One more generative AI rant for the pile

(this one's about summarising text)

LLM chatbots – that is AI, in the same sense that we could just start saying "doctors" to refer specifically to orthopaedic wrist surgeons if we collectively decided to – 

LLM chatbots continue to slosh about the world. I used to try them out intermittently to see if they were any good.

My contact with the technology is only incidental these days. To wit:

• If you google old phrases and terminology in English, a LLM chatbot will still confidently weigh in with completely spurious "definitions" because they're not well-represented in the training data.
• If you google modern bits of even slightly less-discussed technical knowledge like "does a Kickstarter project video appear on the prelaunch page", a LLM will still confidently tell you the opposite of the truth.
• If you need customer support or anything that even looks like customer support, there is an extra quarter-hour minimum of wasted bot effort before you can get it.

Nothing I've seen has suggested the technology has fundamentally changed.

 

He can't be wrong, he writes so confidently.

 

In the previous edition of discussing the emperor having no clothes, I mentioned  

[Wikipedia] editors pointed out that the LLM summaries generally ranged from 'bad' to 'worthless' by Wiki standards: they didn't meet the tone requirements, left out key details or included incidental ones, injected "information" that wasn't in the article, and so on

and 

bureaucratic wonks note that genAI can't summarise text. It shortens it and fills in the gaps with median seems-plausible-to-me pablum. The kind you get when you average out everything anyone has ever written on the internet.

I recently saw an AI booster shuffle their position back to "at least it's good for summarising, it's going to completely replace human effort there". With that motivation, let's drill down a bit.

In (a) summary

Let's not bury the lede. LLM chatbots can't produce good summaries. Sometimes by chance yes, but not reliably. Summarising, like everything, is a skill-based task, and of the various capabilities required to do it well, LLMs lack four of the most important.

1. LLMs won't reliably retain important structure or order in which information is presented. They will just haphazardly obliterate implicit linkages. They will even occasionally discard explicit structures, as when the text itself points out that C follows from A and B, and therefore D.

2. LLMs can't identify the most important information in a text (a necessary first step to preserving it in the summary). In a good summary, certain content "should" be retained, certain content compressed, and the remaining content discarded. Vital information generally isn't identified within the text in a way that's detectable without broader context, language skills, and understanding of the world. Even when it is, e.g., in texts where repetition of a word corresponds directly to importance, or phrases like "this is vital information" are always appended, LLMs still aren't guaranteed to retain important details! And the same applies to cutting out unimportant information.

3. LLMs can't stick to the source text, that is, the content they're meant to be summarising. Because they just generate text (by predicting which bits of text should come next, based on an enormous model of which bits tend to come after which bits, hence 'language model'), there's no internal representation of Things 'In' The Language Model versus Things 'In' The Text To Be Summarised, and no impetus to perform computational operations that keep them separate where appropriate. All of which is to say that as well as not including things that should be in a summary, an LLM will readily include things that shouldn't be. Oops

3(corollary). That includes things that aren't true. Oops(corollary)

4. LLMs will sometimes just negate statements for no clear reason. When processing text, e.g. when directed to "summarise", they'll turn a claim into the opposite claim. I think what's going on here is that a statement and its negation are syntactically and semantically similar, even though their meanings are devastatingly dissimilar. Too bad LLM technology doesn't get meanings involved, instead just taking a probabilistic walk through a model of language features like, oh I don't know, syntax and semantics!

Note what these four crucial capabilities have in common. It's the reason why LLMs can't do them. That's right, they require understanding to do properly.

Or if not understanding, then at least computational models of understanding, like formal reasoning over symbolically-encoded domain knowledge including useful axioms. I mention this because classic AI systems (planners, searchers, problem solvers, etc) can do just that, in their various limited ways. They symbolically represent domain information and then perform operations on those symbols which can then give something potentially useful back once related back to domain information.

And those systems are limited, yes, but LLMs don't do 'understanding' at all. As far as I can tell, on the back of a postgrad compsci degree and a few days spent reading and partly understanding the computational basis, this is a fundamental limitation of the technology. One which can't just be fixed, but which would need a whole new (at most LLM-inspired) technology to overcome. For exactly the same reason why AI "hallucinations" can't be fixed.

 

Presummary (a digression)

This technical basis of how LLMs work also explains something else. These chatbots are particularly bad at "summarising" documents which contain surprising content.

By surprising content, I mean...

➡️ Statements seeming to defy common wisdom. Things that are the opposite of statements well-represented in the training data. When X is generally true of a field, but your text describes how ¬X is true of some narrow subfield or specific context, you'll see an LLM "summarise" X into ¬X more frequently.

➡️ Deliberate omissions of things that are usually in correlating training data documents. If your text looks like a text of type blarg, and blarg texts in the training data typically report on X, but you have not reported on X for your own reasons, an LLM is likely to just make something up about X while "summarising".

➡️ Unusual pairings of form and content. Performance degrades the more you ask an LLM to do something novel.

➡️ Context-sensitive language like metonyms and homographs. When X is a big important noun well-represented in the training data and X refers to something else in the text, you'll see an LLM (appear to) get confused by the statements about X its produces for the "summary".

➡️ Nontextual information content. The LM stands for language model. If you have a report that includes and discusses images and diagrams, a chatbot might be able to stop and parse those, and then incorporate its own description of the image as part of the text to be summarised, and maybe even put images back in the summary. But you'll nonetheless end up with a worse output.

 

In summary (but for real)

So LLM chatbots can't be (consistently, reliably, etc) good at summarising.

Of course people who don't know what a good summary is might not notice this; likewise people who possess the skill but don't carefully check the job they told it to do.

(I would argue that in either case, if the task was worth doing to begin with, you should prefer the task not getting done to having no idea whether your document is a good, adequate, or terrible summary)

Anyway this is why you may have seen people who do know what a good summary is point out that LLMs actually "shorten" text rather than "summarise" it. I'm not certain but I think the first time I saw this was in one of Bjarnason's essays.

The sentiment "this technology sure can't do [thing I am skilled at] for shit, but I guess it might be good at [thing I don't know about]" will continue to carry the day as long as people let it

I'll self-indulgently close by quoting myself again:

A lot of people with a lot of money would like you to think that genAI chatbots are going to fundamentally change the world by being brilliant at everything. From the sidelines, it doesn't feel like that's going to work out.

1d20 Megadungeon Safety Code Violations

It's almost as if the archlich-in-chief didn't have everyone's safety in mind!

(Lousy no-good penny-pinching archlich-in-chief.)

