Welcome to the Playpen, our space for ferrety banter and whimsical snippets of things that aren't quite long enough for articles (although they might be) but that caught your eye anyway.
On the other hand, I think it's more likely that skills won't be optional - but the precise skill system you use might be, if you see what I mean. So everyone gets an allotment of skill points whatever happens, but depending on precisely which skill system you're working with you might be using them to buy nonweapon proficiencies of a fairly simple "either you have 'em or you don't" model, or to buy levels in a number of broad skills covering a wide range of pursuits, or to buy levels in a bunch of very granular skills geared mainly towards dungeoneering.
The skills for dungeoneering are certainly finer-grained than Profession: Butler. As a matter of interest, assuming you were going to use the 3E skill system for some reason, what non-dungeon skills would you like to see? From things you've said previously I get the impression you wouldn't want to get into "social combat", for example (correct me if I'm wrong). Just a finer breakdown of existing ones, or something else?
However! My point was allegedly that 3E wasn't the first complicated skill system out there, and I think that stands regardless of whether NWPs are intrinsic or not. When they wanted to discuss how you did stuff that wasn't combat, the 2E devs offered a complicated set of rules for "skills" and their use. Certainly my impression of reading the 2E rulebook was that their take was: "here are the NWP rules, but if they look like a hassle, you could wing it using broad skill categories, or just use player knowledge", rather than "BTW here are some other rules you could find useful but we don't expect you to use them really". They just were careful not to make them compulsory. However, someone found NWPs sufficiently useful and well-used to make them core for the next edition. I suspect it's one of those examples of Ruleset Creep you sometimes mention.
*Blinks a few times*
*Goes back to playing Tunnels and Trolls*
At the risk of sounding like a complete pedantic dick, I actually think there's a major difference between NWPs and a Skill System, even though Nonweapon Proficiencies were designed to provide rules for things which would normally be considered "skills" in another game.
Crucially, I think the big difference is that Nonweapon Proficiencies, because they were optional, were not an assumed part of the play experience and that had quite profound consequences for the way they wound up working in practice. "Gets lots of nonweapon proficiencies" wasn't a class feature the way "has lots of Skill Points" is for Rogues in 3.X, for example. By the same token, 3.X (and even more so 4E) carried with it the assumption that "using skills" was supposed to be a core feature of gameplay, which by extension meant that the skill list became a kind of short list of the sorts of things players were expected to do in the game.
This is very much one of those "bug/feature" things but it's one that is often misunderstood. When 3rd Edition was released, a friend of mine complained that adding a skill system made the game *more* dungeoncrawl focused, because most of the skills were things which were specifically useful in a dungeon.
Some games use skill systems to good effect to foster a sense of the game (Cthulhu does this quite well with its huge list of prosaic, ordinary-person skills) but I always found the 3.X skill system limiting because it made such narrow assumptions about the sorts of things I would want to do in a Fantasy RPG.
Core 3E doesn't have any optional subsets, just lots of individual variant rules. I'm not sure I'm bothered by the lack of major subsystems that I have to decide on; I like things pretty simple TBH and since the 2E options are basically "here are some rules" or "make it up yourself" I'll take the former. But you're right, 3E & (I think) 4E don't explicitly highlight that entire subsystems can be dropped for homebrew ones, though in 3E at least it's possible.
Basically the later editions are more polished-and-boxed for you. They assume that in buying the game, you're after a ruleset that tries to explicitly cover everything and has established defaults; the older ones offer suggestions for things that might suit you, but assume that a lot of the onus will be on the DM to choose subsystems and invent ad hoc "rules", not just to manage the world, devise adventures and handle NPCs and monsters.
I only play about once a year so 2nd ed works fine for me, but I think for regular play I would go for rationalised 2nd ed (which is more or less what the Baldur's Gate games seem to run off).
I initially picked 4E because it was new, appeared more accessible to wary newbies, and it's more forgiving than older rulesets. Over the course of play, I found the combat tedious, the power system narratively unconvincing, and the high-powered nature of the gameworld not to my taste. Also, I felt like I wasn't giving a fair introduction to RPGs: my group were great, but the game emphasised "have you got X?" and its combat-heaviness distracted from RPing. I'd also started running Call of Cthulhu, with its sparse and flexible rules and complete lack of boards or tactics. So for my new game, I introduced people to 3.5E, with its straightforward combat and more narrative approach to situations. After one scenario, I'm in the process of switching them to Pathfinder, because I've come to believe that it'll streamline some clunky aspects of 3.5E, somewhat rebalances the classes, and better defines their differences. However, the sheer complexity of the character sheets (in both versions) is daunting for some players.
I think one way of looking at things is as a progression of options. 2E has quite specific class and race options, built in particular ways and with no mechanical way to adapt them. Splatbooks offer a load of alternative specific builds, but little individual customisation. On the other hand, this approach leaves a lot to the imagination because there aren't specific mechanical representations of things. 3E builds in lots more explicit customisation rules and a looser approach to archetypes, particularly in multiclassing. The rulesets here let you build almost any character you want, but there's a constant supply of new and more specific classes, feats and races to get exactly the right feel, and of course it can't all be balanced, so some builds end up much stronger or weaker than the game expects. 4E pushes this a different way: powers turn "I attack" into an array of different attack powers, technically giving you more options. However, this itself tends to subtly lock down your options because without a suitable power, how do you do something?
