Mary's Comp97 Reviews [LONG]

Skip to first unread message

Mary K. Kuhner

Jan 2, 1998, 3:00:00 AM1/2/98

This is a long posting with a review of everything I rated: I
was a playtester on a couple of games so didn't rate them, and I
never got "Leaves" to run, nor the arcade part of "Sylenius".

Some general comments first, and then a rundown of games.

I didn't start out with a coherent judging policy, but in looking at
my reactions to games I see four key factors as determining the majority
of the ratings.

(1) Programming. Was it buggy? How richly were synonyms, object
descriptions, multiple solutions to puzzles, and so forth implemented?
What proportion of responses made sense? Could you do reasonable
actions with objects even when they weren't puzzle solutions? Did
it crash? Did text come out when it was supposed to, in the right
order, and not otherwise?

(2) Game play. Were the puzzles fair? Appropriately difficult? Did
the game have long dull stretches such as repeatedly traversing an
area? Did it annoy me with tiny inventories, starvation, or massive
requirements for brute force? Was it appropriately sized for a two hour
game? (I was willing to give some slack to long games, unless I felt
that they'd wasted time unnecessarily.)

(3) Writing. Was the text evocative? Correctly spelled and
punctuated? Did the NPCs sound like people? Did the little responses
to random actions support the tone of the game? Was the technobabble
or magicbabble convincing? If the game tried for humor, was it funny?
If horror, was it scary?

(4) Story. Was the setting a real and interesting place? Did I care
about my character? NPCs? Was I willing to buy into the goals and
methods of the game? Was it all of a piece, or a scatter of unrelated
parts? Did it wrap up well? Did it avoid cliches, or find ways to make
them fresh?

I didn't give any ratings of 10: I was holding out for games with
flawless programming, wonderful puzzles, first-rate writing and a
gripping story, and while I got each of those individually I didn't
quite get all of them in one game. However, the games rated 9 were
extraordinarily good, each of them among the best IF I've encountered.

My single most important piece of advice for authors: playtesting.
All of my high-rated games thanked multiple testers. Many of the
low-ranked ones hadn't been tested at all. The difference is not
subtle. (I think it was the author of "Babel" who noted that responding
to playtester comments had increased the size of his game 25%.)
Also, as a playtester myself I can add: don't wait till the last
minute. Good playtesting takes time. And do go back and re-test
after making major changes.

A couple of pet peeves:

(1) Settings with lots of dull, meaningless locations, like the
endless halls in "Aardvarkbarf". In my opinion a game may need
two or three do-nothing, bare locations for plausibility, but it
probably doesn't need more than that. "You wander the halls for
some time and come to ..." would be preferable.

(2) Starving to death. Especially, starving to death in a day or less.

(3) "Towns" with half a dozen or less buildings in them. I don't
need to see the whole place, but at least there should be some kind of
facade suggesting more space.


Rating 9 games

"Sunset Over Savannah" (TADS)

An ambitious game about beachcombing, magic, and life decisions.
It takes place in a small beachfront area of Savannah, Georgia,
which is wonderfully realized: every location is detailed, none
are perfunctory. I was amazed by the richness of actions which
the game anticipated: I was almost never balked by "You can't do
that" or a nonsense response. I climbed over railings, built
sand castles, tasted assorted foods, chatted with the old man, and
nothing broke the game illusion.

The puzzles are elegant and mostly fair. Some are quite hard--
Andrew Plotkin was listed as a playtester, and one wonders if he
had a hand in a few of them--but extremely good use is made of
contextual clues. When you fail, you often learn why you failed
and get a hint as to what to do about it. One puzzle seemed
unfair, and one contained a bug which led to reasonable actions
making the game unwinnable. Good hints are provided, though in a
mildly awkward interface: an adaptive hint system would be a
welcome grace note.

The decision to avoid lethal states was a sound one. Avoiding
unwinnable states as well would have been helpful, though
there are not very many of them. I would have preferred a
larger or unlimited inventory both to smooth gameplay
and to avoid realism issues like "can I really leave my wallet
on the boardwalk without having it stolen?"

Though the game is set at sunset, it is not timed. This has the
advantage of relieving pressure on the player, but the disadvantage of
feeling slightly unreal. "Savannah" is quite long: I couldn't finish
it in two hours even with extensive use of hints. However, it's
appropriately sized for its theme and contents, and much of that time
was spent playing with the rich environment. I don't see any way to
make it shorter, nor any real call to do so. There is no wasted space.

