Consilience #61: Grim words of perilous science-adventure

I'm audible for the sixth time. A couple new voices too, with Patrick Till and Deon Barnard making noises into mics 2 and 3. Apparently the Meadons have become too weak from their Meadon-spawn and work and shit to stay in the podcast every single episode. (Seriously, kids: Just say "No!" to reproduction.) It was a good recording, and I felt it was interesting and well done.

One thing I wasn't entirely sure of was how well I'd explained the problem of researcher "degrees of freedom." Steve Novella explained it much better a while back. But here's an analogy: You're driving down the road. You decide that staying out of the on-coming traffic's lane 90% of the time is good enough. On an empty road in the middle of the night where there is no other traffic, that might be practical enough. But during rush hour, on a busy central road? Context matters, and the point I tried to make in the podcast and maybe failed at was that the context of each subjective research decision made in the Mossbridge meta-analysis has to include all the other subjective research decisions, and not just external stuff.

Somehow, inevitably, another travel anecdote: I still don't officially get the Gilooly's interchange and fucked up my planned turn through it, putting me on the N3/N12 South. I recognised my error pretty quickly, but there just aren't many places to turn around on that stretch, so I aimed for a spot I know really, really well, the Camarro offramp on the N12 West. I used to live just off Camarro back when I was teaching in the South, and I love that stretch of N12 at night. It's great quality road (it was even before the big '09/'10 highway rejuvenation, which somehow made it even better) and it does wonderful bends and weaves that are a joy to speed along (responsibly), with relatively few other cars in the way. The N12 East, at 01:00 on a Sunday night, was one of my great pleasures back when I lived in the South. That, and never having to put on any clothes, while eating nothing but jars of peanut butter and watching Firefly in my little flat. It was a weird, wonderful time. But the bottom line is, I ended up driving about 1.3 times the total distance of the ring-road around Joburg, thanks to my fuckup.

You can find the file and the show notes for #61 here:
http://consiliencecast.wordpress.com/2012/10/26/episode-61-nobel-prizes-meta-analyses-and-a-real-warp-drive-this-time/

Protestor in pro Imperatore: WFRP3

"Romanes eunt domes!"

I've mentioned before that I like Warhammer Fantasy Roleplay. And while it's a mistake to declare any one system or setting The Best, I will say that WFRP is definitely among the best, both in terms of technical playability and subjective experiencey-ness. Except I have to be clear here that I mean WFRP 1st and 2nd editions. I completely reject the 3rd and most recent edition. This post will be a rant along those lines.

The original WFRP began with a weird foundation, borrowing rules from its miniatures predecessor, Warhammer Fantasy Battles, but also simultaneously borrowing from and competing with Dungeons & Dragons. It came out pretty good, with a simpler, smoother game mechanism than D&D, and on average much better plots, though that may be more of a numbers effect, since waaaaaaay more D&D stuff has been published by a much wider selection of authors and for a much wider audience, increasing the odds of some being written in styles I subjectively disapprove of. But the point is, WFRP1 was a good system underlying great content.

WFRP2, borrowing the skills/feats distinction from D&D 3rd ed., took a decent system and made it near-foolproof (sadly, I've played near fools a lot). It really is one of the best systems I've seen, covering everything I ever need it to (bullshitting, disease, dismemberment, sailing, etc.) in sufficient depth to make the rules worth having at all, but not bogging down play in fiddly detail so much that only accountants and marketing people can care about it. WFRP2 is almost exactly what I want from a system, and the closely-related Warhammer 40,000 Roleplaying system shows how adaptable it is, once you've put in the hard work of re-writing the entire careers system to reflect a completely different setting. But to be fair, "careers" or "classes" or "occupations" or "professions," whatever the equivalent flavoured skills packages in any system are called, are almost always very difficult to re-build and customise, even within the same setting. To compare a rules system across genres, it's more reasonable to look at how the core mechanics translate than anything else. And WFRP2's core mechanics are great.

Then came the premature, unnecessary change to WFRP3, now published under Fantasy Flight Games. I tend to like Fantasy Flight; their boardgames are generally excellent. Twilight Imperium, in particular, is masterful. And weirdly, when they took over the 40K roleplaying, they kept the same system and did brilliant things with it. Yet WFRP, they botched. They took an excellent system with 20 years' heritage and completely junked it. It's not even like original D&D 1st ed. vs. 4th ed. (or the upcoming 5th), where it's gradually become quite different, but you can still identify the common heritage. Instead, they imposed a whole new system from scratch, which doesn't run as smoothly (even though they've stripped out character detail), doesn't convert at all well between earlier editions or with the miniatures game, and requires you to purchase a fucktonne of extra cardboard to play it at all. Before, all you needed was a core book, blank paper and a d100. Now, with WFRP3, you need action cards, condition cards, item cards, creature cards, terrain cards, talent cards, career ability cards, equipment cards, assorted different classes of tokens, and a bajillion special dice, etc., with every single expansion. They've tried to push the boardgames' way of doing things onto a roleplaying game, and it's a mess.