 

 1d20 Issues The Inspectors Bring Up In Their Initial Report:

1. Sixteen cases of poisoned darts shooting from walls without warning signage 
2. Lit torches in corridor RM-70-B burning within two metres of hanging tapestries
3. Thirty-storey staircase lacks railing or other cordon around open stairwell
4. Inadequately ventilated throughout (see attached list of 781 affected rooms)
5. Chasm bridge constructed from inadequate materials. Code requires use of steel cable, tethers
6. Insufficient drainage ducting to prevent floods on floors B2 through B29  
7. As-built construction plans yet to be lodged with regional fire rescue co-ordinators
8. Shrine Of Darkness should be electrically insulated against celestial lightning
9. Enormous boulder insufficiently shored up with mechanical apparatus; risk it could fall and roll
10. Hellfire conjuration chamber not adequately equipped with hellfire suppression equipment
11. Giant mushrooms pose unacceptable spore allergen/asphyxiation risk
12. Several parts of floor B30 are exposed to magma flows (not a safety code violation; brought up as a point of overall concern)
13. Unholy Water Cistern not shielded against heavy metals found in local groundwater 
14. Too few first aid kits (three per floor; code calls for five)
15. Human remains not maintained at morgue temperatures, and allowed to move around freely
16. Floor B7 torture chambers not accessible by ramp or elevator
17. Cursed font of immortality lacks fence to prevent accidental drownings
18. Black, yellow, ochre, blue, umbral, and invisible mold found in food preparation areas
19. High-visibility safety lines should be painted in zones where ambulatory juggernaut roams
20. Giant talking stone head is blocking fire exit

 

 

(Yikes. Well at the pace stuff gets fixed around here, better hope they shut the worksite down before someone gets hurt.)

Smooth vs Chunky game design

Sometimes it helps to look at game design through the "Smooth or Chunky" lens. What do these terms mean? Well, they encapsulate certain vibes. They occupy the middle ground between game mechanics and game feel.

🥛🥛🥛 Smooth is

• small numbers changing incrementally
• things being at the same level
• player-facing subsystems with modest game impact (say, situational dice roll modifiers)
• overloaded dice rolls with small results
• rounded probability curves like 4d6
• anything pre-planned
• predictable consequences, with randomness used as a spice

Smooth is like gradient descent and intricate clockwork and elegant flowcharts. Typical Smooth design elements are sensible, explicable, predictable, fine-grained, subtle, simulationist, world-associated, and introduce enough complexity as they need to be internally coherent.

🥜🥜🥜 Chunky is

• one dice that does everything
• flat high-variance probability curves (say, a 1d100 roll)
• one number that represents a bunch of things
• rollercoaster rides of remarkable successes and sudden catastrophes
• randomness underpinning creative direction
• huge sudden changes to big important things
• hefty modifiers 
• random tables

Chunky is like high stakes roulette wheels and staccato noises and refusing to erase anything. Typical Chunky design elements are simple, wide-reaching, experimental, flashy, coarse-grained, gamist, surprising, central, and do as much as possible with one thing.

Now in game design, unlike with peanut butter, neither one is clearly better than the other. It's situational.

Case study 1

I'm working on a RPG called Overzealous. In this game, you're a well-meaning elder god but your cultists are the typical bloodthirsty crazed zealots. This sets up the tension. You need to become manifest in the world before your cult tears itself apart with ridiculous behaviour.

Early conceptualisation had Overzealous revolve around large random tables, with most of the consequences falling out of those. This played fast but felt random. And it meant that I couldn't model persistent problems, like your cultists getting bored and summoning a bunch of monsters that then hung around, or continuous schisms in your cult leading to further attrition and outrage.

I Smoothed out this Chunkiness by expanding two numerically-tracked stats to five, and adding a granular subsystem for acquiring ongoing "problems" which took trade-offs to solve. The gameplay became a lot richer!

Case study 2

After these changes to Overzealous, you have three "bad" stats (Fervour, Divergence, and Monstrosity) which you want to keep a lid on and two "good" stats (Imminence, Cultists) which you need to get high enough that your cult can perform a ritual to bring you into reality.

In the draft version after the Smooth changes, if a bad stat exceeded 20, it's game over. The bad stats crept up slowly through various mechanisms, all at about the same pace. This increase was slightly faster than you could deal with, which you generally attempted to do by sacrificing positive stats, setting yourself back. That's Smooth design! The intention was that the player needed to find and pursue a good strategy to secure a win before a loss became inevitable.

In practise, though, my introduction of this much Smoothness created two issues. Because the changes were small and fairly predictable,

1. A skilled player could find one obviously optimal strategy, and didn't have to deviate much from that strategy for random events. This reduced gameplay scope.
2. If through poor luck, lack of experience, or exploring other options, the player's negative stats got too high, there was a tipping point where it was obvious that a loss was inevitable... but it took a long time to lose the game.

To deal with this, I eased up on the Smooth pedal and re-introduced some Chunky. I had stats go to 13 instead of 20, took out some of the cases where multiple stats all change by 1, and doubled down on cases where a single stat change by 2 or 3. Also, the ongoing problems that beset your cult (like cannibalism, diabolism, and schisms) only have a chance to come into play each turn, but are more impactful when they do.

And these tweaks got me exactly the gameplay experience I wanted for Overzealous! There's no longer one clear strategy to follow, as the feeling is more one of running around putting out fires. Now the player is tempted to push things, to e.g. just spend one more turn scrambling towards completing the ritual when their Fervour has crested 10, because the end's in sight, knowing that a couple of bad rolls might be their downfall.

New words to use

Was I thinking about Smooth vs Chunky throughout this design process? Not exactly, but I was certainly aware of what was going on with the overall vibes, and now that I've constructed this jargon to talk about it, I suspect I'll be thinking in those terms in the future.

With gameplay pretty much where I want it, I'm ready to get the visual design finished!

Anyway, that is the age-old "Smooth vs Chunky" dichotomy, brought to you by peanut butter on toast.

When player agency meets character compulsions

When you're playing a role, sometimes your character ought to act in some way.

• The goodly hero has a moral code regarding the treatment of prisoners
• The addicted wizard would do anything to get more purple poppy
• The impulsive barbarian wants to destroy doors, spines, and magic items
• The traditionalist dwarf can't rein in her hatred of goblins or desire for gold
• The life priest's role demands self-sacrifice if it comes down to it
• The emergency medic has a duty of care to her patients
• The brash brute can't keep his foot out of his mouth

These traits create roleplaying obligations. The obligation is on the player, constraining their choice down to how their character "should act".

It often happens that a character who acts on an obligation risks getting a negative outcome – perhaps even a catastrophic one. And this may be obvious to the player. So it's natural that players won't always want to have their character act as they "ought" to.

An aside: I'm talking about classic roleplaying games here, where character success amounts to player success. In storytelling games, or even in RPGs played as if they were storytelling games, players tend to be much more open to making deliberately "suboptimal" in-character choices.

At the game table, whose job is it to ensure that obligations are followed through? The rules? The GM? The social dynamics of the group?

