Crap. I have no time to write, but having talked so much smack I feel like I at least need to make a start at responding to these.
It has returned, precious! It has respondedses to our challenge!
(I appreciate it man, don't feel pressured but I find I am warming up to this conversation).
R.E. System Neutral mechanics and railroading.
Now, different systems may ascribe different probabilities to particular actions, so there can be a bias in a system toward one resolution or another. For that matter, if a DM just assigns a probability and rolls a d6, the probability may be different from any system, and for the same reason: anyone assigning a probability to a task without scientific study is just guessing based on their own perception of how the world works. But so what? It neither narrows nor broadens the plausible solutions, it just provides a procedure for resolving them.
Since the railroading tangent only came up because I was trying to illustrate that the system changes the way a system neutral module would be played in response to a comment by DP, and we are now cutting directly into the more relevant subject matter, I propose we drop the matter of railroading.
So I think we both agree that it is
possible to write a system neutral module and then convert that to the majority of roleplaying game systems (with some odd fringe cases like Storygames lacking the mechanical support to describe the action). What I asserted is that a System Neutral module is not going to work equally well or be equally compatible on every system and that the particular system you choose to convert it too is going to make a difference in how well it turns out.
What I think we disagree about is the degree to which a system neutral module is compatible with
any system. Most of the discussion seems to center around mechanical differences between something like Basic DnD and 5e, which is a comparatively small mechanical gap. I assert that as this mechanical gap increases, as well as the underlying assumptions, it becomes increasingly difficult to translate that module into the system while still retaining the same level of playability.
In the case of a system-neutral module, I would expand this hypothesis by stating that the key factor is probably baked in assumptions about scale and proportion ( e.g in Zweihander one high level guy can take on maybe four low level guys if they have the same armament while in DnD one 9th level fighter can probably kill himself a way through 20 orcs without breaking a sweat) or assumptions about the world (even an intrepid cleric can master a spell to determine whether someone is evil in oldschool dnd while in something like Dark Heresy reading minds is extremely difficult, dangerous and has all sorts of caveats attached to them).
In practice, I find that even in games with fewer abilities on the character sheet, players (and DMs) tend to think only about what the rules explicitly say they can do. In my experience, the freeform improvisation that is lauded as a feature of old school systems is actually a feature of old school DMs who have, over decades of play and talking to other enthusiasts, learned to encourage such play.
I don't know about your first game, but when I started playing D&D, fighters (for example) only had two options in combat: melee weapon or ranged weapon. Out of combat they had pretty much no options. It was only over many years of play that players started trying new things - and it took even longer for DMs to let those new things have any chance of success.
(AD&D 2e was my first exposure)
I find your hypothesis of the GM being an absolutely necessary component in this more freeform type of play to be insightful and true to my experience, thus I would propose it be integrated into my conspiracy theory thusly. We might need to parse out rules light systems vs ability light systems at a later date.
1. Rules light systems with few options are limiting at first, but eventually create situations where the GM is encouraged to more outside of the box rulings and thus more likely to give birth to the OSR wonderland of free choice, single-payer healthcare and budd light.
2. Rules heavy systems with many extra abilities like Limp Bizkit Points, Suspenders of Infinite Bezoars and Splintering Boar Strike offer a comforting mechanical blanket of abilities and rules at first but making ad-hoc rulings is both less necessary and more difficult to integrate within the existing framework and thus GMs are more likely to prescribe a by-the-book playstyle and players are more likely to follow it.
I will add that having a certain number of character abilities can actually provide seeds for ideas for creative play - like those weird miscellaneous magic items with no obvious use that you hang onto in case they become useful. Character abilities are in that sense things for the players to interact with. One of the things I did not like about 4e when it started was the dearth of out-of-combat spells, abilities and magic items (there were more later on). 4e assumed all out-of-combat actions were going to be resolved through player creativity (with reference to the skill system). Those options that do have a niche, out-of-combat function are also in competition with combat options that see more frequent use.
I totally agree that there needs to be a sort of basic level of complexity, like say, a decent selection of equipment, to permit complexity in the first place, and I totally see some weird non-conventional item being used for a creative purpose,
because it has no obvious use. So I perhaps it would work if we differentiate between systems with many abilities that are comparatively open-ended and broadly defined versus systems with abilities that are concrete and clear cut, like 4e.
Your description of 4e is essentially what made me dislike it on sight, and it didn't get much better playing through it. I felt much of the flavor, wonder and atmosphere of the earlier games was lacking. Magic was reduced to a utilitarian battlefield craft, rigid and codified. Opponents felt more akin to opponents in Diablo 2 then any sort of being with an independent existence.
Perhaps a worthwhile example of a game where you have a shitload of powers that is at least theoretically more geared towards open-ended rulings and interpretation would be the World of Darkness games.
Your example of 4e is interesting, but I would posit that my initial statement holds true for 3e and to a lesser degree to 5e as well when compared to OSR games. I must further elaborate my hypothesis; I suspect a game with a plethora of well-codified, straightforward abilities means that players will more readily be able to tackle challenges within the confines of their abilities and are thus less likely to improvise and so on.
So 4e was skewed towards combat, not because of too many character options, but because there were too few of them. And the way to address that in a 4e game is to give characters things to interact with, and teach players that creative solutions can work.
The question then becomes, would it not be easier to play/convert another game whose design philosophy is more in line with this type of play?