Worse, they haven't even got decent plots. Everything's simplisitic and mechanical, with no depth or flavour. I can tolerate a bad system that props up decent plots (my friend Arran, for example, did good things with Vampire, which was never a smart system), and I can accept a system that refuses to provide plot at all, leaving that responsibility to the more creative GMs. But to push big, expensive background and adventure books (in fact, always box sets, in WFRP3's model) that don't actually provide much background or plot information is just wrong. The mark of a good roleplaying book is not how big, heavy or expensive it is, but how useful it is to the GM. WFRP1 and 2 books are full of way more information than you need, a lot of it done in a style that conveys the style and tone of the setting. WFRP3's books offer only a bare minimum, much of it in a clunky, step-by-step style that makes sense in a list of boardgame rules, but not in a long description, explanation or exposition.

I will admit that the introduction of nicely-printed character stand-ups was nice. WFRP can (and was designed to) borrow heavily from the real miniatures of its parent game, but that is admittedly a bit more like hard work than many roleplayers are used to, and quick, easy cardboard substitutes are a fair compromise. But that doesn't really justify (or relate to) any other change made to the game.

Now I'm especially annoyed that Fantasy Flight have taken hold of the name Enemy Within for their newest release. Enemy Within was the grand WFRP1 super-campaign, a really monumental thing, declared the best campaign of any sort ever by gaming magazine Casus Belli, and described by roleplaying guru Jonny Nexus as "superb." My group and I are playing through it at the moment, 48 sessions and 3 years in, and it really is amazing, though definitely not well suited to those with short attention spans. WFRP2 is the better rules system, so we use that (it's a very easy conversion between the two), but it didn't stay in publication long enough (thanks to the horrible decision to switch to WFRP3) to produce a new campaign of similar quality. The WFRP3 "Enemy Within" name-steal seems like a mere crass, commercial attempt to make a quick buck off a classic name, and I fear, based on my assessment of WFRP3 so far, that it'll end up spoiling that good name. It's probably not going to be anywhere near as good, but even if it is, and even if you're a fan of WFRP3, it shouldn't steal the name from the classic campaign. It'd be like like using the Star Wars name for stupid movies with a cartoon rabbit and a sulky, emo, teenage Darth Vader. Or like putting the Star Trek name on movies that have no interest at all in Roddenberry's utopian vision.

I don't imagine much can be done to pressure FFG into changing their minds at this point, but I still feel compelled to object... for Sigmar!

Mercenaries: I Hate Being Right, Sometimes

Once upon a time, back in 2006, I did a politics course on international relations, focused on good vs. bad governance. I did many such things in my crazy youth. And as part of this course, I wrote what I considered at the time to be my greatest piece of writing up to that point, an essay on private military companies (PMCs, essentially mercenary companies) in Iraq. It was plainly obvious that the two specific PMCs I had focused on were bad people, as almost all of them seem to be, and not solely because they want to kill people for a living. But what's more important is a historical perspective: By comparing early 21st century PMCs with 17th and 18th century privateers (sea mercenaries), it's possible to do some interesting hypothesizin' about where the modern embrace of PMCs might lead over the next century.

My prediction is not a rosy one: Privateers might have changed sides a few times (fuelling wars far longer than would have been possible without them), but eventually they all went out of business when peace finally broke out, and turned to illegal (but materially identical) piracy to make a living instead. And who did national militaries turn to to fight these sudden surges in piracy? Privateers. This stupid cycle carried on for well over a century. And I'd hypothesize that our century's recent massive swing towards PMCs will lead to similar trouble in future.

You know what makes science great? Rocket-powered dinosaur clones. But you know what else is neat? Empirical evidence. It's still too early to fully confirm my hypothesis, but there is already some evidence leaning that way, most notably this story from the New York Times, about an effort at training mercenary pirate-fighters in Somalia, which has achieved nothing but training and arming a bunch of guys, then leaving them unemployed and free to use their arms and skills however the hell they want to make up for their lack of formal pay. It's the worst possible combination, but it's far from being the only such force, and it doesn't take too much imagination to picture the others in similar positions when they're finally downsized.