The strength of an obligation – how much a player "should" follow it – varies depending on the character, player, group, and game system. For example, in a point buy system where a burdensome personality trait compensates you with extra points, there's a strong expectation that the trait's obligations will be followed. In these cases, there might be rules laid out to cover roleplaying obligations.

Compulsions versus obligations

Now compare in-world things that the characters might encounter like magical geases, mind control rays, charm spells, and brain parasites. These essentially do the same thing. The player character is compelled, rather than obliged, to act in some way.

Of course, modern games almost always structure these things in a meticulously rule-based way, with various "outs", and make them short term and/or weak.

That's because it's a truism of RPGs that it sucks to have your agency taken away.

(Your agency in particular; but your character stuff in general)

After character creation, a traditional TTRPG offers the player the ability to do exactly one thing: make in-world choices on their character's behalf. They can't choose the corresponding outcomes, or odds, or stakes, but they can choose what to try.

Undermining this one capability – even for the sake of in-world things, like established character compulsions or magical mind control – is generally a bad idea. It feels like an attack on autonomy and agency.

So we want to support situations where, through obligation or compulsion, players ought to have their characters act certain ways. But we don't want to take away their ability to decide.

 

How do we square these two things?

Player choice is pretty close to an unequivocal good, making constraint on player choice pretty unambiguously bad. My favourite approach just turns the bad thing into more of the good thing.

➡️ Let the player decide whether their character resists being obliged or compelled, at some in-world cost.

Let's say the player is confronted with some situation where their character is compelled or obliged to act some way. Now they decide whether the character is able to resist, suffering some consequence if they do.

"You resist the call of the purple poppy, at great personal cost. You'll have a morale penalty until you next imbibe it."

"Straining with every part of your soul, you overcome the suggestion spell. You feel a huge fatigue upon you."

"With muscles and tendons clenched, you give the goblin merchant a forced smile and polite nod as you pass. Something in your jaw clicks. Take 2 points of damage."

What is the in-world cost?

The consequence of managing to resist a compulsion/obligation should ideally

• be mechanically grounded (weighty and yet easy to apply)
• be player-facing (rules-based, or at least announced by the GM in advance)
• have an immediate impact (the consequence feels associated with the decision)

Obviously this is easiest in gritty detailed systems with character resources. Stress, Stamina, Focus, whatever it's called as long as it's measured in points that can be taken away.

What if my game doesn't have an appropriate stat?

You could improvise one. 

Or the character could suffer other in-world consequences.

• For an unmet abstract obligation like following a code of conduct, the character might well encounter social problems. Rumours spring up. Hirelings abandon the party. Opportunities dry up. The parent organisation sends a formal rebuke or starts an inquiry.
• For other obligatory traits, the character gets stressed/worried/grumpy/distracted when they don't follow their nature, and end up mislaying an object, getting a bad reaction roll, missing details when looking for clues, etc.
• When an external compulsion is resisted, the character might be disoriented and flub things for a while (rerolling successful rolls, etc).

A specific suggestion

I have long thought that the perfect RPG is one that rewards the PCs with a dice bonus/malus for getting a good/bad night's sleep. I wrote about this the other day. Use a general 'well-being' stat measuring a character's overall mental and physical condition, which the players can push to try to improve, and which modifies most if not all rolls and tests.

This is a perfect candidate for tying in to this choose-to-pay-and-resist-compulsion system. If you decide your character resists, that bonus is reduced (or malus increased).

Otherwise

You could flip this around and offer a player character experience points (etc) for giving in to their obligation. But this doesn't feel the same, for a couple of reasons.

• Offering a mechanical benefit to specifically "good roleplaying that goes against the characters' best interests" will rankle.
• Paying an in-world cost to overcome a compulsion, or choosing to let the character act on it, make it feel like a thing actually having an effect. Just declining a prize feels like nothing happens.
• Players tend to be risk-averse when it comes to stuff like this, so you'll end up with the character whose character quirk never actually gets them into trouble, and the villain's charm spell that never actually does anything.

Other benefits

Sort of buried the lede here, but the big advantage of doing choose-to-pay-and-resist-compulsion is that you can hang lots of stuff off a single mechanic. So there's obligatory character traits, but also...

Thing #1: Mind control. As mentioned, taking away agency is usually unfun, can disrupt a game, and can be too swingy in a tactical combat game (strictly worse than save-or-suck); modern games usually make control spells very weak and brief as a result. (Compare OD&D, where the level one Charm person spell could control someone indefinitely, and even the first supplement only reined it in to the point where a particularly lucky superintelligence might shrug it off after a few days)

Suppose instead that the subject of the magic really feels the compulsion deeply, and can resist at some steep cost. And they can be effected by the spell and decide the point where it wears off, as the 'price' to escape goes down over time.

Thing #2: Other magical effects that subvert character control. Illusion, fascination, terror, berserk rage, altered perception, foolishness, cursed objects that whisper dark thoughts. Games tend to have detailed subsystems for these and/or be pretty weak, because of what they do to characters. Using choose-to-pay-and-resist-compulsion means a simpler, unified system, potentially greater effects, and more player agency through the whole thing.

Thing #3: Social-fu systems that can be used to influence PC action. Games are understandably reluctant to make social skill rolls have actual impact by influencing, let alone determining or constraining, PC choices. Now we have a framework where players will have to decide the lesser of two evils: go along with social influence, or pay the cost (become emotionally run-down, sapped of energy, have low morale, etc).

Thing #4: The same choose-whether-to-pay kind of tradeoff, but introduced to other parts of a game. Possibly as a single unified mechanic.

I'm thinking in particular of things like saving throws and luck points, which aren't very nuanced and involve no in-world decision-making. What if instead you could reliably avoid disaster by suffering some other major long-term consequence? "You fling yourself desperately away but hurt a tendon and can no longer stand up until you receive healing." "You block the blow but are winded and can't succeed on strength rolls until you rest." "You catch a branch on the way past but a strap snaps and your satchel disappears down the mountainside."

Thing #5: In a crunchy character-building system, we now have new options for a character's (in)ability to resist various things. Appropriate traits like "stubborn" or "weak-willed" or "dedicated to the cause" can vary up the cost that a character pays to resist particular obligations or compulsions.

Where'd this idea come from?

I don't remember; I suspect I saw at least the kernel of it somewhere. 

➡️ Lots of games have meta-resources, and sometimes trade in them directly. In Numenera, a GM can suggest something bad may happen and an affected player can either gain XP and let it happen, or lose XP to deny it. Related but not my preferred approach, and not specific to obligations/compulsions.

➡️ In that vein, player-facing story-gaming mechanics may directly affect character actions and/or overall narrative.

➡️ There's this blog post which I found quite compelling. It has a somewhat similar approach to the problem of combat stunts: https://oddskullblog.wordpress.com/2021/11/15/combat-maneuvers-the-easy-way/

➡️ I vaguely remember seeing a GURPS house rule or hack where Disadvantages that compel action can be paid off with FP. I don't recall, but it would make sense to vary the FP cost depending on whether the compulsion was mandatory or resisted with a self-control roll, or more generally if it depended on the point value of the Disadvantage.