The writing is vivid and compelling overall. At a few key junctures it
tries too hard to impress, which is probably the game's main esthetic
failing. I think that the moments of epiphany needed to be a little
less explicit: at times the game tells the player what to feel when
it would be better to simply describe the scene and let the player
fill in the emotional response.

There are many references to other IF and static fiction, but they're
subtle and non-intrusive; the game wouldn't be damaged if you missed
them all.

Overall, I found this a lovely game: ambitious in its storyline
and mostly successful with that ambition. It could use a little
more polishing, but plays very well as it stands. Only the note of
preachiness in the writing, and a couple of bugs, kept it from a
perfect score.

"A Bear's Night Out" (Inform)

A teddy bear pads around the house while his master sleeps, encountering
various challenges including many IF game references.

The programming was first rate. Except for one probable Inform glitch
I found no bugs, and responses (in appropriately teddy-bear parlance)
were provided for practically every action. (I have seen reports of
a bug causing the game to crash, but never encountered it.) This is
one of the very few competition Inform games which implemented player-
on-a-chair and got it completely right (scope, commands to get up and
down, etc.).

I liked the puzzles very much: it was generally clear what to do,
and reasonable obstacles were provided. Holly the cat was particularly
well implemented. I was tickled when I tried several actions that
work with my real-life cats and got good results (one was even a
puzzle solution). In many IF games sound does not propagate among
rooms, but Holly can hear what you're doing down the hall and will
come to investigate if it sounds interesting. Multiple solutions are
supported for some of the puzzles, a nice feature.

The hints were helpful, though I could probably have done without most
of them had I had more time and patience. One puzzle is a little
unmotivated, but I can put that down to bear boredom. Unusually, the
game seems to have been beta-tested end to end: there is no
accumulation of typos and small bugs in the final sequences, something
found in the bulk of competition games (those that were beta-tested
at all).

The writing is simple, but sweet. Although it bills itself as a
children's game the humor and puzzles are not particularly childlike
(nor are the puzzles easy enough): however, it would probably be
fun to play *with* a child. There are a lot of IF in-jokes, which
slightly pull the tone of the game away from its teddy-bear premise,
but I enjoyed them in themselves. (Kudos to the author for getting
formal permission to include other authors' text.) A little more
time to explore the games-within-a-game would be fun, though I
appreciate that it's a huge programming effort. I laughed a good
deal while playing the game, and smiled even more. The AMUSING
list at the end is well worth perusal, and the game is simple enough
that it's painless to go back and try some of the entries. FULL SCORE
is also worth checking out.

It's a bit unfortunate that two very inert objects are right at the
start of the game: it creates an impression of futility that is
at odds with the game as a whole. It is also surprising that this
game managed to make an ordinary house (bathrooms and all) interesting,
given how severely oversaturated the competition was with such settings.

"Bear" is by its nature a less ambitious game than "Savannah" or "Babel"
but is the cleanest and most successful, within its given scope, of
the three.

"Babel" (TADS)

From the first scene to the last, this game is icy, atmospheric, and
tense. It details the exploration of an abandoned research station,
and the slow unravelling (mainly through a kind of object-reading)
of the station's last days. I overran my two hours because I was far
too involved to quit.

The programming is clean, with only a few glitches. Careful
attention was paid to issues like light and darkness. I found the
behavior of the game with regard to automatically opening doors
(sometimes it would open an unlocked door without being explicitly
asked, other times it wouldn't) a little erratic, and in a few spots
the room description did not adequately get across all of its contents.
There are some object-crowded rooms with very few usable objects
(this is particularly noticable in the labs) but the "You can't interact
with that" messages were well phrased and not too obtrusive.

"Babel" is not quite puzzleless IF, but the puzzles are mainly very
easy ones, apparently designed more to increase player engagement with
the storyline and to control pacing than to actually challenge the player.
I felt this approach worked extremely well. A challenge would present
itself, I'd solve it, and new areas of the game/story would open up; I
seldom got stuck or distracted, which would have reduced the sense of
tension. One puzzle depends on an inobvious property of an object,
and could be better cued in the object's description. The bulk of the
game is simply exploring the station and spying on the shades of the
past, but it doesn't feel like being fed a lump of narrative, partly
because of several very elegant devices to draw the player in to the

The game is perhaps a little larger, physically, than it needs to be:
some empty halls and stairways could be omitted to speed up travel,
and there are an awful lot of bathrooms. (I might not have felt
annoyed by the bathrooms, except that the competition as a whole was
overrun with them.) On the other hand, the number of flashbacks
was entirely appropriate, even though not all of them carried essential
plot information; and one of the best scenes occurs in a bathroom.

