Thursday, July 17, 2025

Hell Is Other People's Misconceptions

New Delhi from space, photographed by the ESA's Copernicus Sentinel-2 satellites

Spend enough time on Reddit, and you inevitably learn odd things about other people. I've learned, for example, that if you put an unlabeled map of a fictional world (or even the real one) in front of a bunch of nerds and ask them where civilizations should develop and where large cities should appear, the vast majority of them will point to the mouths or deltas of that map's largest rivers. Which is wrong! And who can stand to see other people be wrong on the internet? Not I.

* * * 

You can get pretty far into the weeds trying to define “civilization” and pinpoint the places where it independently arose, but it's not much of an oversimplification to say that it happened, at least on our planet, in river valleys. In valleys along the Mexican Gulf Coast, the Olmec civilization emerged; in valleys along the Peruvian Pacific Coast, the Caral–Supe civilization. (These were actually pretty close to the sea, with cities like La Venta and Caral being within a day's walking distance of the shore, but still: river valleys.) The Egyptian civilization, famously, emerged in the Nile valley, far upstream of the river's enormous delta. Where exactly civilization first appeared in the Fertile Crescent is kind of an open question, but all of the likely sites are, you guessed it, river valleys, and of course the famous city-states that went on to invent writing, math, the wheel, and boots appeared along the banks of the Tigris and Euphrates. Likewise, the exact spot where civilization first appeared in China, insofar as you can identity such a thing exactly (you can't), is disputed, but basically: the Yellow, Yangtze, and Liao valleys. And the Indus Valley Civilization? It's right there in the name.

So, okay, they form upstream, but then they move down to the shore, right? Well, yes and no. Mostly no, at least until modern times. River mouths were generally unsuitable for ancient cities both because of a lack of fresh water (estuarine water is brackish) and because river sedimentation caused floods and led the watercourse to shift regularly, which could drown a settlement or leave it stranded inland. Dredging, canal-building, and aqueducts (among other technologies) made it feasible to put large cities near river mouths, but even then, a city like Alexandria is actually well to the west of the river. (And of course Alexandria was a classical-era Greek city, not an ancient Egyptian one; every major ancient Egyptian city was built above the Nile delta, and even Canopus and Heracleion, the much smaller predecessors of Alexandria, were just west of the delta.)

Many major ancient port cities, like Carthage and Constantinople, weren't built anywhere near major rivers, but rather on peninsulas that controlled sea lanes; they depended on huge rainwater cisterns and aqueducts to sustain them. Few major Mesoamerican cities were built on the coast, and fewer still along large rivers. The one urban civilization in North America that built along a major river, the Mississippians, did so 1,000 kilometers from the ocean.

Even today, most of the world's largest cities are not coastal. Delhi, São Paulo, Mexico City, Chengdu, Cairo, Beijing, Dhaka, Tehran, Kolkata, Guangzhou, Moscow, Paris, Seoul, London, Kuala Lumpur, Johannesburg, Kinshasa, Wuhan, Riyadh: all inland. Most of the major coastal cities today have colonial-era foundations (Karachi, New York, Rio de Janeiro, Buenos Aires, Chennai, Mumbai, Lagos, Manila, etc.) or are otherwise recent developments (Tianjin, Shenzhen, Saint Petersburg, Dubai, etc.).

There are some interesting exceptions (most big Japanese cities, and particularly Osaka, which has been a major port for almost 2,000 years; Jakarta; Shanghai and Hangzhou; and of course Istanbul), but they are unusual. Even with centuries of colonialism and globalized trade having driven the development of huge ports on every populated seaboard, and in spite of a colossal population boom in the last hundred years, the vast majority of the world's population still lives inland; fewer than 30% of us live within 50 kilometers of the coast.

Monday, July 14, 2025

Skills: Who Needs Them?


