When I was musing the other day about skills it did occur to me that there might be some reasons for bonuses in the initial 10% that everybody has to accomplish anything extraordinary.
Then I read this over at the Land of Nod and thought that maybe bonuses based on ability scores*** is a pretty good idea. Use the standard bonus for higher ability scores in Daen Ral: 16=+1, 17=+2, 18=+3. Or, since I'm dealing with percentages, 16=+5%, 17=+10%, 18=+15%.
Now, this is really DM fiat territory here, though. How to determine if something is best linked to a an ability score. Is jumping across a chasm more strength or dexterity? Is running a long distance more strength or constitution? Is trying to tell a lie more charisma or wisdom or intelligence? So, you get the idea.
If, in desperation, a mage picked up a sword to defend himself (or, since this is heroic fantasy, to defend the innocent child and puppy that are trapped in Cthulhu's pantry) would he have some chance to actually use it? I'd say yes - 10% (probably MINUS 5% or 10% since he's likely to be a weakling). And 01% ALWAYS succeeds in my estimation. There's always SOME chance... But the mage is not a sword wielder. He wants to try it again, no bonus for having been successful once, since he's not training with the sword (and if he wants to do so, he becomes a FTR/MU and deals with the bonuses and penalties that entails).
If the Fighter tries to read a foreign language - 10% chance he'll work it out (I'd do some kind of scaling so that rolling 01 means he reads it all, 02 he gets 90% of it right, up to rolling that 10, just making it...but just kind of getting the gist of what it says - and, if pushed, I might have him do it per page or even per paragraph if it's key). And, yeah, higher INT gives him a better chance to do the reading...but I'd still use the actual roll to determine how much of the text he really "gets".
Hmmm, maybe that's the way to adjudicate all the skills. Rolling a 1 means complete success at the skill attempt - rolling a 10 or more means just making it (modifiers only boost chance for success - not successful-ness...um...is that a word?). So, the fighter who jumps across the chasm and rolls a 2 - made it, just slid a bit on the landing, turns around and yells, "That was easy." The thief tries the jump rolls a 23, no sweat because he's got an 18 DEX (that's what I would modify this jump with) so he makes it...but JUST BARELY . I'd narrate that it doesn't look like he's going to make it but that he reaches out as the edge of the cliff is falling past his eyes and just catches a hand on the ledge. What are you going to do now? (and so on - we'd then play out getting him up the ledge - depending on what he was carrying, whether or not they're being harried, etc.)
So...what about thieves? I guess an additional 5%-15% at character generation isn't a deal breaker - especially since they have 15 thief skills to divide it up among. I'd call it a one time only bonus, though - not something that they get to apply to every skill, either, though that's kind of arbitrary, isn't it? Would starting a thief at first level who has an 18 DEX wtith 25% in the Fine Skills, with a 17 STR starting with 20% in the Physical Skills or with a 16 INT starting with 15% in the Mental Skills really be a problem? After all, the thief is SUPPOSED to be good at this stuff, right? Maybe that doesn't bother me as much as I thought it would...
*** I have a lot riding on ability scores in this game - obviously the initial bonuses, maximum level is tied to ability score, going lower than 0hp is tied to CON, class abilities, spell abilities...off the top of my head. I've never seriously played with MIN/MAX-ers (well, in Junior High...and once after) but the temptation will really be there in this system that I'm putting together...