The game appears to be timed, though I didn't manage to overstep the
time limit (it might be an idle threat). I'm of two minds about this:
the warnings added tension, but if they were actually carried out
(forcing the player to start over) the impact of the game would be
greatly diluted, and I spent some time worrying if I should be
interacting less in order to move faster.

The prose is highly evocative, only occasionally stepping over the line
into purpleness. Adjectives were occasionally used in odd ways--dull
grey light dripping into a room?--but this is not obtrusive. Good
use is made of senses other than vision in a couple of spots. The
author takes some chances in describing the character's thoughts and
feelings, but for me, at least, succeeded very well: places that were
described as eerie *were* eerie.

The key revelations of the plot are very well handled, and several are
edge-of-the-seat disturbing (I particularly liked the scene in
Jonas' bathroom). The ending packs a punch commeasurate with the rest
of the game, and does not go in for Hollywood saccharine.

The main esthetic flaw, for me, was reliance on cliche and
technobabble in crafting the central plot. The characters were well
drawn, but they were stereotypes (the God-Fearing Scientist, the Amoral
Scientist, the Troubled Woman) and their dialog sometimes fell into
worn-out Science Versus Humanity ruts. I'm a research scientist myself
and a bit picky about technobabble: this batch did not convince. (The
strength-enhancing formula would make much more sense if it evoked
hysterical strength rather than causing instant muscle growth, for

However, this didn't distract much from the emotional impact of the
game, perhaps because the game really isn't about Babel and the
conflict between Science and Humanity: it's about being cold and
alone and afraid among the shades of the dead station, and it succeeds
brilliantly at that. I was especially impressed with tiny bits of
descriptive detail work. We learn that a deadly toxin was spilled,
and later find a bit of hall which smells strongly of cleaner. Dead
animals abandoned in their cages seem at first to be sleeping.
Even trivialities like the lights on the ceiling became eerie as
part of the overall effect. There is little bland prose, even in
bare halls. And the cold outside, when it is finally encountered
full-on, is so intensely depicted that I found myself typing faster
to escape it.

Several of the competition games tried to be scary, but this was far
and away the most successful. I'm particularly impressed that it
didn't fall into the trap of constant character death and tightly
timed puzzles: the urgency was generated by the tone and content
of the text, not by forcing manuvers.


Rating 8 games

"A New Day" (Inform)

This is an IF game about an IF game: you are picking your way through
the half-finished creation of a dead author, looking for clues, and
interacting with Winston, who may (or may not) be the narrator of
the unfinished game. As the player experiences it, the game-within-
a-game is fragmentary and broken, with blank room descriptions and
half-coded puzzles. This risky decision (I found it hard to put
aside my initial response to obviously broken code, even once I
realized it was deliberate) is good for some very creepy moments.

The game is in four parts: an initial meeting with Winston, the
machine's indweller; a fragment of an incomplete IF game; a
second fragment suffering from massive bitrot; and a finale
in which you must deal with Winston and his goals. The first three
parts play mind games with the conventions of IF (I cringed every
time I "died" in part 3, even after I realized what was happening)
while the fourth is a little more conventional, and unfortunately
the weakest. Still, it provides a solid sense of closure.

The game's biggest flaw, and the thing that kept it from a rating of
9 even though it seized and held my attention, is that the puzzles
are not very satisfying. I groaned when I saw the green wire, the
yellow wire, the red wire, and the brown wire--in a timed situation,
no less, which I ended up playing through several dozen times before
I could crack it. This breaks the game's premise, which involves
threatening the death of the *player* (not her character) if the
last puzzle isn't solved in time. It was also very hard to visualize
this scene accurately (where are things in relation to each other?) I
missed taking one object because it was "clearly" out of reach.

The game is a good length for competition play, and has useful clues.
It would be a grace note, though perhaps difficult, to incorporate
the clues for sections 2 and 3 into the game-within-a-game conceit.

Overall I enjoyed this game greatly, but it needs better puzzles
(particularly in part 4) to be a champion.


Rating 7 games

"She's Got a Thing for a Spring" (Inform)

Exploration of a woodland and interaction with its animal and human
inhabitants. This is a quiet and mildly romantic game, with few ways
to make a serious mistake. It's enjoyable for its mood and natural
details, and for several good puzzles (the last few are weaker,
unfortunately). It also has perhaps the most fully coded NPC in the
game, with a wide conversational range and complex autonomous behaviors.
What you'll find Bob doing depends on what time of day you meet him,
and also on what kinds of help you're willing to provide.