If there's a notable/obvious game that does this, let me know!

Winning the game by being well-rested

I have long believed that the perfect RPG would reward the PCs for getting a good night's sleep – and penalise them for a bad enough one.

There are many upsides to doing this.

1. It's realistic. In my experience, getting particularly good or bad sleep could have a +/- 50% effect on your baseline competence.
2. It encourages roleplaying. People want to get a good night's sleep. Now characters want to get a good night's sleep.
3. It incentivises investing in comforts and nice-to-haves. Without the carrot and the stick, most players tend to default to only getting functional equipment.
4. It adds more choice to campsite selection. More choice is better. (I think it was Peter D who suggested: concealed, secure, or comfortable; pick one)
5. You can tie it to character injury. In HP-based systems, characters tend to either be fine or dead, resulting in confusion about what HP and damage really "are". Well-restedness is pretty much a score of overall condition, so being wounded can reduce the same score. Now it's actually more difficult to jump around on a stabbed leg.
6. You can tie it to character morale. Generally, telling players how their characters feel is a bad idea. Telling them they must act some way as a result is disastrous. But a hard rule like "a character who sees a friend die loses 1 point of overall condition" is non-intrusive, assuming you don't allow sociopathic PCs.
7. It incentivises problem-solving. To get the best outcomes, you need to acquire and carry around more creature comforts, even through hard terrain. So now you have a to-do list and some logistics to work out.
8. It encourages having a bunch of attendants (lackeys, entertainers, back-scrubbers, cooks, night watchmen, etc). This is in keeping with commonly emulated genres like "low-technology fantasy", and opens up new kinds of gameplay.
9. It makes the weather feel real. Most of us are well insulated from the effects of weather in 2025. Travelling, sleeping, or making a meal in bad weather is really rough going. 

and

10. It's also a soft progression system!

Characters with more resources can spend them to improve their day-to-day efficacy.

Characters who end the first game flush with cash might be able to get good provisions and nice camping gear, and secure a future +1 when there's fine weather and uninterrupted nights. Later, they might hire some camp drudges so they can get more rest per day, porters to carry their extra stuff, and maybe a proper cook. They're looking to get that +2. By the end of the campaign, expect everyone to travel with fine silks and furs, feather mattresses, bevies of personal chefs and butlers, vintage wines, and magical solutions meant to ensure good dreams, chasing that +5.

(Of course, if you want your normal "levelly" game, you can turn it into a treadmill: characters get used to the finer things, and their baseline expectations slowly creep up, so the bonus never gets too big)

I've had worse naps. (Artwork via www.lustigesrollenspiel.de)

How to get a good night's sleep

I think this metric is dependent on six things:

• Being sated (not hungry, thirsty, etc)
• Being comfortable (resting in a quiet, comfortable place, in shelter, at a nice temperature, etc)
• Getting enough sleep and enough down time
• Having creature comforts (indulging in vices, enjoying entertainment, being with friends, etc)
• Being safe
• Being healthy (not injured, cursed, ill, in shock, etc)

Some games have relevant mechanics (like systems for dehydration or sleep deprivation), which this approach could either subsume or integrate with.

Mechanically, how does it work? Is there an individual character roll depending on the factors? GM fiat? Player-facing prescribed requirements for each tier? The implementation will of course differ with the system.

So will the outcome. Games with a single central dice mechanic obviously get a dice bonus or malus. Then it's just a matter of magnitude and deciding which rolls it applies to (I think you could make a very good case for "all of them").

Final thought

You might have noticed I didn't call this "the INSERT_NAME rule". My thesaurus failed me.

Fettle and Invigoration are awkward. Condition or Vigour would be good but tend to be reserved game terms, so might be confusing. Words like Well-Rested and Energised don't cover the negative cases. Restedness is apt, but clunky.

Freshness is pretty good (except for sounding like you're in a supermarket).

'Consume alien': Reviewing Encounter Critical

...You ever hear of Encounter Critical by Riley and Ireland? A long-forgotten piece of amateur RPG design from the 70s, originally published "on a school mimeograph" and sold through hobby stores, and now available only as a scan.

It's one of those early masterpieces in which fantasy and science fiction tropes collide. And yes, that does mean the currency is "gold credits".

Quest into the slaver kingdoms or hurl yourself into the galaxies of space to find wealth and destiny. Your tactics and your character are yours to control as you undertake ENCOUNTER CRITICAL.

...Of course, it's a joke. This "forgotten gem from 1978" was created by S. John Ross in 2004, with flourishes such as advertisements for an official Gazette (a newsletter in four supposed issues), as well as a reference to "a forthcoming line of Encounter Critical miniatures" and a search for a book deal.

Originally it was cast back in time 26 years from the true 2004 publication. Now 21 years have passed in turn.

How does the hoax hold up?

🛸 The aesthetic is perfectly on point: typewritten; amateur; hand-crafted and hand-drawn; retro; photocopied

🛸 The prose is breathlessly earnest, and preoccupied with its own relationship to conventions of war gaming

🛸 The game is full of sci-fi plagiarism and the kind of dated sexism obsessed with 'seduction' and 'doxies'

🛸 It's the classic nine-stat system we all know and love: Adaptation, Dexterity, ESP, Intellect, Leadership, Luck, Magic Power, Robot Nature, and Strength.

Um

The insane retro creativity of Encounter Critical

You roll for character race. Possibilities include, on the fantasy side, 'Amazons', 'Frankensteins', and Hobbits I mean 'Hoblings'. The purloined sci fi IP includes 'Klengons', 'Planetary Apes', and 'Vulkins'.

You can also play as a 'Wooky' and be penalised for wearing armour, although of course 'a Wooky will seek out magical rings or energy armor when it is available'.

You also have a chance to end up as...

You also may be a mutant, and therefore suffer from character traits such as Cannibal Urges, Allergy to laser, Unusual Sexual Gifts, or Self-Consuming Brain.

The writing mostly serves the "lost 70s indie game" in-joke, but also has its genuine funny moments:

 

Amateur indie design 

The mechanics, while harking back to OD&D, very much like they're just being felt out.

🛸 Warriors get followers and/or animal companions, and an underdetermined number of multiattacks.

🛸 Hit points, damage, and number of monsters appearing, are given as ranges instead of the underlying dice codes.

🛸 The nine stats barely interact at all with the (percentage-based) class abilities, only underpinning the general character abilities.

🛸 Instead, a class's key stats give xp bonuses or penalties. So does Intellect, which means it'll stack for the classes with Intellect as a qualifying stat (pioneer or warlock). If they roll maximum Int they will get +20% experience as well as "a 10% chance at experience bonus", i.e., 10% chance of doubling.