So what to do about it? Broadly, I'd say two general things:
1. Stop trying to solve every political problem by shooting people in the face. There are always better options, albeit more complicated ones that require the use of more of the brain than it takes for reloading and trigger pulling, but they just take a little training to become apparent. I've studied and worked with some people from East Africa, for example, and it's not like they don't give a shit; if anything, they're more motivated than anyone else to sort their region out peacefully and might have made some progress towards that if it wasn't for the constant, violent foreign intrusions there. The people hiring mercenaries must be pretty short-term thinkers, but regardless, they have to be stopped. No more of this shit, nor its more legally-clear counterpart of using national militaries to shoot people in the face. It's neither a sufficient nor ethical solution.

2. Give these people other skills. If you're only qualified to shoot people in the face, then you're unlikely to suddenly turn to a job in IT or public health, let alone streetsweeping or shit-mopping. So they'll just perpetuate the violence, making a living the way they know how, unless we help them along and give them something more constructive to do instead.

Easier said than done? Fuck yes. But definitely worth starting to work on, before we get stuck in a cycle of stupid, pointless violence and crime for another century. It's sometimes said that the social sciences are weak because the ability to intentionally change things wrecks their predictive ability. Here, I'd say that's a strength. I want my hypothesis made wrong. Make it wrong just to spite me, if necessary.

Astronyms, part 5: The Shuttles and the ISS

This is part of this series of posts on the history of spacecraft naming.

Space Shuttle Orbiter
(Space plane. In space service 12 April 1981 to 21 July 2011)

Space Shuttle Orbiter

The Space Shuttle Orbiter, more commonly but incorrectly/informally called just 'space shuttle' or 'shuttle', was the central space plane component of the Space Shuttle (which more correctly refers to the entire launch system, including the big external fuel tank and solid rocket boosters). By the 21st century, we'd become pretty jaded about the Orbiter, but try to put yourself back in the mid-1970s, when it was still being developed: Everything until then had been a dinky space condom, 100% disposed after each use, fitting no more than 3 pretty damn cramped people at a time, small enough to fit inside a large-ish two-car garage, landing at some fairly unpredictable spot in the damn ocean or a muddy field in the Kazakh steppes, dangling crudely under parachutes. Then along comes a fucking great jet-looking thing, with seats for 7 (and so much more leg room!), an enclosed cargo bay big enough to fit an entire Apollo CSM and LM pair (or 3 or 4 Soyuzes, or a million Vostoks), yet with the precision to land comfortably on a given runway, and then you could use almost the whole thing again and again and again! It really must have seemed like a massive leap forward.

Counting against it, unfortunately, were some serious flaws, some apparent from the start, some only emerging over time. They gave unexpectedly poor economy and were harder and slower to maintain than hoped. They couldn't get beyond a low Earth orbit, less far from Earth than the Apollos or even Geminis could. And their safety record was horrible, with two destroyed in horrible accidents, partly due to awkward design limits, partly due to terrible mismanagement, as well as a long history of bits falling off in less catastrophic ways. The whole thing was a size and shape that NASA hadn't actually wanted, but was forced to accept because the US Air Force wanted a big reusable launcher for secretly putting big spy satellites in orbit. In the end, the USAF didn't really use the shuttle for that often (opting for cheaper conventional rockets and little uncrewed jobs instead) and NASA seldom needed the full launch capability they had, since construction of the International Space Station, probably the shuttle's greatest project, only began when the shuttle was 20 years old already. It was, in a lot of ways, a bit disappointing, largely because nothing was ever done to really build on it. It should have been an interesting prototype, perhaps more like the X-15 or the cancelled X-20, from which both smaller science space planes and larger cargo space planes and so on could be developed. Instead, its career was dragged out and constantly uncertain. But I guess that's a good metaphor for the whole of post-Apollo human spaceflight.

Anyway, names! Space Shuttle, Space Shuttle Orbiter, Solid Rocket Booster and External Fuel Tank are all pretty descriptive, but so, so very bland. It seems like NASA lost its passion for naming things properly as the rest of the US lost interest in the space race. Skylab was already a little bit of lame name compared with Apollo, but Space Shuttle Orbiter isn't even a name, it's just a description. It feels wrong treating it like a proper noun. Fortunately, someone at least decided that the individual vessels should each get unique names, chosen in a serious, sombre, but at least interesting manner.