Is there anything interesting left to say about TTRPG skills? Probably not. But sometimes the brain itches, and writing down a bunch of stuff that other people have said a million times before, but slightly different this time, scratches the itch.

* * * 

Imagine five people: three athletes, a generic fantasy adventurer (a fighter, maybe), and a regular schmo.

Our athletes are an endurance runner, a weightlifter, and a baseball player. We're going to ask all five people to try their hands at the athletes' sports: they're going to run a marathon, lift an extremely heavy object, and play in a baseball game. And we're going to think about how to represent this in a roleplaying game.

Consider the marathon first. This is something with a low barrier for entry and an almost infinite number of grades of success. Let's assume that all five of our participants can run; it's probable that, given enough time, all five can run (or “run”) the whole marathon. It might take Joe Schmo 12 hours, but he'll get there. Our marathon runner should probably post the best time of the five, and Joe should probably post the worst, but a substantial degree of uncertainty is reasonable. The marathon runner might have trained hard for this but just not be built for it, whereas the adventurer might, despite never having trained for a marathon specifically, have developed tremendous distance-running skills while fleeing gelatinous cubes and the like. The weightlifter is probably much more fit than Joe, but she doesn't necessarily run well. Maybe she pulls a hamstring halfway through the race and has to limp across the line last.

Lifting weights (or heaving a fallen pillar out of the party's path in a dungeon, perhaps) is totally different. It's a strict binary: each person can either do it or they can't. Success might come with costs (a hernia?) and failure might be particularly drastic (crushed to death?), but the weight either is gonna get lifted or it isn't. The professional weightlifter's training will ensure that her technique is good, and she'll be most likely to avoid injury; the baseball player, who probably lifts weights frequently as part of his training, will also have good technique but probably doesn't have the same maximum capacity. If the weight is significant enough to pose a real challenge for the weightlifter, Joe and the runner won't be able to budge it, and the ballplayer and adventurer are likely (if somewhat less likely) to similarly just not be able.

Finally, we send them into a baseball game. Again, the level of challenge could vary quite a bit, and there are many grades of success (from just putting the ball in play to getting a base hit to hitting a home run). Any of these people might have some amateur baseball experience, but assuming that our pro ballplayer is good, and assuming the level of competition is enough to challenge him, the bar for any level of success is extremely high. The ballplayer himself might strike out on three pitches; it happens frequently enough. The other four are very likely to strike out, although the adventurer's combat-honed hand-eye coordination might give her a better chance to put the ball in play, and the weightlifter's massive strength might help her muscle a hit out of the park if she does manage to make contact.

* * *

Now, what the hell kind of RPG skill system is supposed to represent all of this? These are three challenges that, in 5E or any D&D-based system with a similarly stripped-down skill list, are likely to be represented by an Athletics check. If we have a system that doesn't lock certain skills to certain attributes (e.g., my trusty Stars Without Number), we might get a little stat-based variation (using CON bonuses for the marathon, STR for the weightlifting, and STR, DEX, or maybe even WIS for baseball). But how can the same flat skill bonus apply to all three?

Said flat skill check works passably well for the marathon, at least for separating the pros from the schmo. A d20 check isn't ideal—creating enough separation between Joe and our endurance runner that there's only a 2.5% chance of the former beating the latter requires giving the runner a bonus advantage of +15 (and that 2.5% chance is still pretty excessive)—but a 2d6 system will handle it easily with some comparatively small bonuses. Of course, if we only have a single Athletics skill, moving to 2d6 won't make the runner meaningfully better than our other pro athletes, which is weird, but we'll come back to that.