Unfortunately the game is quite buggy in spots, especially as soon as
you stray away from the main solution path. The blackberry is a classic
example of why Inform floating objects should not be portable. Bob
sometimes responds "Default Order Property" when asked to do something,
and his attempts to share his lunch fail in a frustrating way. And
one mishap involving the sandals should be preventable by taking them
off, but isn't.

The writing is quite good, especially in describing Bob's actions and
the antics of the wild creatures. Towards the end it seems to change
tone a little, and not, I thought, for the better: I found myself
mildly put off by the romantic finale. Perhaps it's that the
protagonist is too clearly a real-life person on whose privacy the
player is intruding.

Most of the puzzles are fair and reasonable, but some would profit from
richer coding of alternative wordings (especially the shed puzzle) as
they tended to become guess-the-verb once you understood what to do.

With polishing, this could be an excellent game: it's nice to get away
from the standard building setting.

"The Edifice" (Inform)

A rather large game (I was only able to play about half of it in the
given time) involving re-creating moments of insight in the life of
prehistoric humankind or proto-humankind.

I very much liked the tone of the writing, with the capitalization of
key nouns suggesting a kind of personification: this isn't just my
branch, it's Branch. In some of the location descriptions it's a
little cold, but overall it works for the game. There is even some
attempt to suggest the growing intellectual sophistication of the
protagonist, though allowing the sections to be played out of order
weakens this effect.

The language puzzle in part 2 is both difficult and intriguing, though
I wasn't able to solve it in the time allotted: you're trying to
talk with someone with whom you have no words in common, by a
combination of pointing and picture-drawing to learn vocabulary. The
NPC involved has quite a range of conversation, even if I was never
quite sure what he was saying. The part 1 puzzle is a little less
satisfying: it is easy to get into an unwinnable situation, and many
alternative solutions hinted at by the environment (for example, driving
Beast into the mud) are not allowed for.

The game is too large for the competition format. This might be
improved by trimming the physical scope of part 1 (it is easy for
the player to get lost for a long time--it would help if the Edifice
were visible from a distance) and implementing multiple solutions.

Hints are provided via a within-game mechanism (etchings slowly
appearing within the Edifice, which fills the role of a 2001
monolith). The player is required to go out and interact a while with
the puzzle before receiving another clue. Unfortunately, the clues
are not adaptive, which means that you may have to make a great many
trips if you breezed through early subpuzzles but got stuck on a
later one. It also proved impossible to clue the language puzzle
usefully with an etching. This is a nice idea, but could use some
further work.


Rating 6 games

"Madame L'Estrange and the Troubled Spirit" (Inform)

A psychic detective tries to unravel two interlinked crimes in Sydney,
Australia. This is an unusual game in a number of respects. It mostly
dispenses with conventional IF movement and puzzles in favor of a
name-the-location movement system, interviews with NPCs, and psychic
scrying. It is also written in third person ("Madame did
such-and-such" rather than "You do such-and-such") and mostly, though
not quite universally, in past tense.

There is a *lot* of text in this game, and it stands or falls on the
strength of that text. Mostly it stands quite well: people and places
are well drawn (there is no doubt at all that the city is Sydney, not
some generic city) and there are some effective bits of horror.
I particularly liked Madame's first encounter with the "doggie".
The tone is melodramatic, but not annoyingly so. Most of the large
cast of NPCs are only sketched in, but they avoid cliches fairly well.
Madame herself has a definite personality which comes through in her
dialog and reactions.

There are a few breaks into present tense, probably oversights (it is
not trivial to convert all the library responses).

The format of the game and the third person/past tense writing combine
to create a certain detachment between player and protagonist. I'm
not sure I consider the person/tense decision to have been wise: as
_Christminster_ shows, it's possible to strongly characterize the
protagonist without using third person, and the past tense just comes
across as awkward in spots (especially when it is trying to get across
the concept "Madame can't do that"). Still, I was more engaged than
I expected to be, and overran my two hours a bit to see the ending.
A set of questions from the employer at the very end help to reward
paying attention to the plot and not just its tokens: a nice concept.

Unfortunately, the game is buggy, especially near the end. Madame
maintains a pad of locations which she can visit, but at least one
key location never appeared (I found it only by the accident of a
disambiguation question) and another is confusingly mispelled. Some
events which should happen only once repeat whenever Madame revisits
the location: for example, her client claims to leave her home, but
in fact he's there to tell his story again whenever she returns.
And the final puzzle sequence, the only conventional puzzle of the
game, is full of run-on text and daemons which don't execute in quite
the order they should.

