

It would be really useful if the layout wasn't reset ...

Yeah, but the problem is that you can get smth stuck (for example you can't move the "comment:" once it changed location) and the only way to undo that is reload the page.

At the moment, you can move around the neurons and change the layout to something more intuitive. However, when you run a new simulation, the layout is reset to the default.

It would be really useful if the layout wasn't reset ...

Just a little design bug: if you write "1-2-3" instead of "1,2,3" for targets (as in the example), it shows an error message "enter only number." and this message stays until you save/reload the page.

If I hook the same noise generator to several neurons - does everybody see exactly the same copy of the noise? Same for the Poisson generator?

rotter
For the Poisson generator I would expect to be able to specify both the rate and the weight...

This is possible (see my post above). One has to click on 'more' to bring up the extra fields, including mean firing rate ...

After playing with different parameters, neither Poisson input nor noise input is now having any effect on the neurons... I will now reset to the factory settings...

jarvis
Although we have the connectivity matrix, it's difficult to view this on the network figure as many connections terminate/begin at the same place. Would it be possible to designate one side as input, and the other output, so at least the user can determine which neuron is pre/post for a connection?

Ah. discovered that the network does show input/outputs better, but only if you mess around with the neuron locations. Then, I modify my suggestion to "choose the starting locations for neurons better". It's also now a more trivial piece of feedback. Ahem.

Therefore, I modify the suggestion to: color coding the neurons themselves.

For the Poisson generator I would expect to be able to specify both the rate and the weight...

maybe of interest to you the nuSPIC team:
http://iqr.sourceforge.net/?file=kop1.php