🛸 The terminology is slightly inconsistent.

🛸 The game is stuffed with percentage tables.

In these ways and others, it's all deeply connected to OD&D (while taking sniping pot shots at that game).

The book is laid out as if you pull out the middle pages to use at the table. Pages 23 and 27 are ("mistakenly"?) transposed, breaking the table of monsters.

Which includes extremely compelling types such as: Asteroid Worm, Bee Girl, Dragon of Wisdom, Rogue Robodroid, Sky Piranha, and:

(The entry for the Phasic Wolf helpfully notes that it is "phasic in nature".)

The intended 70s style rings true, IMO

An incautiously designed disease table gives you a chance of getting a brain disorder from sex work. Or you could get the "Pestilence of Dark Withering" or "Curse of Seven Hundred Minds" from a rusty nail.

There is a combat system which somehow combines simplicity, percentage rolls, underspecification, and assuming an understanding of how OD&D did things.

Does the ranged weapon table have "sling", "musket", "tommy gun", and "phasic sniper rifle" on it? Yes. Is it possible to do more damage with the sling than said rifle? Yes.

The text is very persnickety and period-appropriate, with pot shots at house rules and gaming styles, etc. It describes its spells as "correctly balanced", snubs spell levels, and says the authors prefer "a more science fiction approach where a spell is a spell".

(Speaking of spells, I love that a science fiction game has a spell that lets you travel... 500 miles.)

The inscrutably-named abilities are pretty great. Consume alien, Ensorcel, Illicit, Machine friend, and See the future all on one character sheet!

...And Seduce, of course. I've draw a veil over all the gendered stuff, but it's there – doxies, amazons, succubuses, etc. While it's clearly intended as a send-up of the 70s trappings, and works as such, the benefit of another 20 years makes it feels awkward and unnecessary. Rather than making a lot of content actually hinge on it, it could have been presented as a one-note joke.

 

The world of Encounter Critical

In the accompanying adventure, goblins are stealing the brainwaves of abducted girls to fuel up a spaceship in a warlock's lair.

This takes place in the game's setting: "Vanth, a fantasy world of adventure". (Is Vanth just Xanth, but different?)

It comes, of course, with a map:

 

It feels delightfully like the kind of thing I would have been making as an early teen.

My copy of Encounter Critical is from almost twenty years ago. These days you can find it for free on DriveThruRPG. Apparently there's an updated version.

To me, this game reads like a how-it-could have been of the original D&D, like an alternate history Dalluhn Manuscript. The game that resulted from Arneson watching some slightly different films.

Well done.

Out now: Monsters! Horrors & Abominations

 
My new book, Monsters! Horrors & Abominations, is now published!

It's a 215 page bestiary crammed full of weird and flavourful monsters for D&D 5e/2025, like the...
👹 Endlessphant (wrinkled grey elephant's leg with a million knees)
👹 Big scaly one (gormless rubbery hound-god of trolls)
👹 Void oyster (beautiful nacre-petrifying abomination from beyond the stars)
👹 Stillness mage (undead weather-worker and master of doldrums)

Streamlined for combat encounters and delivered with tactics information, modern design innovations, great illustrations, and a touch of humour! Between the base creatures and tons of variants, there are 240 monsters from Challenge Rating 0 to 20.

All focused on the unsettling, underground, unspeakable, unholy, and undead!

Now available here as a PDF or a neat hardcover book!

And as a blog reader, feel free to use this code for 10% off the PDF, or if you would like the hardcover, this code for 10% off the physical book!

Review: Secret Party House of the Hill Giant Playboy

It's Reviewn June! Revebrewary! Octobereview! Janreviewary!

It's August, and here is a quick review.

The adventure 

Secret Party House of the Hill Giant Playboy (hereafter SPHotHGP) is a location-based adventure for Swords & Wizardry, BX, OD&D, etc, released by Jason Sholtis in 2013.

You may know Sholtis from adventures like Operation Unfathomable, as well as blog The Dungeon Dozen. From the latter, years of incomparably high-density weird ideas got fleshed out into a book and then into a sequel. I consider those two of the best books in the entire TTRPG space when it comes to inspiration.

SPHotHGP is 20 pages, released for free. To an extent, it's a pastiche of the classic 1978 module, Steading of the Hill Giant Chief. Some of the set-up is the same, and there are several little nods. For example, in the 1978 module, the hill giant's name is Nosnra (that is, almost Arneson backwards), whereas in SPHotHGP it's Sadrox... who crudely disguises his identity by calling himself Lord Xordas.

Sadrox/Xordas is the Hill Giant Playboy of the title, and his party pad is a den of iniquity and a hoard of much treasure.

As well as giving adventure hooks, Sholtis writes a great background story about how the evil Xordas acquired his huge wealth and party house, although it's not clear how this would become known to the players.

The content 

SPHotHGP is absolutely full of tables. Find the type, nature, extent, and progress of the party. Determine how the PCs are received. Roll to see what kinds of brawl break out, and what giant-sized and gruesome banquet foods are being served: Megatherium done 5 ways. Sweagledactyl, a swan stuffed in a giant eagle inside a pterodactyl.

There's a fascinating array of evil VIP guests and regulars:

• Raver Ylyach the Swamp Hag and her putrescent servitors
• Zogorion, the surprisingly scheming Lord of the Hippogriffs
• Zhemorna, caterpillar-headed high priestess of the Worm Sultan
• Glurt “Beef” Wellington, possibly the world's handsomest hill giant
• The Piper from Beyond Comprehension
• etc 

And Xordas has a very weird treasure hoard indeed.

The presentation 

SPHotHGP is semi-professionally put together. There's a keyed map, and this being for an oldschool game, inline stat blocks for each key being.

I noticed one unfinished page reference, some stray punctuation, and Glok the ogre captain seems to be incorrectly statted. Content-wise, I liked everything except that the characters and complex are largely reactive, with only a few VIPs having guidance for active agendas. I certainly can't complain about getting a good adventure for the low low price of free.

Sholtis's description is terse but evocative. You can practically hear and smell the place.

A few illustrations by Sholtis also help to really tie it together.

 

Normally I'd leave a link for you to take a look yourself. I found SPHotHGP a while ago on Sholtis's blog, but the download has sadly succumbed to bit-rot. I hope the author will repost it some time, because this is great stuff!

Playing Scrabble for keeps

So I've been hooked on Word Play, the Scrabble-based roguelite.

I play some video games here and there, and when I find one I really enjoy, I try to squeeze out all of its challenge juice (technical term). That usually means at least getting all the achievements.

As a result, when it comes to Word Play I have been putting far too many hours into beating Ultramarathon mode specifically: 20 rounds at the most difficult scoring.

Here's how I finally beat it.

(Roguelite players will be completely unsurprised to hear that this did not involve me being particularly good at Scrabble.)