The skill check doesn't work so well for weightlifting. Even if we create the same kind of skill gap (Joe only succeeds 10% of the time; the weightlifter only fails 15% of the time), that's weird. This doesn't feel like something where everybody should be balanced on the razor's edge between success and failure. Sometimes, doing something difficult is a matter of raw aptitude. Training and experience might give you a slight edge and a lower chance of disastrous failure—you know just where to grip for maximum purchase, and you're not going to throw your back out in the attempt—but in the end, you're simply strong enough to lift that heavy thing or you're not. When games call for skill checks in situations like this, they probably just…shouldn't. Dragging that stone pillar out of the doorway simply requires 14+ STR. Maybe you can substitute something else (weightlifting experience? being a dwarf?) for one missing attribute point, maybe you can make an Athletics (or similar) check to see whether you injure yourself, but just doing the thing is a binary.

And baseball? It doesn't seem like a normal set of dice is going to be able to model how hard it would be for an untrained amateur to, say, hit a major-league slider. Nobody does that, not even the best amateurs in the world, until they've seen a few dozen in the minors, at least. So now even our 2d6 skill system is under some strain. If an untrained, physically average person in one of the XWN games (-1) goes head-to-head against somebody with maxed-out skill (+4) and the maximum stat bonus (+2), there's still ample room for an upset—a roll of 10, 11, or 12 by the amateur will beat snake eyes from the pro, for instance. Even in Traveller, where the untrained penalty is harsher (-3) and there's no hard cap on skill level, our pro needs a total DM+7 just to be guaranteed not to lose a contest—which is more than a starting character can have, and that in a game where starting characters are mid-career professionals and further advancement is slow and difficult.

* * *

Now, do we really need our TTRPGs to be able to model the difference between an average person and a top-tier professional baseball player? In a literal sense, no—when are our characters ever going to play pro baseball?—and metaphorically, probably not, because the literally one-in-a-million talent of an MLB player is a degree of aptitude games don't need to encompass (even if they need better granularity than “nobody can ever have better than a 95% chance of doing anything, or less than 5%"). But something like this comes up pretty frequently, especially in games with non-fantasy settings. The difference between a trained pilot and somebody who's never been in the cockpit, or between a university-educated physicist and a high school dropout, or between a professional programmer and somebody who's never touched a computer? Pretty vast!

Traditional skill checks work for feats like the marathon: anybody can attempt it, but physical aptitude and training go a long way toward making high-level success possible. Simple ability checks (or ability gates, even, without a roll) work for feats like weightlifting. Either you can, or you can't. For this third category, in which people are trying to do things that are challenging for them, in spite of their extensive experience and training, and would be simply impossible for an untrained person, who wouldn't even know how to approach the problem (or what button to push, as it were), we might have to combined the two types of check, and conceptually separate skill from training. You need training to even undertake the challenge; your skill determines the likelihood that you succeed.

Now, it's not completely impossible that an untrained person could fly a jet, or a spaceship. Maybe they've seen other people do it enough to have a rough sense of how it's done, even without any training. Maybe they should, as in Traveller, face a steep penalty instead of being locked out of the skill entirely. But is that penalty determined programmatically? Is it something you put in the rulebook? (E.g., Program is -5 for untrained characters, Pilot is -3, Surgery is -4, etc.) Don't characters with different backgrounds and experiences have an argument for different modifiers? If you're a soldier who's deployed from a dropship a hundred times, you probably have better odds of piloting one than a farmer who's never even seen a spaceship, even if neither of you formally has either training or skill.

And, again, there's that problem of too-general skills. Advocates of skill systems often say they help to define and humanize a character, but any system that makes a marathon runner, a weightlifter, and a baseball player mechanically identical—and equally good, or nearly so, at one another's sports—isn't doing a great job of defining and personalizing its characters. Then again, does anybody want to go to the level of GURPS (or even further) in terms of granularity? If we start splitting piloting into myriad sub-skills (sailboats, large merchant ships, helicopters, fighter jets, space shuttles), we run into the problem that any given character's skills are useless (or at least not optimally useful) 99% of the time, unless the GM finesses everything to make sure there's always a sailboat or helicopter handy.

