channel selector switch array - thoughts?

[analogical]

Kicking around the idea of making a switch array/buss selector card and I'm curious to know if you drive a fader bumped signal into multiple active submix buses, does it in anyway negatively effect the source signal in terms of load or sonic integrity? Is there a need to isolate the breakout paths with resistors, reduce crosstalk? Is that even an issue when the source signal is all the same and you are just splitting it out to various submix busses?

I'm almost ready to buy something like this:
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&Item=280398826710&Category=23785&_trkparms=algo%3DLVI%26its%3DI%26otn%3D2

and just try to reverse engineer it to learn.

If anyone has a schema of a simple channel selection switch array where the switches are latching, i'd love to take a gander.

Oh and lastly your opinion- there's always the option of getting interlocking switches that defeat each other when pressed- this would rule out running a signal to multiple buses simultaneously, but I'm ok with that if you think this is the superior way to keep the routing clean.

 

[analogical]

In the example below of a WBS shema it looks like switches would not need isolation.
Is this true in the case of running to active buses?

 

[analogical]

http://sound.westhost.com/project30a.htm

suggests using a rotary switch; I might just keep it simple and go that way.

Again, any experince you have when working with splitting outputs, feel free to chime in.

Thanks!