I'm really not convinced we're all talking about the same thing here. (yes Vinnie that is the example under discussion as I understand it). The illustration below is the rule that is being clarified... at least that's what I think is going on... maybe I'll learn something new...
In the illustration -- Three walls meeting at a common vertex circled below... The red circle represents a blocked LOS between the german in FF5 and the Brit in DD8. The blue circled set of walls do not block LOS and there is LOS between AA8 and CC5.
There is a case where the
addition of a wall magically creates a LOS where there was none previously, this is the case the OP was talking about as I understand it:
Imagine there are no walls at all along any hexsides common to EE6 FF6 EE7... los would be obviously clear. Now we add walls, one at a time:
EE7/EE6.... LOS blocked.... just that 'half' wall blocks the vertex along the LOS. (Of course it would be the same if only the FF6/EE7 were the wall section added.)
Ok... so now there is one wall section only at EE7/EE6 hexside and LOS is blocked... now we add back the EE6/FF6 wall along the hexspine of the LOS and
zap there is LOS. But add the third wall at EE7/FF6 and bap, LOS goes back to blocked. In my opinion this makes for consistent playable rules around wall LOS, and they have nothing to do with common sense -- LOS regarding these particular wall situations could be treated elsewise with equal commonsense -- but some rule however arbitrary must be consistently applied.