The next thing I'm looking at is the skill system. While I really appreciate the d6 method, I was tinkering with a d12 method to make it more granular and to allow the specialist to improve somewhat in many skills rather than just two. This works rather well, but my final desire was to allow modifications to the various skills based upon the ability score modifiers (+/- 1 to 3) of the characters.
Looking at that, adjustments by d6 pips are 16.67% each, while d12 pips are 8.34%. This of course makes anyone with a score of 4 or less have a 0% chance to succeed in any given skill attempt, while those with 16 or higher end up being supermen. As much as I eschew d20 type mechanics for skills (I have no desire to model the 3.x system), I like the idea of using a single die and not get bogged down in the minutia of % skills - although I like them, in practice runs of my homebrew game I found calculating these to slow the game down far too much.
That said, I gave in and decided to look at converting the skill system to a d20 "roll low" method. While it's fine and dandy to allow non-specialist characters to keep the 1 in 6 (16.67% chance), those with ability scores outside the 9-12 "dead zone" should have their modifiers come into play IMO. As such, one has the choice of giving a flat 3 in 20 (15%) or 4 in 20 (20%) chance, which is modified by the relevant ability score modifier.
It is interesting that the Lotfp method of giving specialists 2 d6 pips to assign (an overall 33.34% increase in skills per level) is rather close to the 30 % points given to a thief each level in the 2e system. Assuming that an increase in sneak attack/backstab multiplier is worth 16.67 % "points", after 4 levels the "extra" 3.34% per level = 13.36%, which is pretty close. (I wonder if Jim ran these numbers in his design - I would think so).
Converting to a d20 pip method, it's not too much of a stretch to have the cost of a sneak attack/backstab multiplier = 3 or 4 d20 pips for a "cost" of 15-20%. I've been thinking of allowing specialist in this possible d20 pip method to sink pips into the sneak attack category which would remain dormant until they allotted 3 or 4 over time.
Considering all that, one could allow 7-8 d20 skill pips per level for a total of 35-40% "points" to be spent per level by the specialist (the difference simply being how many skills you wish the specialist to increase in).
On that note, out of the 10 named skills (not including the 2 "blank" skills allowed for DM tweaking), the one that sticks out as odd to me is the open doors one. I've decided to keep that one a d6 roll and not allowing specialists to increase in that, and at the same time converting it to the more familiar "basic" method of a base 2 in 6 chance modified by the STR modifier. those with a -2 from a STR of 4-5 have only a 1 in 12 chance of opening a stuck door, while those with a STR of 3 either have a 1 in 20 chance, or must roll a 1 on two d12s, or perhaps a 1 on a d12 and then on a d6. (it should be a miracle if they can open it)!
However, actual "breaking an entering" in terms of using various tools to break into a tomb could simply be washed into the "tinkering" skill, which already includes open locks, remove traps, & set traps.
In terms of what ability score would modify what skill, I think the obvious would be DEX for sleight of hand, & stealth & STR for climbing. Tinker is a bit harder, and I could see that the average of INT and DEX might affect that (perhaps rounding up). Likewise, for search, one could make equal cases for INT and WIS. My gut level "make the game cool" feeling is to allow the higher modifier to prevail, but those with high INT would actually notice the actual issue, whereas those with high WIS would simply be told "you have a feeling that something isn't right". Bushcraft and Architecture would seem to be mostly INT based, but I could see arguments otherwise.
In keeping with the vibe of the D&D thief, I've been looking at having the 6 arguably "thief skills" being sleight of hand, tinker, search, stealth, climb and sneak attack. Non specialists would start out with 3 of 20 in all with none in sneak attack, whereas specialists would be given 16 d20 pips (80%) to assign (given that S/A costs 4 pips), which is of course a bit more generous than the 66.67% that 4 d6 pips @ 1st level allows, but I don't think it's an insane difference. Assuming the specialist spreads them around, they could end up with x2 sneak attack, and allows them to place 3 pips in two of the remaining 5 thief type skills, and 2 pips in three others. This would bring them to 25-30% success rates in these categories, and that's before considering any bonuses or penalties they might get from ability score modifiers.
Of course, one could make a good case for making baseline 4 pips (20% in all categories), but one is free to tweak. After all that the specialist could have 8 d20 pips per level and assign them at will. I don't think it's too crazy to allow the other classes one or two d20 pips per level to increase somewhat in some areas, - once again, not wanting to mimic d20 type games, but a 5% increase somewhere isn't the end of the world.
Finally, i think I shall leave the languages where it is as a d6 method, and not allow the specialist to improve in it. however, one of the two "free" skill categories would be well served to become a read magic/use magic item category (IMO) to kind of give that flavor that these sneaky bastards might become somewhat skilled in the dangerous practice of trying to use odd artifacts that they found (generally with disastrous results for sure)!