The internal documentation, oddly enough, is also buggy (for example,
implying that time passes and the player might try a second seance the
next day--I don't think this is true). The game shows some signs of
a hasty late revision. The walkthrough was apparently written by a
playtester, not the authors: its tone clashes painfully with that of
the game.

This is a good game as it stands, but could be a much better one with
another round of beta-testing.

"Sins Against Mimesis" (Inform)

This is a spoof on IF in general, with references to famous games
(particularly _Curses_ and _Jigsaw_), newsgroup gossip, and perhaps
other things I didn't catch. Generally I don't much care for this
sort of thing, but somehow this game made me laugh. It's hard to
describe why without spoiling any of the jokes.

Coding quality is quite good: not much attention was paid to scenery
(as will be seen by examination of the furniture) but that's
deliberate, and those objects which are important are coded
correctly. I would have liked to see a little functionality in Text
World, though it would have been a bear to code. When the game drops
into an emulation of a different parser it is quite convincing, a
trickier feat than it sounds.

The AMUSING list is worth perusing, though I was disappointed in the
lewdness of LEWD. The game is short enough that you can go through the
game, and then through this list, in two hours: quite sufficient for
humor, which often flags if prolonged.

The writing does what it needs to (though Black doesn't quite sound like
him/herself). Parts of the second half actually have a descriptive
quality that would not be out of place in a serious game. The major
NPC responds to a goodly range of dialog, in his smart-ass way.

Since it is essentially one big in-joke, this game will be almost
completely lost on anyone who isn't part of the IF subculture. It
tickled my funny bone pretty successfully, though.


Rating 5 games

"Zombie!" (TADS)

Evil zombies are taking over Indiana in a low-budget horror movie.
The game's appropriately sized, and has a couple of nice puzzles.
(Also maybe a few more red herrings than it needs: I was disappointed
that the skeleton key didn't unlock everything as promised, and that
at least one visible "puzzle" has no solution.) The prose is nothing
very special, but it gets the picture across. A little more unity
of tone--is this hokey horror, or is it a satire on hokey horror?--
would help.

Unfortunately there are rather a lot of bugs, ranging from outright
TADS runtime errors to misplaced text (the umbrella is always hanging
on a peg when you drop it, even outside in the garden) to weird
effects like items teleporting out of your hands. I also wonder how I
managed to get 21 points out of a possible 20 (confounding the
score routine, which called me a " "). There are a lot of
objects which don't have as many synonyms as they ought, and some
verb problems. I spent a long time pushing, pulling, and turning on
switches before hitting on "flip". Game play, as a result, is
clunky and awkward.

Logically speaking, the measuring-cup puzzle would have a trivial
solution involving storing water in the cake pan; disallowing this
makes the whole exercise seem rather arbitrary.

The most serious game-play flaw is that there is no way to know what
you need to bring into a certain restricted area of the game, except
by going there and dying: you have many objects and can carry very
few, and the essential one is inobvious. I resorted to hints a lot,
and often felt when I got them that there was no reasonable way I
could have known what to do.

Not a bad game, but unpolished. The bugs are particularly distressing
in the key scene where the mad scientist is introduced: a pity, since
it's a nice conception. I especially liked the puzzle involving "sweat
of the damned."

"Glowgrass" (TADS)

A miniature game about exploring the ruins of a futuristic civilization.
The text and atmosphere are quite nice, and the plot is appealing. The
puzzles are rather prefunctory, mainly consisting of putting parts
togther into widgets: since there are no spare parts to speak of, this
is easy enough to do.

Game play could have been improved by more careful choice of object
names: a game this size should not need constant disambiguation
questions. Also, one item changes its name without warning, leading
to confusion when you try to refer to it (hey, it was here a minute

There are a few bugs, but not very serious ones: getting out of
the chair is problematic. Also a few logic questions: where is the
master bedroom? Overall the game works well enough, though
the environent is rather bare and restricted. With less than
ten rooms, the author might have put more detail into each one.
In particular, the posters in the girl's room could be described,
and the contents of the kitchen, and a glimpse of the automation which
must keep the lawn alive.

There is a hint of a sequel at the end which I'd like to see: the
game as presented really feels like a prologue. The setting and
situation have a lot more potential than this tiny game can fulfill.
In spots, the development of the background simply feels like a lecture,
since the player will never be able to build on that information.

"Friday Afternoon" (Inform)

The player is a MicroSun employee trying to finish a bit of work
before going on a hot Friday evening date. The setting is
a single office, with a couple of co-workers and a handful of desks
and printers: six locations in all. It's an appropriate size for a
competition game; I was able to finish it in about 1.5 hours, with a
little use of the on-line hint function.

