Grog said:
By this reasoning, there's no point in ever making any changes to the system.
It depends on what you are trying to change. I hate analogies, but I'll try one for the sake of illustration.
I think that gaming technology is just technology. It's a new technology, and its evolving as we better understand the technology, but eventually it will become a refined technology. At that point, its going to be running up against real world limitations.
For example, architectural technology is reasonably robust at this point. I can argue that there is no such thing as a perfect building, because there isn't. There are always tradeoffs in construction. If someone tells me that they can produce a building that sturdier, more attractive, and faster and cheaper to build than existing construction techniques, I'm going to be really skeptical. That doesn't mean that I think existing techniques can't be improved, just that I doubt that you've got something that revolutionary to do so given how long we've been building buildings.
I think we could all agree that improvements could be made in the CR/EL system. I've previously been a part of discussions like that. What I've never claimed and what I'm skeptical of is that fixes to the mechanics of the CR/EL system - the mechanics Mearls chooses to focus on in the article - would make it easier to run large groups of mooks against high level characters and still challenge them, while simultaneously running large groups of mooks against low level characters and not risking a TPK. First level characters aren't threated by 20 goblins because of the CR/EL system. They are threated by 20 goblins because they are 1st level characters. All the CR/EL system attempts to do is systematically document that.
First, I'm extremely skeptical that any algorithm can accurately document how big of a threat X monsters are to Y PC's in Z situation.
Secondly, when Mearls hints at the real mechanical change that might allow you to send 20 goblins against 1st level PC's as a normal encounter, he brings up an extremely gamist idea - 'minions' - an idea that you are equally skeptical of. I fully agree with you that if I want a weaker monster, I'd just like to use a weaker monster. Why I need a vague gamist concept like 'minion' to balance a monster against PC's I'm not sure. It might superficially sound ok for orcs to be mooks, but then its a concept of utility only at very low levels of play. If the concept scales, then its going to create really weird situations like 'minion' ogres, 'minion' fire giants, 'minion' priests of Set, 'minion' 10th level fighters, 'minion' dragons, etc. depending on what BBEG has as minions and what the level of the PC's is. One guys minion is another guys boss. It's another concept like 'encounter' that is uncomfortably vague and gamist. I actually fear all these gamist concepts will be so vague outside of the narrow context of the design, that they'll be more constraining on encounter design than a more universal simulationist approach would be.
We seem to be moving back to '0 level fighters' and 'turns' and other vague gamist concepts (remember when you had different scales depending whether you were inside or outside) all of which I was glad to move away from because they had such a narrow application and created so many problems when applied outside of that narrow application. These gamist approaches don't create flexibility - they pigeonhole - and so far Mearls is not showing that they will in turn 'streamline' the rules. I think there is a good reason for that. No game system in the industries history has ever been successful by moving from high detail to low detail. 'Rules Lite' systems are rarely successful, both because players become uncomfortable with them and the economics of publishing work against them. So claiming we are going all simple and streamlined in this edition requires me to fail a save to disbelieve check right off the bat.
And yes, I know that as a DM, I could wing it and build a big encounter for 3E. But that doesn't change the fact that working around a broken system is an unnecessary pain in the ass for a lot of people, and trying to change the system so it works better is a worthy design goal (though it remains to be seen how well it will actually work, of course).
So anyway, yes, fixing the system for assigning XP should be high on thier design goals, but doing so won't solve any of the problems of encounter design he focuses on in his article. If he wants to convince me that 20 goblins vs. 1st level PC's is no longer a problem, he's going to have to do something that convinces me that he hasn't just created another problem someplace else.