There's a great, and reasonably popular, solution to all these problems and more (like the weirdness of level-based skill advancement, wherein a character crosses some abstract experience threshold, and poof, now somehow instantly knows how to do new things, potentially including things they got no new training or experience in; and the problem ). Ditch specific skills, let players define their characters' expertise narratively (via “backgrounds” or similar), and work out the numbers by some combination of negotiation, consensus, and GM fiat:

  • “I used to be a shuttle pilot, so I don't think it'd be a huge stretch for me to figure out piloting this fighter, given there's no immediate time crunch or danger."
  • “Remember when we did a bunch of climbing down in the catacombs a few sessions ago? We learned a lot from that, and climbing this ivy-covered wall seems to be more straightforward than that was.”
  • “Naah, I don't think a few hours studying at the local library was enough to substantially deepen your knowledge of arcana.”

Of course, I get why major (or “major”) publishers rarely present games like this—a mass-market game needs to cater to all players, including those who can't abide by loosey-goosey rulings and those who don't trust their fellow players with this kind of freedom—but for a home game, with the right players, it works like a charm. And yet the dream of a perfect skill system persists, even in the sparsely populated and OSR-inclined backwaters of the blogosphere and Reddit.

It should be robust, yet lightweight. Flexible, yet a vehicle for deep character personalization. Suitable for a simulationist approach, yet not excessively granular. I look forward to reading about! I'm sure as shit not going to come up with it myself.

Friday, July 11, 2025

Flyover Country: Chapter 10

Sketch by Kunrong Yap

 

Continued from Chapter 9.

* * *

The crew heads back to Morrow, making the drill out around midday on the 24th (Coordinated Atomic Time), arriving in Betharan late on the 25th, and getting down to the planet around the end of the 26th, all thanks to BQ's expert piloting. Having looked over the data they pulled from the ruins under Freeport, which included the sites of a dozen or so other pretech planetary engineering observation sites, they head straight for the most accessible source of extra pretech junk.

It's in a swamp, half-submerged but undefended and fairly easy to access. Unlike the site in Freeport, this one seems to be entirely inert, unpowered, unaware of their presence; they blast its door open with a demolition charge. It's not as full of goodies as the first one, but it does have the same computer equipment, wall panels, etc.—good enough to impress Ashbrook or her proxies. Orlando plays a little cat and mouse with her via email, trying to walk the line between appearing too eager to meet and too wary or too demanding. The crew spends a bug-plagued but otherwise uneventful day dragging old tech out of the ruin, and they eventually nail down a rendezvous that might give the crew just enough time to install a reaper battery on Quora before meeting Ashbrook's people.

They just need a reaper battery, and the right contractor, and a favor to skip the queue. Elias connects them to a Milieu-affiliated operation at the Freeport docks that has some used weapons lying around. For friends of Elias—and such a charming, gracious friend as Sarai, who naturally handles the negotiations—they'll sell the crew the weapon at cost and expedite the installation. Elias will owe the contractors a favor, which means the PCs will owe Elias another favor, but what else is new? By midday on the 29th, they're ready to go.

The rendezvous site Orlando has arranged is a desolate salt flat northwest of Freeport—not far, at least for a TL4 starship. Arriving, the PCs see a gunship (a shuttle with some armor and a small weapon, really) and a gravflyer parked not far from one another. Ashbrook is taking no chances; not present herself, she has sent a team of 10 mercenaries to handle the handoff. The gravflyer, it emerges, is meant to shuttle the crew back to the city after the mercenaries take possession of their ship. The PCs land, drop the ramp, let the mercs confirm that there really is a load of pretech salvage in their cargo bay, and arrange for the mercenary leader, with three of his soldiers backing him up just in case, to come aboard, inspect the goods up close, and prepare to authorize the transfer.

