So if I formed a contract with Saber and Berserker, Saber and Berseker would both exist sharing my magical energy, so their abilities would considerably decrease.
Yeah, not to mention that replenishing Bersekers magical energy would be a painful experience 
Actually, there are two types of decisions people often (unintentionally?) confuse.
First is when one decides which path to take. This is a kind of decisions called “choice” – a decision based on estimations of benefits from each alternative. For example, one might consider buying a car or a buying bicycle instead, or even keeping riding metro because it’s cheaper and more convenient that way.
To make a proper decision in such a case one needs to account everything he knows about all the possible alternatives and their consequences. The choice with the most beneficial consequences will be the best choice. By failing to account every known consequence or constraining oneself to only a subset of known consequences, one may end up choosing the bad or even the worst.
In such a manner, to make a proper “proof” decision, one needs to suggest a logical explanation for a chosen alternative, and then, no matter how convincing the explanation might seem at it’s own, disprove any arising counterarguments.
Basically, when you’re dealing with a “choice”-type decision, you have a pair of scales. Every pro and contra you bring on goes to the appropriate scale, shifting the pointer. In the end, some of these contras might remain unresolved, but as long as they’re outweighted by pros, it’s all right.
On the contrary, with “proof”-type decisions you’re at a trigger gate. You need to be perfectly clear to pass. Weight does not matter; as long as you have even one discrepancy unresolved, it’s not good enough. A single contradiction is enough to sink your proof.
The thing is, people often confuse these two. They end up “proving” their statements by bringing up a shitload of pros (which does not matter as long as they ignore contras) in a hope that outweighting a contradiction would somehow diminish it. Or they do argue for their choice by stating that “it’s the only choice that allows us X” and completely ignoring the fact that it is also the choice that denies us Y, Z, A, B, C, et cetera.
Don’t do that.
To further train myself in this skill, I’ll write some of the posts in english from now on. If you spot an error – and I mean any error, not only grammatical errors, even bad phrasing will do, – please be sure to correct me.
Now to the topic…
Responsibility comes uninvited. It’s not a thing that can be measured and limited to any desirable extent. You cannot throw it off by saying “I don’t want to be responsible for this so don’t rely on me” – and that’s where life differs from the world of software licenses (in which similar claims are pretty usual things).
In real life if somebody relies on you, it’s already too late to state your “limitation of liabilities”. If you’re being relied on, that means you’ve already done enough to provoke this; so by saying “hey, don’t rely on me” you’re basically saying “looks like I’m too scared to face the consequences of my previous actions, but I want to look like a responsible guy nevertheless”. And that’s exactly what’s the irresponsibility is. Even if you did everything you could to make people not to rely on you – with no success though, – you still can’t say that “it’s not my problem anymore if I let somebody down, because they were warned”. That is the real irresponsibility.
If you really don’t want anybody to depend on you, don’t do anything that might make them think they could. And if you did then go all the way and support them till the end because it’s already your problem from the time you’ve first messed with it.
Your actions and not your words define your responsibility.
(a bit more)Maybe my point will be easier to understand, if I introduce some explicit example. The case that brought the whole topic into my mind.
As some of us know, Raymond Chen is a popular Microsoft blogger who writes about Windows history, system internals and backward compability stuff. He’s easily one of the most popular and widely-known Windows-related bloggers out there. His articles are always superb, and constitute, to my opinion, an excellent reading for everybody from system administrators to kernel programmers.
But again and again Chen continues to state that the whole blog thing is but a mere storytelling for fun and nobody should rely on any information there.
Well, I see his point, some of the topics concerned are out of Chen’s jurisdiction and far beyond his profile; he can only make guesses in these, so the information in his blog can possibly be unreliable (although, as far as I know, nothing came up until now). But then again, he’s the blogger who gained so much trust and popularity; all of this was not given to him in an instant, – he earned people’s trust by writing good, reliable and informative posts – and now he wants his readers not to believe him out of blue? But if he really wanted not to be depended on, he’d just never write anything potentially misleading. He’d never publish a book made of his blog posts!
And even if he really did not want anybody to depend on him, it does not matter. He still can’t get off with that lame excuse. Of course nobody’s going to blame him for his mistakes (it’s his own blog, after all), if such are to be found, but by taking a role one takes the responsibility coming with that role. It’s impossible to separate one from another.