They already had production numbers, in the format OV-0xx or OV-1yy, where OV stands for Orbiter Vehicle, xx is a descending number starting at 99 and y ascends from 01, with the '-1' intended to represent an actual, flying vehicle and '-0' a non-flying ground-test hull, until they turned STA-099 (standing for Structural Test Article) into the flying OV-099 without giving it a new OV-1xx number. In principle, they might also have had OV-2zz and OV-3ww numbers for future advanced versions, but never changed the orbiter's design significantly enough to warrant that. Compare this with the very complex Mercury production numbers like "7"; I believe in bureaucratese, that's considered progress. Fortunately, NASA's inflexibility with production numbers at least makes it very easy to tie any number and unique name together.

The first shuttle, OV-101, was to be called Constitution, after the still-fucking-sailing US Navy frigate that was among the first vessels commissioned by the new Navy when it formed in the late 18th century. But Trekkies, who hadn't been much of a coherent demographic until the late '70s (I guess it was slower and took more dedication being organised fans of something before the internet), suddenly got it in their heads that it should be Enterprise instead. And there were a lot of them. And many of them were NASA employees. And some of the older, more senior NASA employees had served on the WWII carrier USS Enterprise (CV-6). There might also have been a little extra bit of input from the then-new nuclear-powered carrier USS Enterprise (CVN-65). And so, OV-101 became the shuttle Enterprise. And it never went into space. Sigh.

[Edit: Some newly released documents confirm that's pretty much exactly how Enterprise got its name.]

The actual spacey space shuttles, in order of first space flight, were:

• OV-102 Columbia. 28 flights from 12 April 1981 to 1 February 2003. Destroyed on re-entry during the 28th. Named after an exploratory sailing ship and the poetic name for America. Although it's inevitable that the Apollo 11 CSM of the same name was in mind when this one was chosen, it seems that they reached the same name from different roots.
• OV-099 Challenger. 9 flights from 4 April 1983 to 6 November 1985. Destroyed on launch of what would have been 10th flight, 28 January 1986. Like the Apollo 17 LM, it was named after the HMS Challenger and its Challenger Expedition, an early major exercise in oceanography.
• OV-103 Discovery. 39 flights from 30 August 1984 to 9 March 2011. Named after four sailing ships, primarily James Cook's last ship, the HMS Discovery, but also Henry Hudson's colonisation and exploration ship Discovery, George Nares's Arctic explorer HMS Discovery, and Scott and Shackleton's Antarctic explorer RRS Discovery of the Discovery Expedition. Frankly, I can't think of a better name for a science and exploration vessel; it's a sentiment, a goal, a boast, and to my ears it's also a nice-sounding name.
• OV-104 Atlantis. 33 flights from 3 October 1985 to 21 July 2011. Named after RV Atlantis, first research ship of the Woods Hole Oceanographic Institution. Am I the only one who finds the name Woods Hole pretty damn funny?
• OV-105 Endeavour. 25 flights from 7 May 1992 to 1 June 2011. Like the Apollo 15 CSM, it was named after James Cook's first exploratory ship, the HMS Endeavour.

I count a total of 134 flights into space there. I think they were originally aiming to get at least double that many by the time the shuttles retired, but it's still not a bad number, and will exceed even the whole Soyuz family's total number of flights for several more years.

I'm quite glad Constitution wasn't used, as it could have set a precendent of giving the shuttles ugly, exclusionary US nationalism-themed names. Instead, with the set actually used, you end up with a pretty decent theme there of shuttles named after famous research and exploration sailing ships, with some ties back to the previous generation of Apollo CSMs and LMs. I'm personally in favour of looking for new, more original or at least more intentionally-chosen names, but if you are going to recycle old names, then this is a pretty good way of doing so. It also shows that impersonal naming committees made up of people who'll probably never ride the vessels themselves don't have to settle on bland-shit names like Space Shuttle Orbiter and International Space Station every time.

[EDIT: I've just come across this 1972 White House memo, trying to get the shuttle a new name before it went into production. It seems some people were thinking about proper names, but Nixon fucked it up by disregarding the suggestions. Fucking Nixon. Anyway, the possible class names listed there are Space Clipper, Pegasus and Starlighter. I like Pegasus most, and just don't get Starlighter at all. If it had been Space Clipper, the first of the class (OV-101?) would likely have been named Yankee Clipper, same as the Apollo 12 CSM. Still a historic ships theme and possibly the root of the actual naming theme.]

International Space Station
(Space station. In space service 2 November 2000 to present)

International Space Station with Soyuz and Progress docked

Our greatest defence against the Borg, often abbreviated to ISS. This thing is really vast compared with everything we've had before it (just look at this comparison), but admittedly a lot of it is uninhabitable, especially those giant solar panels. Remove those and you're back to about the size of Mir, and with no single module as big as the girthy Skylab. Still, the solar panels are there, and so is the 100m-or-so truss they sit on, and the whole thing is apparently big enough to make it the second brightest object in the night sky, after the Moon, though I've never been able to spot it myself. Stupid city living.