The PCs, naturally, double-cross and murder these poor schmucks. Thanks to yet another godawful initiative roll by the NPCs, the crew manages to teratically overload the mercenary leader and wipe out the other three (including the team's most capable, and best-armed, fighter) without a shot. The remaining mercs immediately launch their gunship and try to disable Quora's new weapon, but instead quickly find themselves facing a hull breach and fire in their cargo bay, then a disabled engine. They try valiantly to make repairs, but can't quite manage, and all the while the PCs' faster, more powerful ship is breathing down their necks. They crash-land their gunship into the accommodating desert plain, and after some negotiations (they aren't eager to leave cover and their one decent bargaining chip, the intact ship), the PCs take the six remaining mercs prisoner.

An uncomfortable debate promptly begins. Should they murder these guys? Batias is strongly in favor, of course. Krissa is uneasy about the ethical dimensions, but not keen to let a bunch of guys who might be able to work out that she's a psychic walk free. Roman, Mustang, and BQ are various shades of indifferent; they set about inspecting the downed gunship and trying to get it back into flying condition rather than getting bogged down in the argument.

In the end, Sarai, despite being the only person strongly opposed to summary execution, is the most eloquent and persuasive of the bunch. She proposes dumping the six prisoners in farm country farther north, beyond the desert belt, without any comms equipment. They won't starve or die of exposure, but it'll take them days to find their way back to Freeport, and by then Quora will be long gone. Roman and Mustang stay behind to work on the gunship while the rest of the crew carries out this errand of mercy.

Once the gunship is up and running and the others have returned, BQ loads the gravflyer aboard Quora and they blast off for Freeport, where they'll leave the gunship and the soggy pretech junk for Elias to dispose of. Krissa, having stabilized the downed mercenary leader almost as soon as she disabled him, straps the guy into a bed in the medbay, slaps a Squeal patch on him, wakes him from his brief biopsionically induced coma, and interrogates him. Between that and searching the gunship's computer systems, the crew learns a little about Ashbrook and Mputu, but not much. They learn more about the mercenaries and their ship, for whatever that's worth.

* * * 

Eminent Alternatives, it seems, is the PMC Ashbrook hired for her various Freeport jobs; this is the same outfit that attacked the PCs as they were trying to unload their trucks back at the city dump. The field commander who led the team whose asses the crew just kicked, and who is now their prisoner, is a guy named Wei Khan. Ashbrook has had EA on retainer on Morrow all month. Like her, they're based on Khabara, and their previous job was in Istanu, one of the systems that connects Sanasar (Khabara's star) to Betharan (Morrow's). EA is a small company, with only about 25 employees total (some of them administrative types), including those who went to the handoff. The loss of four soldiers (including Wei) represents a pretty disastrous setback—that's a quarter of their combat strength.

Wei did some digging into Ashbrook's background and reputation when she offered EA the job. She's worked with all the major Khabaran security outfits (OKR Personal Defense; Patil, de Lima & Company; Rodrigues+Chen) and various smaller ones, and she's also brokered contracts with Juman privateers via the Mundan firm Müller, Bhatnagar & Yang (probably off-books mercenary work by Imperial military personnel). Wei got in touch with a couple of the smaller outfits that have worked with Ashbrook, Cloudy Mountain Defense Services (based on Khabara) and Sable Company (a local Morrow operation)—they found her reliable, professional, demanding, and unfriendly.

Between the lengthy retainer and some beefy hazard-pay bonuses, Ashbrook seems to have spent almost 40,000 credits on the mercs, and the money came from her longtime client Mputu Manufacturing. The mercenaries' gear was not entirely uniform—their armor was mostly from Ogenne, a Khabaran company, but their weapons a hodgepodge from Sayed+Spiner, Laxardal Forge (both Mundan corporations), Omni Armory (from Opis, in the Commonwealth), AKHI (the massive state-owned conglomerate in the Directory), and other sources. The ship itself was of Khabaran manufacture (from Gemini Interstellar, a subsidiary of Seneschal), and was provided by Mputu, not owned by the mercs. Far more than the mercenaries and their gear, the ship represents a huge loss for Ashbrook—a setback second only to the loss of the computer equipment and data from the Freeport dig site.