 

Cracking open this game like an egg

It's all about synergies, of course, which means you need luck plus strategy. This run had the 'more rare and legendary modifiers' modifier, and I just doubled down on the first synergy I saw, which revolved around Upgrades.

My engine is made of modifiers:

➡️ I get a random common Upgrade when I play a word with 8+ tiles.

➡️ Each Upgrade gets +1 use.

➡️ +1 bonus point each time an Upgrade is used.

➡️ I gain a refresh when an Upgrade is used up completely. (This was switched out near the end of the run)

So if I play exclusively long words, I get a bunch of Upgrades, and I get more bonus points on every subsequent word. I also get refreshes (to help me get long words), but I don't use them, because so many of the common Upgrades let you refresh selectively. I quickly build up 40 refreshes and a hundred bonus points per play.

Not crucial to the engine, but I also get a modifier for x2 score with 3+ unplayed special tiles. All the Upgrades are turning my entire bag into a mess of special tiles, so this doubles all my scores without effort.

Finally, I get my first potion tile, an 'M' worth 3 points. Potion tiles give you plays equal to their score, but break, when played. I hoarded this until I lucked out and got my final modifier: if you play a four tile word, add a copy of the first tile to the letter bag.

So now, with my huge numbers of refreshes, I can in principle just refresh until I get my potion M, play it at the start of a four letter word, and have it break but add a copy to the tile bag, for a net +2 plays. This is huge when you start the run with 20 plays and only get a few more per round. So: rinse and repeat, interspersing with long words (to get more Upgrades (to get more refreshes)).

In practise, though, that's slow and unreliable. I didn't end up spending many refreshes getting the potion M out there. Instead, I was careful to have an Upgrade slot open at the end of each round, and at about round 12 I got exactly what I was hoping for: the uncommon Upgrade which adds your refresh count to a tile's score.

You can see where this is going. I used it on my potion M three times, discovering in the process that a tile's score maxes out at 99. Now I gain 99 plays each time I put the potion M at the start of a four-letter word. Over the course of a round I gain more plays than I will ever need.

That's why the little number in the bottom right of the screenshot says "1538", not the "15" or so that you would normally expect.

 

Descent into absurdism

At this point the run is essentially won, so I rejoice, but it will clearly be a slog. Even with most of my tiles turned emerald or golden with the bounty of Upgrades, I only get something like 800 points per play with a long word, so I'm going to need to play 100 good words in the final couple of rounds.

Aware of this, I have been burning my essentially-limitless plays rerolling modifiers, and it pays off at the end of round 17. I get the 'multiply final score by number of special tiles' modifier, one of a couple that would reliably boost my scores even further.

So I wave goodbye to 'gain a refresh when an Upgrade is used up', you made all this possible. Now if I spell a word like LAVENDERS, it scores 5936 points. I can and do cruise to the finish in a handful of plays per round.

And that is how I got the hardest achievement in this damn spelling game.

 

Your mileage may vary

None of this strategy is reliably reproducible, of course, due to randomness. But I think it's interesting that it worked, because it was the first Upgrades-based build which I had tried. Part of that is luck in the early rounds, naturally.

Builds that I tried and failed with, for the record:

➡️ All gold tiles

➡️ Fast-growing diamond tiles

➡️ All the emerald synergies

➡️ Double length points, board expanders, and lots of plus tiles

➡️ Dozens of attempts that never got the smallest synergy.

 

So that's most of the challenge juice squeezed out of Word Play! I recommend this game if you're a Scrabblehead. It's on Steam.

—

Update a few days later: I translated my run into "whoops, all wildcards". 300+ tiles of golden and dotted 99-point wildcards took me to round 50.

 

I spent scores, maybe hundreds of rerolls trying to get the "dotted tiles multiplier increases with each play" modifier which would have given me desperately-needed multiplier scaling, and which could have taken me even further. I never got it, though, so I called it at round 50, where winning just meant typing "******************" over and over again and waiting for the scoring to finish.

 

Trying not to be a Gell-Mann Amnesiac

I sometimes wonder how much Gell-Mann Amnesia people experience. Paraphrasing Crichton, when you're a domain expert, you'll sometimes read an article that gets every aspect of your field completely and absurdly wrong, have a little laugh about it... then keep on reading and trusting articles that are about other fields, even from the same publication or writer.

As if they're some pure spring of wisdom which only coughed out a lump of mud when it came to the thing you happen to know about.

It's just an idea from a novelist, not the kind of cognitive bias that's supported by real-world studies that I know of, but you have to admit that it has a kind of... truthiness to it.

Stack this up with Dunning-Kruger and it's easy to become cynical. You might decide that actually, all the loudest voices are talking complete nonsense, all of the time. That might be too far. But I do think it pays to put deliberate hard effort into distinguishing domain experts from overconfident bullshitting pundits.

Now, anyone with their ear to the ground and a weather eye out for Gell-Mann Amnesia should have arrived at the obvious conclusion about generative AI. To wit, that the current state of the technology is that it is an overconfident bullshitter.

On being a piece of software and being confidently wrong

The case studies are easy to find, and the ones from domain experts sound pretty different from the ones from the tech industry and the reporters too busy and/or demoralised to do more than repackage their press releases as articles.

➡️ I am not a historian. The historians I've read say genAI gets softball history questions mostly right and deep ones mostly wrong. Sometimes subtly, sometimes dramatically. It just makes things up when the evidence is scarce. It makes errors of commission and omission as well as having misplaced focus and drawing weird conclusions from premises.

➡️ I am not an artist. The artists I listen to say genAI art looks bland and awful and organic because it doesn't understand composition or anatomy or separate objects (because it doesn't 'understand' anything). It can't make an image that isn't well-represented in the training data, like a camel and a steampunk automaton jousting from the backs of sumo wrestlers. Same in other kinds of media: filmmakers say genAI can't do film because it can't take direction or keep track of characters or have a consistent shot.

➡️ I am not a Wikipedia editor (except incidentally). Earlier this year there was a wretched moment when the Wikipedia editors were going to have genAI article summaries foisted on them, although I think that's turned around now. The skilled editors pointed out that the LLM summaries generally ranged from 'bad' to 'worthless' by Wiki standards: they didn't meet the tone requirements, left out key details or included incidental ones, injected "information" that wasn't in the article, and so on.

➡️ I am not a manager. The managers say genAI can't even collate timesheets reliably.

➡️ I am not a novelist. The novelists say a genAI book reads like a statistical summary of all creative writing anyone has ever done, including all the embarrassing teenage fanfiction. It sucks at originality. And because it doesn't have an internal model or understanding of its outputs, it can't keep track of things and make a coherent satisfying story. Things are vague, tropey, or contradictory.