Incidentally: Use Rope as a skill isn't that excessive, given that climbing, tying up, untying and balancing are fairly important to adventurers (what about Track and Forgery?). A glance at my 2E rulebook displays a vast range of Nonweapon Proficiencies ranging from Direction Sense, Rope Use (!), Astrology and Tightrope Walking to Ancient Languages and Running, with individual class-related modifiers.
Spiders in real life aren't that bad.
Perhaps, but they don't win many points in my book with that whole "inject their prey with paralyzing venom that can also digest them from the inside out" thing. But, yes, it's a phobia on my part. (Curiously, I don't have any problems with scorpions or crustaceans, aside from coconut crabs, which I still think look like they should be tearing their way out of someone's chest cavity. Snails freak me out, but that's Junji Ito's fault.)
Still, I know they're valuable to the ecosystem and all that, so I don't actively hunt them down or bother them if they're small and/or are going to scurry out of my sight immediately.
On top of that, having an explicit skill system is seen by some as a step along the way to 4E's "pick an option from a list" playing style - the argument is that when everything you can do is codified by a skill (with a level of granularity scaling down to "Use Rope", for crying out loud) people start thinking in terms of the list rather than thinking outside of the box.
This isn't a problem which is necessarily new to 3E; magic, in particular, has always suffered from it, as this blog post points out: when you've got a prescribed list of a bunch of things that work, people are going to be reluctant to branch out into headachey and deliberately difficult stuff which is less structured even if it does have rulebook support, and many won't even consider the possibility of exploring territory which is completely off the map. The difference is that what used to just be the case for magic ended up becoming the case for everyone to a certain extent in 3E, with skills and feat lists standing in for spell lists, and of course in 4E there is literally no difference between a spell and a martial feat except the person who's doing it. A lot of the stuff the Wizards people have been saying after this announcement (and leading up to it) seems to suggest they really do want the next edition to encourage more thinking outside of the box, so I'm feeling hopeful on that score.
To be fair, it does seem to be 4E which attracts the most bile - there's enough disaffected 3E fans who never made the transition for Paizo to viably market Pathfinder to them, and by all accounts Pathfinder's very commercially successful. The 2E-and-earlier crowd have their various retro-clones to play with but that's a cottage industry compared to Pathfinder. But if you go looking in 3E you'll find a lot of the new ideas kicked off trends which were taken to an extreme in 4E.
I guess one way they could do it is to make the new system highly, highly modular - make the core mechanics extremely simple and incredibly sparse, and provide heaps of optional add-ons.
Oh hey, turns out that's precisely what they're doing.
I'm a bear oh yeeeeeeeeeah.
I think that WotC have very correctly identified that a fractured fanbase is D&D's biggest problem at the moment - its big strength was never that it was particularly good at any one aspect of the tabletop RPG experience, but that before 4E it was a broad enough game to accommodate a bunch of people's preference, so whilst 4E was clearly a much better game for people really into miniatures and strategy it was also clearly going to put off people who weren't into either of those things.
The big difficulty's going to be trying to incorporate enough people's ideas that there's a general feeling the consultation process made a difference, whilst at the same time avoiding becoming a massive, contradictory mess. Ultimately there's going to be some issues people feel really strongly about which Wizards will have to make a call on one way or another, and whatever they decide there'll be neckbeards screaming and crying and kicking game tables over; the trick's going to be making choices which ensure that it's just a few big babies who get stroppy and stomp off and not significant sections of the fanbase, because then you have the fractured base problem all over again.
It's going to be a tall order. I guess one way they could do it is to make the new system highly, highly modular - make the core mechanics extremely simple and incredibly sparse, and provide heaps of optional add-ons. My group wants a game where we spend a lot of time tweaking and optimising our characters but where combat flows quickly, so we use the Advanced Character Gen and the Streamlined Combat optional rules; your group likes to roll up characters quickly and then spend ages over the fighting, so you use the Quickstart Character Gen and the Tactical Combat modules. Essentially each group would take the core mechanic and bolt on different flavours of subsystem to suit their tastes.
That would risk losing some of the cohesiveness, but then again it'd also acknowledge the fact that no two D&D groups have ever really played the game the same way in the first place. The real difficulty of such an approach would be making the game accessible to n00bs, but then you could flag up some of the modular system components as being particularly learner-friendly and design accordingly.
Either way, it'll be interesting to see how they do it.
So I'd be interested to hear what the other Ferretneurons think. A slightly glib starting point would be "work out what sort of game you want it to be". At the moment 2nd edition seems to have the classic lowish fantasy market with strong archetypes and social structure, Pathfinder is probably on top of loose flexible fantasy, and 4E is there for fans of tactical combat and cinematic fantasy who don't want too much realism. What do they actually think they can offer?
NB: okay, that's weird... I managed to read the article, just went back to recheck it, and now it's behind a paywall. Good timing there.
The flamewars over this are going to be legendary.
I'm sorry, your supply of what?
Bile, presumably. It's like regenerating health; you have an infinite supply, but it runs low, so you need to hide behind a wall and wait for it to recharge.
'Course, in my day, we had bile packs, and we had to ration our outrage accordingly. It made arguments more challenging, knowing you had to make your points while never being certain when the next dose of anger would keep you going to the next point. Honestly, the day regenerative bile when mainstream is the day rhetoric went to hell. "Faster and more immersive arguing", my ass. All it does is turn arguments into white-hot jets of rage which eliminates the highs and lows of traditional debate and eliminates all sense of personal risk from the arguer.