International Space Station is a crap name. Like Space Shuttle Orbiter, it's just a literal description, nothing better.The first crew to arrive on board did try to give it the new radio call sign of Alpha, but they made it clear that they intended it as a full, permanent name for the station. This annoyed Russian management, who would have preferred something like Mir-2 or something similarly uncreative (the Russian cosmonaut on Expedition 1, Sergei Krikalev, seems to have been happy with Alpha). Space Station Alpha had been the name of a planned US design that was later folded into the ISS mix, and it seems Russia felt it unfair to pick a name implying all their past stations hadn't counted. As a result, it's pretty hard to find references to the station as Alpha much beyond 2001. I've seen one astronaut call it that in a 2006 interview and the ham radio crowd may still use it from time to time, sometimes in the form of ISS Alpha. But absolutely every official source today is void of the name Alpha, wasting everyone's time with bloody long and boring International Space Station instead. I have few strong feelings about the name Alpha - it's a bit bland without context, but not awful - but at least it's a real name. I might buy that it could represent a fresh start for human spaceflight, a chance to start over in a cooperative way and put aside the Cold War competitiveness of the Space Race, but that's just my own thought; I have no idea why NASA actually picked it. That interpretation is spoiled quite a bit by the exclusion of China, which led to the separate development of a Chinese station program.

Like Mir, the ISS is made of lots of modules, which each have their own names, so there at least is a bit more to discuss. The pressurised (inhabitable) modules are:

• Zarya (meaning "dawn") or Functional Cargo Block
• Unity or Node 1
• Zvezda ("star") or Service Module (mentioned previously, basically an upgraded Salyut)
• Destiny
• Quest Joint Airlock or Joint Airlock Module
• Pirs ("pier") or Docking Module 1
• Harmony or Node 2
• Columbus
• Poisk ("search") or Mini-Research Module 2 (formerly Docking Module 2)
• Tranquility or Node 3 (which should rightly have been Colbert, or at least Serenity)
• The Cupola
• Kibo ("hope") or Japanese Experiment Module
• Rassvet (also "dawn") or Mini-Research Module 2 (formerly Docking Cargo Module)
• Leonardo (formerly temporary, now permanent) or Permanent Multipurpose Module (formerly Multi-Purpose Logistics Module 1)
• Raffaello (temporary) or Multi-Purpose Logistics Module 2

In addition, there's the truss or Integrated Truss Structure, made up of 12 modules, numbered according to their position. One of them, the Z1 truss segment, has some pressurized space, but the rest are just cold support structures. There are also some robotic arms and unpressurized modules, but I'll skip them for now, since too many of them are annoying backronyms. There are several planned additions, but many additions are cancelled before they ever launch.

The whole thing is a messy jumble, representing the fact that a lot of these modules had originally been intended for several separate national or slightly-less-international stations, before international agreement unified them here. The Russian modules have a weird scattering of unrelated names, and I don't know why two of them both translate as "dawn". The American modules have a clear theme of abstract states of being ending in Y. The European modules are named after dead old white guys (or possibly teenage mutant green guys; see official MPLM patch below). And the Japanese module kind of fits in with the abstract sentiments of the US modules. Taken together as a whole, it's not a great use of names, there's no apparent logic to it and I'd probably be happier if they'd kept to plain, literal names for the individual modules and saved their creative juices for the station as a whole.

Logo for the Multi-Purpose Logistics Modules Leonardo, Raffaello and the unused Donatello.

Astronyms, part 4: Project Apollo, Skylab and the last US Condoms

This is part of this series of posts on the history of spacecraft naming.

I found a bit of a kindred spirit in Curtis Peebles while writing this, as I stumbled over his 1978 summary of US astronyms. It's like a more concise (and outdated) version of what I'm trying to write here, but he does have a few factoids in there that no other source has been able to give me. It's surprising how little interest there is in Apollo vessel names after Eagle. He also gave me Ladybird to add to my Gemini names, plus a few clarifications of the intent behind some spacecraft names. See 'Astronyms, part 3' for those updates.