It's a lightweight but amusing game, and would be good for beginners.
I especially liked the calendar puzzle. There are a few problems: some
guess-the-verb, especially around the telephone and in NPC dialog (i.e.
"ask NPC about X" is needed when you want the NPC to give you X, as "ask
NPC for X" responds "You don't see that here"). I would have liked to
see more complex use of the telephones, and a few more objects (much
of the scenery returned "You don't see that here" when examined). One
puzzle solution struck me as mildly unfair, but the others were good.
The tight setting is used to good effect. To flesh the game out, I
might add a little more complexity in the Marc/password sequence (i.e.
something you have to do after obtaining the password).


Rating 4 games

"The Lost Spellmaker" (Inform)

This is a small quest-game set in a fantasy village. It's much
more seriously designed than "Phred" or "Town Dragon", but still
somewhat tongue-in-cheek.

Some quite nice detail work has been done on the town, such as the
characterization of Digga and his/her gender changes (though the
pronouns didn't seem to follow properly). The magic concept is a
novel one, and its implications have been worked out in a bit
of detail. The game avoids stereotypes; in fact, sometimes it feels
as though it's straining to avoid them.

Unfortunately, I didn't find it satisfying as a game, for two
reasons. The first is that the protagonist is mostly a spectator.
Too many scenes revolve around someone else solving the problem or
making a speech. This is an interesting defiance of stereotype in
itself, but it reduced my sense of engagement and accomplishment.
I ended up feeling as though my employer ought to have gotten
someone else to do the job, since I clearly wasn't up to it.

The second problem is the puzzles. I had to use the walkthrough
extensively because I couldn't make any progress. One key item
was hidden in a busy room where all the other details returned "You
don't see that here"--I had given up trying too soon, assuming there
was nothing there. Key actions were unmotivated: how was I to
know where to lead the cow, for example, or what to do with the
villain at the end? (The "right" answer to that last question
strikes me as wildly out of character.)

There are some programming bugs, especially associated with the
basket and rope. I crashed the game once by tying the rope

Hints are given via an in-game mechanism: a nice idea but not
really appropriate for a detective story. If your employer
really knew the situation well enough to give hints, he wouldn't
need to employ you to solve it--thus, the in-game hints create
the suggestion that your employer is just making you jump
through hoops.

The game ends with an interview with its author, which is a neat
idea but not thoroughly implemented (it turns into a long string
of "I don't know anything about that"). Probably two or three
times as many responses would be needed in order to give any
feeling of a real interview. As it is, this scene, like the
game as a whole, is intriguing to think about but frustrating
to experience.


Rating 3 games

"Down" (Hugo)

This game involves rescuing people from an airplane crash. The
situation is novel and interesting, and the beginning of the game
(which throws you right in without explanation) packs quite an
emotional punch. Unfortunately, this punch is rapidly dissipated
by sketchy and flawed coding which breaks the sense of reality.

There is little indication of which directions the player can go in,
and no explanation of why, for example, you can't go back to the
first part of the game after leaving it. (You might save your
life by doing so, after all.) Scenery is not adequately defined
and many objects don't really exist. Few verb synonyms are available
and at least one puzzle demands a rather unusual verb.

More of a problem for me was the obscure nature of some of the find-
object puzzles. If I need to search everything in sight, I want some
positive feedback on my early searches, or else some hints as to where
to look. In this game, a crucial object is found where I would have
thought, from the room description, there couldn't possibly be room for
it. The puzzles are not overly complex, but they are often hard to
solve for reasons like this. This is particularly annoying since
the situation presented is urgent (and probably timed): random
searching breaks the mood.

I was deeply exasperated by having a strip of cloth and a bleeding
man, but being unable to do anything with the combination. It turns
out that the game didn't expect me to do anything, but why? Also, I
tried to tear a second strip of cloth off my pants (I had other uses
for that first one) and it chided me "You won't have any pants left if
you do that." I'm trying to save a man's life, for heavens' sake.
I don't care if I have pants left.

Disambiguation questions sometimes got a bit weird: "Which little
small do you mean, the little boy or the little girl?"

A good idea, but probably needs three or four times the current amount
of code in order to really flesh out the environment.

"Phred Phontious and the Quest for Pizza" (Inform)

A silly-fantasy game in which the player is a cook's apprentice looking
for specific ingredients.

