Traveller in my case, but I'm not worried about stats.  My thinking is there's all kinds of stuff I don't necessarily want to put on the standard equipment list and have everyone be carrying one of from the first session, but that would still make for a good game to have in the hands of a player from time to time.  So items cool enough to work as loot/treasure/reward/one-offs for sale, but still appropriate to a science fiction game.

I'll start, with a few things I posted elsewhere, apropos of something else.

• tunable electromagnet, bulky but powerful
• high tech grapple gun with line and integral heavy duty motor
• bag of large synthetic gems suitable for industrial applications - possibly valuable on truly uncontacted worlds, but anything reachable by standard Jump isn't going to fall for it
• high tech, ship-strength cable, shape and movement controllable from a pad on one end, but frustratingly difficult to control without the Remote Ops skill
• an entertainment unit pre-loaded with Beethoven's symphonies and hours of whale song, one or both of which are known to please 90% of all sapient races that can perceive the sound
• cheap knockoff grav belts, bulky and prone to burning out after repeated use
• telescoping staff the size of a thick coin when retracted, practically unbreakable extended
• Rocket boots.  Yes, grav belts are better in almost every way, but hey, rocket boots.

More like this?

I've figured out what I want out of magic, that rpgs don't model.  It's those one-time, supreme efforts that, for whatever reason, don't get reproduced.  Think Gandalf summoning the white horses/river flood in LotR.  Or the magician in The Face in the Frost tearing down a bridge to stop pursuit, but he barely knows how to do it and loses his Tarot deck in the process.

I would call this a great work, but I see that has an established meaning, so I'll call it a supreme effort.  It's a part of magic in fiction and myth, but it's hard to pull off in an rpg.  Players -understandably!- think if they can do something once, they should be able to do it again.  That's certainly the case in DnD; once you research a spell or magic item creation method you have the formula.  And I'm looking for the opposite, that you may get any particular impressive feat only once in your career (though you might have several such over time).

Is there a game that does this already, that I'm not aware of?  Free-form, or word- or element-based, magic systems look like a partial start, but still suffer from the problem of doing something once establishing a precedent.

Or failing that, how would you do this from scratch?  I don't want something purely free form or magic tea party, I'd like some rules or guidelines on what you can or can't do, it's just by it's nature it should exceed the power and scope of daily spells.  I think I'd impose a chance of failure, a chance of blowback, or both, to keep it from being a gimme play.  I'm hung up on the guidelines for what you can do in particular though.

Say I wanted to run a play by post hexcrawl/exploration game.  In person I could simply hand the party a blank sheet of hex paper and tell them mapping is up to them.  But for a play by post game, how would I do that?  Is there a website or an app that would manage that?  I guess worst case everybody gets hexmapper or hexographer and a dropbox account, but it seems like there should be a better way.

ACKS Qelong is going well enough we have the highest level PCs at 5th, and we're outgrowing some of the dungeons I originally seeded and hooked.  In ACKS that's moving into overland travel and lair territory, which is cool, but the lair in the last session almost kicked the party's ass, and they're looking to slum it with a dungeon a little bit.  But dungeons for 1st or low level characters are pretty well covered by the OSR, so it's a good time to ask if anyone can recommend any good ones for mid level characters.  Ideally something they can leave and come back to rather than a quest adventure, as the players present can change from week to week.

(And yes, I am going to start rolling up my own, but by the nature of a sandbox I'd like to place more than one.)

I have long hated niche protection, and I think I've finally figured out how to express why.  To explain, let me start with a game where, in theory, it kind of works.

So take D&D.  Fighter, cleric, mage, thief.  In combat, the fighter fights, the cleric is a secondary fighter who decides when or whether to heal or turn undead, the thief sneaks around and tries to set himself up for a backstab (or just stands back with a shortbow once he sees his actual chances), and the mage, while he may have the least to do on a turn by turn basis, decides when to bring the artillery in with a Fireball or Sleep spell.  There's some strong niche protection, and not everyone's equally involved every turn, but in principle everyone has a role in combat.

And out of combat (and again, in theory) everyone should have something to do as well.  The fighter is a leader of men, the thief wrangles traps and locks and spy missions, the cleric can heal and preach to villagers, the mage can research spells or scribe scrolls.  I'm aware this part starts to break down in later editions, and especially where the words "face man" or, even worse, "diplomancer" can be uttered with a straight face, but that was the original ideal, especially in games that assumed downtime.

So we've had that concept of class protection from the very start.  But that's been combined with people getting used to stable groups of four or five players (which wasn't the case right at first, though it emerged pretty quickly), and we got stuck with the idea of niche protection.  And now I've seen groups and individuals in everything from Traveller to Savage Worlds to L5R start talking spontaneously about niche protection as if it's either a good thing or just a baseline assumption.

But stop and think about that.  Take Traveller, or sci fi generally.  If your niches shake out as, say, "pilot/ship guy", "combat specialist", "party face", and "ship's doctor/medic" you've just accomplished the opposite of what the original class system did.  Because now there's only one player with something certain to do in combat, one player with something to do in social situations, one player to do in chases or flight scenes, and so on.

That in turn imposes a burden on the GM, to "challenge" each character - except really it works out as name-checking everybody's separate skill sets.  So every session or every adventure ought to have a social test to get past (but not one that lasts a full night of play), every adventure needs a combat (which is somehow going to have to challenge the combat monster without pasting the noncombatants, an added burden), every adventure needs some kind of technical challenge which just so happens to require whatever engineering, repair or computer skill one character has, and so on.

But if involving all the characters is such a good thing, how much better would it be to just have a party full of generalists who can all be involved in multiple types of challenges?  If everyone's got a combat skill of some kind, but no one's head and shoulders above the rest, the GM can involve and challenge the whole party without turning the worst combatant into tomato paste if he stumbles across the combat monster's intended foe.  If the game has different social skills, why not split Diplomacy, Bluff and Seduction across three different characters instead of stacking them all on one with a maxed out Charisma, who's magically always present to talk to NPCs for everyone else in the group?

I think this is one reason I like Traveller's random character creation.  Because you can try for certain things, but out of a group, at least one person's likely to get Mechanic whether they like it or not, you may well see some skill overlap around Pilot or Engineer, and in general you get a mix of skills you may not have expected or desired if you were trying to min-max and match your skills to your attributes and your expected party role.  That's a good thing!  Skill overlap is a good thing for involving more players in challenges.  And, tactically it's good to have some backup if someone gets shot (or the player just misses a session).

Similarly, the last time I ran L5R I just flat out told the group, "hey, try to make well rounded characters.  Everyone should have something to do in combat, everyone should have one of the social offense skills, every samurai should have a courtly skill of some kind, whether it's a Lore or Perform or Artisan skill."  I didn't actually veto anyone making a combat monster or a helpless courtier, but I warned them up front that if they did I wouldn't be tailoring any challenges to them.  So they might well make paste out of any one challenge they came across if it matched their skills, but I wouldn't go out of my way to challenge them or involve them on their highest strength every session.  And that campaign ran better than when I've seen min-maxers go for niche protection in L5R.

So I'll make a claim, though it applies more to point-buy and skill based systems than well designed class systems:  at a meta-game level, niche protection is just plain bad for games.  It's bad GMing if the GM is the one encouraging, and it's bad play if it's players seeking it out.  For games that differentiate them, by all means distinguish characters with different weapon skills or fighting styles, different technical skills, and different social skills, but it's far better to spread those out across the group than stack them on one character.

Anyway...  Am I wrong, and there's a counter-argument I'm not seeing?  Or is this not even news to anyone here, and I've just been unlucky in some of my face to face groups?