The skills in 1st ed. were much like the skills in 3.5; each class has a set number of skill points per level which is increased by an Intelligence modifier, and these skill points were spent on skills relevant to the class. There was room for customisation, room to breathe if a character, say a wizard or some such class, was on horseback for a long stretch of a journey. Even though the DC to ride a horse normally is only 5 our hero might not have the skill points to do it consistantly, and as a result may fall off from time to time. Certainly, if the party is ambushed they might end up doing a Gimli-son-of-Gloin and face planting into the mud. Thus, our Wizard might drop a few points into Ride when next he levels up, a symbolic feature suggesting his horsemanship is increasing over the course of his adventure.
One aspect of skills I disliked in 4th edition D&D was the distillation of the skills list. In retrospect I understand why it was done; the length of skills in 3.5 was cumbersome and colossal, and it meant that certain specific skills would be left by the wayside and ignored. Use Magical Device is a good example, as it can be used to tinker with magical requirements, disable traps, use artefacts and other awesome stuff. The 4e skill list removed something from character creation, and I sought to address in this project.
When Providence was first conceived in strength, the list was compiled from 3.5 with a few adjustments, the biggest of which was the addition of Endurance and Perception, the removal of Spot, Balance, Search and similar skills. As time has ground on and campaigns have been tested, it occors to me that the new edition doesn't fit a skill list very well.
The old skills points worked because you had a large selection of skills to choose from and a few points to spend every at every new level. Growth was fluid and natural, and diversifying your skills added a bit of neat customisation. With the new straight up bonuses from Character Abilities, this element of customisation doesn't exist anymore.
Each skill is represented evenly, there are none that have been ignored or neglected. However, unless a character chooses a talent that increases a skill, 2.0 skills tend not to grow. The exception to this is the increase to ability scores, and even then the increases are negligible.
This change in dynamic has forced a change to the meta game as well. In the current testing campaign, skill check difficulty has had to slow significantly. The skills, the strength, the quickness of characters has not grown quickly, and in two cases, not at all.
As 4th level characters, I feel a simple locked wooden door with metal bracing shouldn't be so much of a problem. Give it a trap, and it's a nice little road bump in a dungeon. However, whereas a 1st ed. Character's Disable Device skill (we assume) has risen by 4 at this point, there's no guarantee that a 2nd ed character's score has risen in any way. The trapped door becomes even more difficult, and unless the heroes can think of a way to bypass it, they may be forced to walk a road they don't want to. Alternatives, such a beating the door down and taking the trap to the face, become an easy fix.
I wouldn't expect this door to be a problem forever. At the end of the day, I couldn't beat down as door like that, and I'm not dexterous enough to even consider disabling its defences, but our heroes can.
However, there is a danger of our heroes, when they reach their maximum level that same level where they're supposed to challenge gods and knock on the doors of greater demons, will still have that average Disable Device ability.
For a Rogue esque character, the chances are slim. The trend to put Ability score points into DEX almost guarantee Disable Device growth. If we assume this god-tier rogue maxes out, then Disable Device will get a +10 bonus, in addition to any bonuses achieved through talents. If we assume our Rogue is something of a monkey, and chooses Sneak Attack, Quarry and Lightning Affinity, then he can gain a +6 bonus to Disable Device from starting abilities alone. With the Skill Tricks, Skill Focus and Thief talents, our Rogue can benefit from a +8 bonus. This totals a whopping +24 to Disable Device, a minimum of 25 if we include the d20 roll. A 25 is enough to auto-success most of the complicated traps ever seen in dungeons, with an average of 33-36 to deal with the endgame traps such as acid floods or accidental Spheres of Annihilation.
And here we reach a problem. If we assume the player is experienced enough to recognise the need for Talents that increase or enhance skills, he must also realised that, in order to get to that level, he has wasted three of his talents and consumed his skill bonuses to the exclusion of everything else. The rogue will never be as good at sneaking, stealing, backstabbing, sniping, conning or looking good than another rogue who has generalised. One might argue that this is the point of specialising in something, but consider that the rogue's skill with any of the above might not have improved much since 1st level. If a Providence character specialises, I would want it to be at the exclusion of a handful of things and not of everything else.
So, what to be done?
For starters, the skill list needs to be truncated. Not much, mind you. Although I love it with a passion, Use Magical Device will be rolled into Spellcraft and Knowledge (Nobility) will become one with Knowledge (History). I have no idea what to do with Professions. Probably drop them altogether, since they never get used, and I'm at a loss as to what to do with certain DEX based checks.
In terms of skill growth, I have no idea what to do. The variables we have to consider are thus; ability scores are going to go up as characters grow, increasing skills as they do so.
Skills are used in combat to perform actions against enemies and for allies. In 1st edition, this created characters who could auto-trip and auto-tumble boss monsters, due to the difference in size between a skill and a defence.
The difficulty of dungeons is variable, but generally has an upward curve. The DC to pick a lock will rise if it has been made and enchanted by dwarves, or go down if it's a rusty, dirty thing scrounged up by Kobolds.
DC modifiers tend to be realistic - a colossal reinforced iron door with defensive spells and trapped with an electric keyhole will require multiple high-end checks to disable or damage.
Types of people don't change, but their attitudes do. At 1st level a barbarian wouldn't be able to intimidate the tyrannical baron into lending his party aid, but their slick-tongued bard might be able to convince or connive resources from him. Only later, when that barbarian is known as a slayer of dragons and wrestler of ogres does his intimidation have any weight. Even then it will be a difficult sell to rattle a hardened and feared ruler. Thus, the skill check changes from 'Impossible' to 'Very Hard' (to the 30's).
I'm open to ideas. The goal is to have a skill system that rewards a player with scale and options, but doesn't overpower combat or fall beyond the realms of reality. Play your favorite tabletop games, see what they did, see what you would change.