The tone of the writing is uneven: sometimes it seems to be
aiming for either creepiness or black humor (as in the graveyard
and dentist scenes), other times for parody, other times for plain
scenic value. The game would probably have a stronger effect with
a more consistent tone, either black humor or broad parody.

Some of the puzzles are well worked out: I particularly liked the
malevolent dentist and the use to which you can put her object.
Others are annoyingly buggy (for example, the name used in the
text for a key item is not the name the player can use) or trite,
such as repeated get-x-give-x.

At least one puzzle is tightly timed, but lacks basic support for
that, such as a good exits list. (I hate dying because I can't tell
in what direction the exit lies!) It is also very choosy
as to nouns and verbs. I probably died two dozen times trying to
solve this puzzle, which did not endear the game to me. A related
problem is that it is easy to get into unwinnable states and
have to go back to the beginning. I was not able to finish the
game in two hours due to this. UNDO is some help but not if the
lethal move was not the most recent one.

There's probably the seed of a decent game in here, but it needs pruning
and polishing. Consistency of tone, more polish in the coding, and
more careful selection of puzzles would all help.

"The Frenetic Five Vs. Sturm and Drang" (TADS)

This game follows a team of hapless young superheroes on a mission.
The idea is nice and not often seen in IF. However, it's rather
ambitious in that the NPC members of the team are on stage a lot,
and need to be exceptionally good. In this the game doesn't really
succeed. The NPCs come across as annoyingly dense rather than
amusingly so, and there is just too much scutwork involved in getting
them to accomplish anything.

Several aspects of gameplay seem unnecessarily hard. For example,
it's necessary to repeatedly ask each member to put their bus token
in a farebox, because there's no apparent way to tell which bus to
take initially. One of the NPCs should presumably know, and indeed
I thought I was getting hints to ask him, but I didn't succeed in
getting an answer. Similarly, the player's inventory is much
smaller than the number of objects he's likely to want later on.
He can give them to NPCs, but this soaks up time (get an object,
give it to an NPC, ask for it back, try to remember who has it....)
and is not all that exciting. I feared that if I ever reached
the finale I would find I had the wrong subset of items and no time
to ask for them back.

The game is quite buggy, with problems in disambiguation (those bus
tokens again) and daemons running inappropriately. It also gives
responses appropriate to player-does-X when the player has asked an
NPC to do X. I managed to crash tadsr once with a "tie" command (a
well known IF weak spot, admittedly.)

I didn't finish the game, having bogged down in the bus token
fiasco. It may be that the finale is better, but I didn't have
the enthusiasm needed to go back to the start a fourth time.

"Zero Sum Game" (TADS)

You're the murderous hero/ine of an adventure game...except now you
have to give back everything you've stolen and placate everyone
you've injured, because your mama's really mad at you.

I like this premise. Unfortunately, I approached it thinking "How
can I redeem myself?" but the storyline does not support that: to win,
you must behave even worse than before, including killing everyone
who's been nice to you in the course of the game. I was never going
to solve it, because I simply wouldn't have tried that (I know about
it only from the walkthrough).

The game is not very richly coded: location descriptions are good, but
most of them are meaningless and empty, and objects are poorly detailed.
(The weed clump doesn't know the word "weed", which kept annoying me.)
Dead people and animals are "a dead such-and-such", which I guess is
meant as a reference to Roguelike games, but just comes across as lack
of effort. (And you can carry an amazing lot of them--shouldn't
people react to Mr/Ms Corpse-Monger?)

The puzzles are not bad in themselves, but they feel as though they
belong in a game with a rich environent where trying odd nouns and
verbs is rewarded, not constantly met with error messages.
If you behave all the time the way the puzzle solutions need you
to behave, you'll be drowned in unhelpful responses like "I don't
know that word."

You'll also die a lot: this is one of those games you probably have
no chance to win without dying (how to know that something is a poison?)
The game provides a command to give you warning when you are about to
get into an unwinnable state, though I'm not sure it works (I feel I
got into unwinnable states several times without warning, but perhaps
I just missed alternative solutions). However, there is no warning
for sudden death, though UNDO is available.

I thought the scene with the naked S&M wizard, which was quite
cleverly described, would have been way cool if you could *do* something:
but it seemed to be just a set-piece to look at. Like the rest of
the game, it promises more than it can deliver.

"VirtuaTech" (TADS)

A computer-oriented variation on the "student life" theme. The setting
is somewhat interesting, but unfortunately we see very little of it:
the semi-dystopian future is really just a backdrop for some mildly
exasperating puzzles. Future tech seems to involve having to jump
through hoops (for example, you have to attach an address thingie
to your phone in order to use it, and it can only hold one address).
The game is a bit buggy, lacking synonyms and occasionally saying
unhelpful things like "Which do you mean, the link or the link?"
(Particularly odd since there was only one link present.) A supposedly
helpful NPC lacked the vocabulary to do much.

