A couple of things in the blog rang warning bells for me - it suggests they're still not grappling with some fundamental issues around encounter/challenge design.
One was this:
Rather than improvise and come up with something unexpected, I found, in my own gaming experiences, players combed the skill lists on the character sheet to determine what they could and couldn’t do when presented with a challenge. And if the character didn’t have mastery with the given skill, the player, more often than not, chose not to do anything since failure could and did (with skill challenges) adversely affect the group.
So long as encounters are designed in such a way that
failure is not an option (and that is the traditional D&D approach), then players will always feel the pressure to bring their biggest numbers to bear. You don't make this pressure go away just by limiting the game to 6 numbers.
If you want the low-CHA, untrained dwarf fighter to make social skill checks, you have to create situations in which the consequences,
for the player, of not making the check are worse than the consequences of failing the check. This is easy enough to do, and their are plenty of RPGs that give advice on how to do this: HeroQuest, Burning Wheel etc. The simplest approach is to set up a situation in which (i) the PC will look like an idiot if s/he doesn't talk (and I'm assuming here that the player doesn't want his/her PC to look a fool), and (ii) if the check fails that
doesn't mean that the PC looked like a fool regardless, but rather that for whatever reason, the check failed to achieve what the PC (and player) hoped it would - eg despite the PC's entreaties, the duke can't agree because he feels bound by his earlier promise to the PC's rival. (Where that promise, or the duke's degree of commitment to it, is retconned in by the GM as part of the process of action resolution.)
Here is the other passage that makes me concerned:
Even if you ask the DM if your knowledge of history reveals additional information, ultimately the DM decides whether this skill is useful or not since the DM has to have historical information on hand so you can use the skill, has to make up something that might later be invalidated, or just says no.
If they're thinking like this, no wonder they can't make History skill worthwhile! The key to making these sorts of skills worthwhile is to take a more relaxed and free-flowing approach to worldbuilding. The GM (or perhaps the player) is
expected to make things up, and part of the GM's job is
making sure that these aren't subsequently invalidated (eg by keeping notes).
A proper system of linked skill checks, or augments, can also help a skill like History. On its own it won't often resolve a situation, but it can be a steady source of bonuses to other, more definitive checks (knowing the history and culture of the minotaur empire, I have a better sense of where their secret caches might be; knowing the history of the duke's family, I have a better path to befriending him; etc).
Unless some of these fundamental issues are tackled, I don't think the two non-combat pillars can be made as central to the game as combat (or, at least, as central to the
mechanics of the game - and once you're using freeform resolution, all this stuff about class balance, and classes being balanced via their varied contributions to the three pillars, becomes irrelevant).
Love the two trait examples, Language and Workshop. They aren't the same power level, they aren't really even quite the same category of trait.
Who cares? Both allow a player to create the desired PC. I also love the idea that some things have to be worked out on a case-by-case basis. Rules aren't going to cover everything, so it's better to admit that in the design explicitly.
I can already see Workshop creating headaches in play.
Suppose a PC captures an enemy alchemist's workshop. Now s/he has the benefit of the Workshop trait without having spent any PC build resources on it. In which case, can another PC spend a month studying with a teacher to get the benefit of the Language workshop without spending resources on it?
Or will the rules say, in these circumstances the workshop is lost by the time the next session begins, unless the player spends the resources to cement the gain? (OGL Conan, HeroWars/Quest, and The Dying Earth all use this sort of "lose it between sessions" mechanic.)
I'm not saying that these sorts of issues are fatal. There are many RPGs that solve them fine. But D&D has never really been one of those RPGs.
Won't that always be true though, in any skill system? It's not like Spot, Use Magic Device, and Tumble were ever balanced with Decipher Script, Use Rope, and Knowledge (Local).
In a system where skills can be opened without spending highly limited PC building resources, this is not a problem - or, at least, nothing like the sort of problem that it is for D&D, with its very limited skill point/feat slot mechanics.
Examples that I have in mind include Runequest, Burning Wheel and to a slightly lesser extent Rolemaster.
