I'd like to see these teamwork boosting measures talked about more in a later beta when we have netcode and can see the effects online, but I really like the idea of something, because as it is now, the rook can chain a hammer slam on the end of a boulder roll and put the other two demis near death without much chance of defence, and with other stun abilities, you can be killed before you know it.

This is true.

I'd be tempted to go double and shorter for two reasons:

1) Frozen stuff loses flexibility and strength. It's brittile.

2) This makes permastuns considerably harder to pull of, which is definitely a good thing.

I know that. I'm saying that if the freezing time were very short, it would be very different. If the freeze lasted a second tops (and generally less time than the heavy attacks' animation), then getting in a hammerslam would be an epic feat of team coordination worthy of instapwn. It swings both ways, as you say. I think that if freezing could behave synergistically with heavy attacks in the way I describe, it would make it worth having even if it weren't a very effective stun, and would help make the two feel drastically different. I understand that it'd be devastating if executed perfectly, that's the entire point. It needs to be perfect to work, and if it isn't, it's a mediocre version of stun. It would provide a great opportunity for the best teams to prove how good they are and get a serious advantage for it, and would make the skills much more different than a stun that reduces damage. It doesn't even need to make hammer slam an instapwn, just a nasty attack that puts you on death's door that requires a lot of skill to pull off. Now that I've clearly said what I actually meant, what do you think?

I'm sorry I keep doing this.  

Dammit, you're right. However, timing everything into a window of opportunity shorter than the actual attack animations is very hard and considering it's a pair of skills, then it would be fair to be mediocre most of the time and awesome very rarely. Still, I think your solution is better.

I do too, it would make the worst abuses of stun nearly impossible. What about giving a recently stunned demigod a half second's immunity to stun so he gets a chance to do something? It's probably a terrible idea for some reason.

I agree with innociv. A half second ingame is a long time, when stuns are concerned.

I think I wrote something like that before and I totally agree.

I think there should be both active items/consumables which give temporal stun immunity and passive items which give either a chance of not getting stunned or reducing the length of a stun.

something like:

Shield of Sturdiness : you gain stun immunity, mana cost for activation 1000, mana drain per second 200, stops working once you deactivate it or you don't have mana left

"Potion of I don't Care" : you gain stun immunity for 5 Seconds, item is consumed on use.

"Gloves of TheBigOne" : stun length is reduced by 25%, they also look incredibly good.

yeah but what if stuns are just made less broken, and lots of them are replaced with snares, roots, blinds, silencing?

Do you have a consumable for each?  Or does the one item reduce the effects of all? 

I'd really like to see more skills that just snare, slow attack speed, make attacks miss, make it so you can't cast spells, or root in place rather than make it so you can't do anything at all all at once like stun does.

And/or most stuns reduced a lot.  GW has stuns, but they're called knockdowns, they're used but not spammed to highhell and keeping someone knocked down the entire game..

With stuns as strong as they are at the moment, introducing items to counter stuns would essentially force everyone to use them (because noone likes beeing a free kill), wasting eqipment slots and money. This also redueces the number of possible item-builds, since all of them have to include something to counter stuns.

I think weakening stuns, replacing them with other debuffs (-->see innocivs post) or implementing some kind of diminishing returns on all crowd control skills is the way to go here.

Actually, make items that resist one of the effects. So you could make a demi who will always be able to move at top speed, or one that will be able to attack when stunned 50% of the time, or cannot be silenced. How does that sound?

I agree with the Big One and was about o post something of the same kind. Consumables that allow stun immunity for a short period, or equipment that would grant 33% chance of resisting stun would be excellent. (You could still be frozen though)

Other possible solutions for the Stun problem:
- the Torchbearer stun seem to me most balanced, because it doesn't begin with a long duration, grow in potency with time. The problem is the possibility to chain the specific stun ability + frost nova.
The fact that an enemy is frozen could grant damage reduction, like in Titan Quest (like 50% damage reduction?). Thus, freezing enemies wouldn't be exactly like stunning them. I think innociv suggested it and I think it's a good idea.

- The boulder stun from the Rook is definitely the most annoying. It shouldn't stun for such a long time. 5 seconds feel like ages. You face two Rook with an ally and you are close enough : the first Rook stun you and your ally, close by and hammerslams you, the second rook does the same, you're both dead.
a. it shouldn't stun that long
b. the field of vision should be large enough to enable you to see the Rook launching its boulder in every situation
c. the boulder could be slower so that you can avoid it more easily if you pay attention.

It does, but the thing is, it takes a lot of the control out of the players' hands, and I'm not sure that there isn't a better way to achieve it.

This. A game that offers choice is only fun if the choices are choices. If it's too automatic, then most of the fun of making decisions goes away.