➡️ I am not a lawyer. The lawyers are, um, well, by the sound of it a lot of them are being sanctioned for using generative AI to cite completely nonexistent caselaw. (☉__☉”)

➡️ I am not a public policy wonk. The bureaucratic wonks note that genAI can't summarise text. It shortens it and fills in the gaps with median seems-plausible-to-me pablum. The kind you get when you average out everything anyone has ever written on the internet. If you try to have an LLM summarise or draw conclusions from a study, it will usually do a bad job, fabricating statements more along the lines of what an average person would guess if they'd only read the study's title.

➡️ I am not a software engineer. The software engineers seem to have mixed opinions. They say that genAI works as code autocomplete (something that has existed for fifty years, but this new kind has pretty sophisticated lookahead, neat). At least some are saying it can't do principled software engineering, it introduces security flaws, its performance drops off for obscure languages, it overconfidently generates bad code, it plagiarises from code repositories that it doesn't have the rights to...

I could go on.

I'm no longer a domain expert in anything, this many years after my stint in academia. I think I'm halfway to being an expert in a few different areas, though. I deliberately concocted some thoughtful questions at the intersection of those areas, just to see.

For example, I asked about the (obvious) mapping of choose-your-path text adventure books onto mathematical graph structures, which the LLM chatbot identified. I followed up with technical questions about the features of those graphs in context: what would the game be like if they weren't digraphs, would you expect cyclic vs acyclic, would a finite state machine be more appropriate and if so why, etc.

And lo, the generative AI output was absurdly, hopelessly, and confidently wrong when given questions that needed expertise.

A lot of people with a lot of money would like you to think that genAI chatbots are going to fundamentally change the world by being brilliant at everything. From the sidelines, it doesn't feel like that's going to work out.

Sometimes I read posts from experts along the lines of

"I've noticed it's almost worthless at [my field], but it sounds like it's pretty useful for [other thing]."

But less so lately, maybe?

So I'm left wondering: are people experiencing massive Gell-Mann Amnesia about these chatbots? Or does everybody know that the emperor has no clothes?

(But oh no, we've invested so, so, so very much money into the emperor's finery, and all the wealthiest people at the imperial court agree: pleeeease could you keep squinting to see this amazing new clothing?)

 

Reviewn June 2025 part 3: Game books

In which I review the books I read in the first half of 2025

Part three: TTRPGs and game content

 

What games have I checked out in the last six months? Here I'm including published books and files, commercial or free, but not blog posts (which I read too many of).

It's Reviewn June!

 

Tiny Spaceship

An alien exploration to understand Planet Earth 

By Tanya Floaker (https://floaker.itch.io/tiny-spaceship)

A wholesome diceless three-page mini-RPG about a spaceship that's come to visit us, and the dangers it faces (from other players, in the roles of e.g. "a big flock of birds"). Has a beautiful cover by August Charters, and the layout is nice (although it would benefit from a proofread and a looser line spacing).

I appreciate the extensive endnote about inspirations and mechanical sources, and the use of CC0. 

Mechanically this is a collaborative story-building game with simple natural-language rules. The tiny spaceship plays off exploration (drawing attention to itself) against understanding (thereby resolving problems). With the caveat that I haven't yet played this, I would call it story-telling gaming done right: a game at a high level of abstraction should be fuelled by simple, abstract, broadly-interpretable rules. Looks fun.

 

RPG Design Zine 

A how-to zine about tabletop roleplaying game design

By Nathan D. Paoletta (https://ndpdesign.itch.io/rpg-design-zine)

Published 2019, it's 28 pages of quite abstract, but useful, advice. The thesis is "I'm not here to tell you which [design decisions] to make, but I do want you to know why you’re making them." That's largely borne out. The content covers pragmatic approaches to, and implications of, various design angles: articulating inspiration, interplay of structures, roleplaying-as-conversation, iteration, etc.

Overall, the zine doesn't align perfectly with my own conception of TTRPGs, but I found it all the more useful for that. For example, it makes some interesting points about granularity and utility, different to my usual approach of complexity is a cost you pay, and left me with lots to think about.

The scan quality had me squinting a few times. Corrections were made to the physical paper, which adds to the zine aesthetic, but a fair number of typos stayed in. Also, a lack of cite marks on the pasted text might have played well with the aesthetic, but meant the works cited list was less helpful than it could have been.

I'm looking forward to reading the 2025 edition to see how the author's approach has developed.

 

Sinister Hovering Orb

(Sinister Hovering Orb)

By David J Prokopetz / Penguin King Games (https://penguinking.itch.io/sinister-hovering-orb)

A single-page solo mini-RPG (plus title) which interested me because of the complete lack of strictly-flavour text: all the themes, implied goals, and flavour are purely wrapped in game mechanics. We have to learn the game to find out what it is about (to the extent that it can be said to be about something in particular).

For example, the reader discovers in the rules that they have four choices of activity/situation, used variously "when you are simply and suddenly present", "when your very presence brings down calamity", "when those who ponder you know what they must do", and "when you radiate a palpable sense of doom".

There are no player goals, high-level structures, rules about direction, best practises or suggestions to serve as connective tissue: only a series of game elements, presented with a sense of finality. I played a quick game and found it quite a compelling story generator, though a very open-ended one.

PUBLIC GUEST 5

Living With The Certainty of Death by the imminent explosion of our orbiting artificial planet PUBLIC GUEST 5

By Turtlebun (https://turtlebun.itch.io/public-guest-five) 

A solo RPG on a poster, with a delightful visual style. It's about, well, exactly what the subtitle says. It's a sort of journalling game, meant to be played a sentence at a time over the course of a long real-world period (probably about a month on average).

The diary mechanics are very simple, with two dice informing how you spend a week on your doomed world. When you roll boxcars, PUBLIC GUEST 5 explodes. (I broke the rules and played through in a single sitting, for a fairly melancholy experience.)

PUBLIC GUEST 5 and Sinister Floating Orb were interesting diversions. Together they're inspiring part of my approach to some upcoming projects which will explore themes of cosmic doom, inscrutability, inevitability, and loneliness.

Writing With Style

An editor's advice for RPG writers

By Ray Vallese / Rogue Genius Games (https://www.drivethrurpg.com/en/product/217525/writing-with-style-an-editor-s-advice-for-rpg-writers)

A 44-page book of advice for TTRPG writing. It had some tips and reminders I found useful. For someone without a background in technical writing, I bet this book is absolute gold.

There's plenty of generic advice, but plenty is specific to RPGs: sins to avoid (like read-aloud text that dictates the players' choices), lists of RPG-specific overused and confused words, writing to a game publisher's specification, etc. My only criticism is with the visual design: some odd typeface choices and the line spacing is too tight.

Worth the $5 on DriveThruRPG, especially for a new writer.

