Examining the ideas presented here:
•Factor DEX into AT/PA. OK, that's possible, but it doesn't seem logical. If the primary reason for keeping the two stats separate is the fact that fine motor coordination is not the same as full-body coordination, then it should be pretty clear that fine motor coordination does nothing to help you dodge a blow or swing a sword. Having nimble fingers does not give you an increased ability to dodge. The GURPS advantage "Fine Motor Coordination" outlines this idea pretty well--it's a very inexpensive ability that gives you +1/level on all fine motor skills. This gives you a bonus to painting or fixing an engine, but not to swinging a sword or avoiding a blow.
•Use DEX in techs for precise attacks. This has been done before. Strictly speaking, it can be done again, in the revised system--my new sheet for James is a good example. Why should a rogue character have to pay tech points every attack just to approach the attack accuracy of a character who chose STR and AGI over DEX?
•Use DEX in AT/PA, but only for certain light weapons. This is reasonable. The problem with it that I can see is that that revision still does not make DEX and AGI equally valuable, because a tricky, fast rougue who uses his points on DEX will be sacrificing initiative bonuses (AGI), even if he's still getting AT/PA bonuses. While lots of skills can utilize DEX, the character is still getting an overall better benefit if he only has one stat to worry about.
On the subject of "stats to worry about," we've discussed this before, and the basic idea is that character "archetypes" each need different stats. Generally:
•Fighters need COU, STR, AGI, STA, and maybe INT or WIS if they want skill points.
•Mages need INT, WIS, MAG, COU (mostly to resist enemy spells and potentially raise their AT/PA for defense), and STA doesn't hurt, either.
•Rogues need COU, AGI, DEX, INT, and WIS, because they need to keep their skill points per level high. They might also benefit from STA (increased TP).
Oddly enough, RPGWW's character list, from my experiences, has fewer rogues than any other archetype, but that's not really my point. In order to be effective, a rogue-type really does need to invest in an extra stat. The fighter can get away with 4 if he doesn't need a lot of skills (Choark is a good example). The mage can get away with 4 if he doesn't mind being a little fragile and not having a large TP reserve. The rogue absolutely needs 5 if he's going to do his job properly, and with STA now contributing to TP and tech combinations being a definite requirement for the rogue to hold his own, that fringe statistic is a bit more valuable than before.
Before anyone attempts to refute the idea that a rogue might need WIS, keep in mind that without a few points in WIS, a rogue is only going to be able to improve three of his skills (by one point, at that) every level. For a character that is likely to have a lot of skills, this is a serious drawback. COU, I feel, is necessary for all characters--which is intentional, though I have seen people play very interesting characters without it--but the assumption being made is that the player wants to create a character that is at least somewhat combat-capable. If you want to create a character that's a terrible fighter, that's easy, but not everyone wants to do that. As a result, the system should allow for players to create characters that will be combat-useful and have other skills (flexible non-combat magic, "rogue" skills, knowledge skills, etc) as well to contribute to a party.
On the subject of missile weapons, I can't say I actually see them get used much. Clearly, DEX factors heavily into missile weapon accuracy--but obviously, the drawbacks of focusing on DEX (such as decreased AT/PA and initiative) are not worth the increased missile accuracy in most cases.
Pervy, it seems, is trying to out-ludicrous me at my own game, especially since he earlier stated that he'd rather have the fourteen-stat system I facetiously proposed earlier, but the idea of four-stat Philsys (plus MAG) actually appeals to me in some way. However, four-stat Philsys would really require a total overhaul of the rules and formulae that have been laid out, a complete rebalancing, and a massive character sheet conversion that would basically involve starting over and re-doing every character created, likely somewhat differently from before. It's a very interesting idea, but seeing how much resistance there evidently is to one relatively minor change, I don't foresee GURPS-Philsys being a viable option at this time. A very interesting suggestion, however.
While we can hypothetically "do a lot more to make DEX useful," nothing will make it "more useful" than simply merging the two stats together into one, single, unified stat. <p>
<div style="text-align:center">
</div>
</p>