Apollo Command/Service Module and Apollo Lunar Module
(Space condoms, or arguably space condom and drop ship. In space service 11 October 1968 to 24 July 1975, and 3 March 1969 to 15 December 1972)

Apollo Command/Service Module

Apollo Lunar Module

Strictly two completely separate vehicles, it's pretty much impossible to discuss the spindly Apollo Lunar Module without discussing the Apollo Command/Service Module (a giant three-person Gemini) that always teamed up with it. Apollo CSMs flew a few missions without LMs, but never LMs without CSMs. You might even say they were designed that way. The Apollo CSM was also a new thing for NASA in that all the Mercuries had flown under Project Mercury and all the Geminis had flown under Project Gemini, (and all the Apollo LMs were flown under the Apollo Program), but the Apollo CSMs were put to use in three different projects, pairing up with different partner vessels in each.

The Apollo Command/Service Module was to the Gemini what the Gemini was to the Mercury: Bigger, more advanced, and with a crapload more stuff dangling off the butt in a jettisonable module. In Gemini, this was called the Equipment Module; in Apollo, it was the Service Module. Apart from just being bigger, it included a big go-forward rocket, in addition to all the little navigational thrusters, so that it could push itself back to Earth from the Moon. That, paired with the crewed bit known as the Command Module, was jointly called the Command/Service Module. Why Command? Because it was intended to be the mothership for the Lunar Module. The LM was also made up of two halves, a descent stage (all rockets and legs) that carried the whole vessel to a safe landing on the Moon, and an ascent stage (more rockets and astronauts and computers that are famously less powerful than even the most basic ones we ever use today) for flying away from the Moon's surface and back to the orbiting CSM. I'm undecided, based on my definitions, if this counts as a condom (because it's only useful for a single mission) or a drop ship (because within that mission, it does both land and launch again). I suppose it'll have to sit uncomfortably at the edge of one category or another until a more perfect scheme of categories is created. The LM was originally the LEM, or Lunar Excursion Module, but apparently NASA management felt that the word 'excursion' made it sound like a vehicle for going on Sunday picnics, and so quietly dropped it from the name. That's right, back in the '60s they dropped an E. Astronauts still pronounced LM as "lem" though, as it rolled off the tongue better.

Piloted Apollos were always paired up with the Saturn family of rockets, either the smaller Saturn IB or the enormous Saturn V. So, following established NASA naming, the missions got compound module-rocket names, i.e. Apollo-Saturn (though for some reason, they were also named the other way round in some contexts, as rocket-module, or Saturn-Apollo). To distinguish between the different Saturn variants, the mission number was lightly coded, with 1xx for Saturn Is (and the xx filled by a sequential launch number), 2yy for Saturn IBs and 5zz for Saturn Vs (where yy and zz are different sets of sequential numbers, not the mixed jumble of Project Mercury). Then there were also the simpler, more public mission names that were just Apollo 1, Apollo 2, etc., but these got a little confused by the Arabic numerals for the Project Mercury Missions and the Roman numerals for the Project Gemini missions, and so the Project Apollo missions were sometimes seen with either of the two types of numerals. Also, each CSM and LM had its own production number. So Apollo 10 (also written as Apollo X) was also AS-505 (or SA-505), which included CM-106 and SM-106 (jointly known as CSM-106, or Charlie Brown) and LM-4 (a.k.a. Snoopy). The individual bits of this bureaucratic coding system all mostly make sense when viewed in isolation, but together they do make quite a mess.

The name Apollo itself, after the complicated Greco-Roman god, was chosen to be impressive and commanding and dynamic, with visions of galloping horses and blazing suns (in summary, Blazing Saddles). It's kind of hard to impartially judge this choice any more, since it's become such an iconic name. Does it sound good because it is, or because it's so hard to imagine it being anything else? It's like trying to judge whether the Mona Lisa is a good painting or not; there are just too many preconceptions in the way.

Apollo 1 never flew, the first mission NASA opted to completely skip as a sign of respect, after its crew were killed in a training accident. Heading that crew was Gus Grissom, who had named Liberty Bell 7 and the Molly Brown, so apart from all the real tragedy, it's a minor tragedy that we never got to see what he'd name a Moon mission vehicle. There were also no official Apollo 2 or Apollo 3, but these code number-only flights, along with Apollos 4 to 6, were used for unpiloted test flights. Then came Apollo 7 and 8, the first piloted missions, using only the CSM, to check that it would work at all. Apollo 8 became the first human vessel to orbit the Moon. Apollo 9 was the first mission for the LM, and this created a radio communications problem, since CSM and LM sounded too similar and risked identity confusion over a crackling radio, while their full names were too long and awkward, especially when you consider that every new bit of radio chatter was required to begin with "[receiver's callsign], [sender's callsign]", e.g.:

"Houston, Apollo 11. We're in process of maneuvering to P23 in desired attitude. It likes roll 8.37, pitch 61.33, and yaw 339.87. Over."