Dwarf Mine

A mapping game of adventure 

By James Hron / Paper Dice Games (https://paperdicegames.itch.io/dwarf-mine)

Perhaps not strictly an RPG, but a 25-page tabletop exploration game. You play as the location, and instead of levelling up, you add levels further down. It sort of feels like how Dwarf Fortress felt ~20 years ago: one plane, inscrutable dwarf characters, simple mechanics, inevitable doom, etc. But the cutaway is cross-section, not top-down.

There's a points (prestige) system for scoring, and a huge block of random tables. My trial game was quite enjoyable, and I definitely want to come back and give it a more thoughtful treatment.
 

The presentation has some rough edges, with typos, misplaced page references, inconsistent capitalisation, and the font size varying between tables.

A few of the rules could do with a rewrite:

• I'm a bit confused about multiple attacks vs entering combat multiple times, with persistent enemies who attack back.
• The connective geometry is mostly explained by example, not rules. Some of the rules contradict the examples, as in: "The walls of rooms, hallways, and ladders must all have at least a 1x1 square separating them", and also in "Rooms must be built in the orientation shown".
• Goblin invasion mechanics seem underdetermined.
• The appendix leaves the crucial word 'once' off the end of the sentence for Trap room. 

There's lots of neat mechanics, especially with the big monsters. I like that the Cave Wurm can damage the map itself.

The First of the Stone Folk puzzles me. It must be entombed in a 6x6 room – internal dimensions 4x4, so you only need to make 16 squares. Using 21 dwarves per square to minimise deaths, you will lose on average 168 dwarves. But I don't get the impression that total population is ever going to be over 100. Am I off the mark there? Or are you meant to wall this monster off really slowly? 

The dwarf leader ability adds a little more strategy. More inheritance tracking with the 'Bloodline' system would be interesting. I do like the idea of the Achievements list giving all future leaders a title. The room name generator appendix is clever, although wacky; you could get a barracks named "Bottom Watch" or a tomb named "New Trophy".

The author has written lots of official expansions and supporting content to keep the game fresh.

 

GURPS "Discworld Also"

A companion book for the original GURPS Discworld RPG

By Steve Jackson Games, Terry Pratchett, Phil Masters. Generally good, but most of the best stuff made it into the new edition (which I already read, and wrote some reviews for last year). So because I'll focus on the stuff that didn't, this review may sound more negative than I feel!

Sean Murray's illustrations feel quite Discworld-y, sketchy pen/pencil pictures that don't try to rip off Kidby's or Kirby's style. I like them. I'm not a fan of the book's layout, though. There's weird stuff going on with the kerning and text wrap. The column breaks are terribly confusing when there's a subheading (e.g., p 48 reads left to right then top to bottom, but p 49 reads top to bottom then left to right).

It's poorly copyedited for a professional product. For example, on just one page (29) I noticed

• A mangled seven-clause sentence
• "almost" should say "always"
• "could" should say "would"
• "fall off" should say "falling off"
• "unaging" / "ageing" spelling inconsistency
• "They not" should say "They are not"

The alternative troll stats, and the extra information on elfkin and gnolls is good to have. There's a nice robust peasant character template, with various background skills. We get a 'sentient animal' template, a disappointing miss on terminology (it should of course be 'sapient animal').

Translation of Discworld magic is done well. The additions are a bit uninspired, e.g. a magic item which replicates an air conditioner, with maybe a subtle joke about Maxwell's Demon.

The book provides some campaign settings. I mostly already read this content in the new edition GURPS Discworld RPG, with a couple of exceptions: Smarlhanger and Fourecks get more space here.

• Smarlhanger is a frontier boom town in sheep country. It doesn't commit to some details ("there may or may not be changes made", p 86), which is frustrating in a setting guide. Lots of other details to work with though.
• Fourecks cart races is a new subsystem, a spin-off of the GURPS Vehicles rules. The Mad Max pastiche from The Last Continent is fleshed out to a whole societal thing here. (It doesn't say so, but Fourecks definitely has a bunch of ultra-hardwood trees that would be used for the highest tier of armour.)

The book also has three adventures and three 'adventure seeds'. Again, I mostly already saw these in the newer edition. These seem pretty good, barring some vague patches.

'Walking the Spiral' is an adventure with an excellent idea: druids come to look for megaliths in Ankh-Morpork. (I really liked that the mouldering records were in the form of an ancient school essay.) The problem is that the implementation is painfully linear, with very little for the PCs to actually do until the setpiece fight. Especially if the druid emissaries are NPCs. The adventure also implies that the University's NPC wizards are perfectly capable of saving the day by themselves at the end, making this an adventure that doesn't need player involvement at all. The adventure premise is good enough that it's worth rescuing by redoing the structure, though.

Highland Vice

System-neutral Martian cowboy hexcrawl

By Strange Ian (https://ianstrange.itch.io/highland-vice) 

A 'kitchen sink' futuristic setting that meshes well. The concept: Mars has been scarred by poorly-thought out terraforming projects. It's reminiscent of Red Dwarf and Fallout and Futurama and Borderlands and Charles Stross, with extra Tibetan, Mongolian, Peruvian, Japanese, and Weird West elements. The goal is to find and retrieve a chunk of computer megabrain.

A 100-entry hexcrawl in 28 pages. With four hexes per page and system-neutral style, Highland Vice favours high-level terse description over gameplay details.

There's a great density of evocative ideas, which is what I want most out of content like this. An algae plantation heiress hunting with an overtuned laser rifle. The electromagnetic rail launcher used by wire-crowned shamans to send captives into orbit. A buried chest full of mismatched halves of different treasure maps. A hillbilly cyanobacteria moonshine shack guarded by a truck-sized amoeba. A monk riding a solar-powered robot turtle terrarium. An android executive assistant living for years impaled on a mammoth's tusk. A colossal mining engine meditating in the middle of a mountain, trying to free itself from the material. Gauchos duelling over a holographic pop idol in a field of cherry blossom. The primitive town built by a crashed trainload of plush robotic cuddle dolls.

Many of the hexes are linked, but some of the conceptual through-lines could be highlighted. For example, 'satellite shamans' come up in 0006, 0104, 0107, 0307, 0408, etc. If I ran this I'd make a quick setting overview using key concepts as subheadings, and add relevant pointers to/from each hex:

• Satellite shamans
• Teratorns, poebrothers, and other revived megafauna
• Robot prosthetics
• Hive drones 
• Sky bandits
• The Spinal War
• Sindovar
• The Maze 
• Josung (game)
• etc 

Definitely worth a look. It's recently published and free on Itch.
 

In progress

By my quick count, I have six different rulesets and splatbooks bookmarked and partially read. Possibly there's more, lurking in browser tabs or forgotten on the desktop.

I'd like to put some real effort in there. My ambition is that when I next fling reviews together, my "unfinished list" will be completely different!

Reviewn' June 2025 comes to a close.

So how'd I do? I didn't read as much as I hoped, but overall, not bad. For me, reviewing is ultimately about self-accountability, but hopefully you got something out of it.