The gunship is called the Stormy Petrel. Stripped of its sandthrower (which the PCs want to hastily install on their own ship), it's worth about 230,000 credits on paper, but it's now both stolen and damaged, which dents its value—Elias might get close to full freight for the armor plating, which can be stripped off and reused, but repairs will cost a good 10,000 credits, and refitting the hull, changing the drive signature, counterfeiting new registration, and so forth will cost another 100,000 or more, so even if Elias can resell the ex-Petrel for close to full price, the profit is “only" going to be in the neighborhood of 110,000 credits (and he'll take a hefty cut). The new pretech junk might sell for 80,000 or so, after expenses getting it where it needs to go and all that, and Elias will take his cut there too; still, the crew are looking at a payday well above 100,000 credits. It'll take at least a few weeks to come through, but hey—not bad for a couple days' work.

The Echelon Altostratus gravflyer the mercs brought (also the property of Mputu Manufacturing, big enough to carry 10 crew and passengers, and crammed with bells and whistles) stays in Quora's cargo bay; it seems like a handy thing for the crew to have, even if it takes a big bite out of their cargo capacity. The onboard computer had a flight plan prepared that would have taken the vehicle back to a Mputu Manufacturing facility in Freeport (at the north edge of Center City, near the Havre)—it seems Orlando and co. were meant to just get a lift back to the city, then return the gravflyer to its owners and find their own way from there.

* * * 

With a bunch of stolen goods and ever more enemies gunning for the PCs here on Morrow, it's probably time for them to make themselves scarce. Elias points them toward Istanu, one star over. It's a neutral system, long dominated by megacorps from Khabara but now sliding into the Kyran Directory's sphere of influence, and its on its way to the Directory's core systems, where the PCs have vague designs on learning how to counterfeit atrament. The Directory's main base in Istanu is Seven Miracles Station, a port in LEO above Salafai, the system's one Earth-like planet. The Milieu is well entrenched there, and the local boss is an associate of Elias's: Nour Mbuyu.

First, though, a brief entrepreneurial interlude. The PCs may have just double-crossed a powerful corporate fixer, stolen a ship worth hundreds of thousands of credits, killed three mercenaries, and captured a fourth; they may be fighting against the clock as the mercs they didn't kill work their way back toward Freeport and could be getting in touch with Ashbrook any minute now; they may still be on the bad side of the entire Najeeb crime family and harboring the traitorous Orlando Ilunga; but they still have to make time to look into a good business opportunity.

Hoop Cola, you see, is going gangbusters over on Rustam. Jenny Beck may have been an indifferent engineering student, but it turns out she's got a real knack for marketing and soft-drink distribution. And Roman and Sarai have, in their spare time, cooked up a Diet Hoop recipe that they have every confidence in. They schedule a quick meeting with Hoop Barrett himself while Elias's people get the Petrel's sandthrower installed on Quora. They hash out a business plan: Roman, Mustang, Sarai, and Krissa will each contribute 2,500 credits to an investment that'll get Hoop's bottling operations going again; in exchange, each of them gets a 12% ownership stake in the company, with Hoop narrowly maintaining control with his 52% share. Dividends will be paid out quarterly from profits. Hoop is advised to get Diet Hoop production going and to concentrate on the burgeoning Rustami market. And then we're off.

Poor Wei Khan definitely knows that Krissa is a psychic, and a biopsion at that. Overriding Sarai's moral squeamishness, they chuck the guy out the airlock, in drillspace, during the voyage to Istanu. Krissa performs a little Juman ritual—a salt circle, some burning alcohol, some drinking, a bit of chanting—to keep the spirit from catching on the hull of the ship. Does it work? Well, they don't hear any ghostly screaming on the way in to Salafai, so…maybe?

Hell Is Other People's Misconceptions