That was all good and well while the CSM and LM were docked and acting as a single unit, but as soon as they split up, you'd have to give them each different, practical callsigns. And so NASA was forced to drop the Molly Brown ban and unique vessel names returned, still chosen by the crew, but with some management oversight. And so Apollo 9's CSM and LM were respectively named Gumdrop and Spider, simple names pointing roughly to the shapes of each, which was not very sexy, but quite practical. The NASA PR people still weren't happy with this, but were overruled for practical reasons.

Apollo 10, as I've mentioned here before, was given a more lighthearted pair of names, CSM Charlie Brown and LM Snoopy, because NASA had a weird thing back in the late '60s for Peanuts. I love those names, but it's just occured to me how much more awful they would have made things if the mission had gone wrong. Imagine a newsreader having to report, "NASA today announced that it believes that Charlie Brown was destroyed in an as yet unexplained explosion during manuevers in lunar orbit. Snoopy remains missing, but may have been at a safe enough distance at the time. Even so, there remains no hope of bringing Snoopy home." The funny thing is, in reality, Snoopy's ascent stage probably remains the only one of its kind launched into space that still exists intact, in some unknown solar orbit. Those seen in museums today are all spares that were never launched (or replicas), while all of the other 8 active LM ascent stages were all crashed into either the Earth or Moon. Of course, the surface of the Moon is still home to the 6 used descent stages from the successful lunar landings.

Apollo 11's crew was originally going to go back to simple, descriptive names, Snowcone and Haystack, after NASA management decided that lighthearted was not a style they wanted to embrace. But then someone realised that, "The Haystack has landed," might not be a very good line to put in the history books, so more interesting but still serious names, Columbia and Eagle, were chosen instead. And from then on, NASA's unique vessel names were almost all kept serious and sombre, unfortunately. The official convention became that the command module pilot got to name his own CSM, while the other two crewmembers had to agree on a name for their LM; there does seem to have been a bit of inter-crew discussion and name sharing, so these wouldn't have been purely personal, unadulterated name choices. Allow me to go through the whole list of unique names and the reasons given for them.

• Apollo 7 CSM: No name, but Phoenix was considered, refering to the mythological fire bird, before being ruled out as too stark a reminder of what had happened to the Apollo 1 crew.
• Apollo 8: No name, but Jim Lovell had wanted to name their Saturn V rocket the Columbiad after the space gun that launched Verne's fictional lunar expedition. This would have been the first real spacecraft named after a science fiction one. The name was instead given to the Apollo 11 CSM.
• Apollo 9 CSM: Gumdrop. Approximate shape of CSM.
  
• Apollo 9 LM: Spider. Approximate shape of LM.
  
• Apollo 10 CSM: Charlie Brown. Named after Charlie Brown, to go with LM Snoopy. Second spacecraft to be named after someone named Brown. (The only real spacecraft I've seen in person so far.)
  
• Apollo 10 LM: Snoopy. Named after Snoopy, NASA safety mascot.
  
• Apollo 11 CSM: Columbia. Named after the Columbiad space gun from the Verne story that Apollo 8 had considered using. This is not quite the same root as the shuttle Columbia's name, but I'd guess the -d was dropped from the end of Columbiad to bring it closer to the same Americany sense. I can't find any pre-shuttle claims that this CSM was named after the same sailing ship Columbia that the shuttle was, so I think that assertion may be an erroneous back-connection.
• Apollo 11 LM: Eagle. Named after US national bird, the bald eagle. Oddly, the mission patch with the bald eagle was designed first, and the LM name taken from that. The module and the site it landed on were also subsequently named Tranquility Base, which Armstrong seems to have improvised, but which is now the official name of that place. This was not repeated with other LM landing sites.
  
• Apollo 12 CSM: Yankee Clipper. Probably named after the type of sailing ship.
• Apollo 12 LM: Intrepid. Probably named after one of the US Navy ships by that name.
• Apollo 13 CSM: Odyssey. Named after the movie 2001: A Space Odyssey.
  
• Apollo 13 LM: Aquarius. Named after the constellation and the song named after that.
  
• Apollo 14 CSM: Kitty Hawk. Named after Kitty Hawk, North Carolina, where the Wright brothers first flew.
  
• Apollo 14 LM: Antares. Named after the star, which was a planned navigational aid for the mission.
  
• Apollo 15 CSM: Endeavour. Named after HMS Endeavour, the ship commanded by James Cook. This is the same root as the shuttle Endeavour's name.
• Apollo 15 LM: Falcon. Named after the mascot of the US Air Force Academy. Carried an actual falcon feather, as seen in this video.
  
• Apollo 16 CSM: Casper. Named after the cartoon ghost, apparently chosen because Mattingly was reminded of that sort of cheesey ghost image by the white space suits they wore. I'm not sure how they snuck in another lighthearted name like this.
  
• Apollo 16 LM: Orion. Named after the constellation.
  
• Apollo 17 CSM: America. Probably named after the short version of the USA.
  
• Apollo 17 LM: Challenger. Named after HMS Challenger and its Challenger Expedition. This is the same root as the shuttle Challenger's name.
• Skylab 2 CSM: No name, see below.
• Skylab 3 CSM: No name, see below.
• Skylab 4 CSM: No name, see below.
• Apollo 18 CSM: Apollo. The US half of the Apollo-Soyuz Test Project, see below.

But are there any nice patterns in there? Not really. A bit of US nationalism (watered down a bit by the ones named after British vessels) and a few unrelated astronomy names. A stray mention of a real place on Earth. A few fictional characters and one fictional spacecraft. Charlie Brown and Snoopy were the only pair given names that work together well as an obvious, themed pair, which seems like a missed opportunity to me. Yankee Clipper and Intrepid both refer to boats, since they had an all-Navy crew, but that's a much thinner connection. Odyssey and Aquarius were both movie references, but slightly disguised. Apparently one possible pair of names considered for Apollo 11 was Romeo and Juliet, which might have set an interesting precedent.

After the Moon program was ended early, the station Skylab (see below) was launched using a modified Saturn V, and Apollo CSMs were used to ferry crews up to it, in much the same way that Soyuzes ferried cosmonauts to and from Soviet space stations. Skylab itself was considered the Skylab 1 mission and the three Apollo CSMs that flew up to it were designated Skylabs 2 to 4. They had no unique names, as the station and the CSMs would only ever be occupied in space one at a time, so there could be no radio miscommunication. They were also known erroneously as Skylab I, Skylab II and Skylab 3, both numbered incorrectly and with the return of the annoying mix of Roman and Arabic numerals.

Finally, the very last Apollo CSM to fly was the US half of the Apollo-Soyuz Test Project, which was much condemned by scientists and engineers who would have prefered to use it as Skylab 5 instead, but which was an important political statement that helped end the Cold War and laid the technical and political foundations for subsequent US-Russian cooperation on Mir and the International Space Station. This CSM, officially but very rarely called the Apollo 18 mission, simply went by the name Apollo (no numbers or anything) and docked with the Soyuz 7K-TM (a variant customised specifically for this mission) that was officially Soyuz 19, but which also went by the plain callsign of Soyuz (making it arguably the first piloted Soviet space condom to receive a proper unique name). And with the end of that mission, the US completely stopped using space condoms for decades.

15 crewed Apollo Command/Service Modules were launched into space, flying once each. All were successful, except for Odyssey (Apollo 13), which had its famous fuck-up, but which was still non-fatal.

9 crewed Apollo Lunar Modules were launched into space, and all were successful. 6 successfully landed on the Moon, with only Aquarius (Apollo 13) failing to do so as planned, but only because of the failure of Odyssey.

Skylab
(Space station. In space service 26 May 1973 to 8 February 1974)

Skylab docked with Apollo CSM

Skylab was, for all practical, engineering purposes, part of the Apollo family. The Americans' first space station was itself made out of a converted Saturn IVB rocket stage, a left-over from cancelled Moon missions, its hollow tube providing lots of room inside for people and gear. It was launched on the last Saturn V to fly and it was serviced, as noted above, exclusively by spare Apollo Command/Service Modules, docking with the station the same way they would with an Apollo Lunar Module. All of this meant that ground staff and astronauts needed little conversion training. Even so, Skylab had its share of technical problems, almost failing to get started at all and then killing itself off faster than necessary. It's interesting to think what might have been if it had stayed up long enough to become a regular shuttle destination. It did get some good science done, though, and showed the Americans the value of a long-term space presence over short hops.

Sadly, the name Skylab marks NASA's turn towards the dullest, most literal sort of naming, where there's no creativity, no subtlety and nothing interesting about the names. As it happens, I'm ok with the name Skylab, as it's at least an attempt to make a new word, and Sky- isn't quite as literal as the Space- in the later Spacelab shuttle modules. But they could still have done much better. The silly thing is, there'd been a whole name-making committee to find a name for the station, which the name-choosing committee completely ignored.

Skylab operated well enough until it was finally de-orbited and fell to Earth in pieces, not in the Indian Ocean just south of South Africa, as planned, but in Western Australia. A second Skylab was built but never launched.