My main esthetic objection is that if the player really knew what
the character knows, there would be hardly any puzzles in this game:
almost all of them involve difficulties in manipulating a widget
the character presumably knows all about. This does player
identification with the character no good at all.

"Travels in the Land of Erden" (Inform)

I was wondering how a competition game could require .z8 format,
normally reserved for monsterously large games. The answer is
rather simple: this is a monsterously large game, way too large
to be solved in competition time. I really never progressed
beyond wandering about looking at things, having visited probably
three or four dozen locations.

There is some good nature description involved, though the author
occasionally lets her fondness for a graceful description get in the
way of mundane information like an exits list. The land of Erden
is a fairly standard fairy-tale country with a castle, a witch,
a princess, mysterious ruins, and so forth. It is depicted
earnestly, without the tongue-in-cheek tone of the other fantasy
games. The writing is reasonably graceful though not catchy. I
particularly liked the seacoast and rock bridge. The only major
writing flaw I noted was a lack of facility at getting across spatial
relationships such as where the castle sits in relation to its

Among the myriad locations there are a relatively small number (as
far as I could determine) of usable objects and talkable people. (There
are an awful lot of non-talkable people, including a whole market
where "nothing is for sale" and no one will speak to you.) None of
the objects had obvious uses to me (I tried the ladder in a few
spots, but it never helped me). The game is so big, and puzzle/
object relationships so unclear, that the task of solving anything
seemed daunting. So I have a sled. What on earth do I do with it?

This was really my major objection to the game. The landscape was
interesting to traverse once, but running across five to ten locations
every time I wanted to try a new puzzle solution got tedious very
quickly, and the sheer scope of the problem was daunting. (Had I
missed searching some object somewhere?) I had a goal--find the
jewel--but with no idea how to prosecute it I floundered around,
trying things at random.

The bit of the walkthrough I read suggested some rather nice puzzles
with multiple solutions, but I could not come to grips with them and
didn't solve any (except entering the castle) in two hours of play.
I wasn't stuck--I still had object/obstacle combinations I hadn't
tried--but I was baffled and bored.

I feel that what this game needs is either a clearer problem statement
with some sub-goals, or a restriction on the travel space so that
puzzles and solutions are closer together. I've seen many IF games
which are too constrained and linear, making them feel railroaded:
this is one of the few I've seen that are too big and open, at least
to my tastes. It's clearly oversized for the competition, but I
don't think I would be pleased with it even as a full-sized game.

"Poor Zefron's Almanac" (TADS)

A tongue-in-cheek fantasy game about a wizard's apprentice. It seems
to have been inspired by _Enchanter_ though the spell system is easier
to use. A twist midway shows that the situation isn't what you think it
is, and the game is really a genre crossover (with the new genre as
silly as the old one).

There is almost nothing present that isn't a puzzle solution: once you
notice this, the puzzles are not too hard. Most of them involve turning
into a critter, since that seems to be the only spell you know. It is
easy to die or get into an unwinnable state this way. Save often.

The documentation may have raised false hopes: it talks about how
to use the almanac (grimoire) in some detail, as if you'd actually
have call to do so, and implies that there might be more than one
spell in it. This text would grace a larger and more complex game,
but is a bit annoying here.

The biggest bug probably involved disambiguation of various bits of
partially toasted bread: I was constantly taking or dropping the
wrong one, as the game would leap to conclusions. It's a good thing
there were plenty of pieces.

There's nothing really wrong with this game, though it is a little
buggy, but it just left me cold. The humor didn't click for me: it
is supposed to be reminiscent of _Hitchhiker's_ and _Enchanter_ but
came across as derivative instead. The setting is a cliche, and the
plot twist simply piles on another cliche. When I made an inobvious
error that lost me the game, I wasn't motivated to go back and try


Rating 2 games


This does not really feel like a game: more like being teased
by the author. Hey, you're dressed as a hotdog. Hey, you
blow up. Hey, you win--all equally arbitrary. You're supposed to
be trying to send a piece of email, but none of the obstacles are
really puzzles--there's no thinking-out involved. You just need to
try stuff to see what will happen.

The teasing was a little funny, but mostly annoying. It's a good
thing the game is so short.

There are a few bugs even in this short bit of game, mostly
involving trying to pick up the pile of mail. I didn't see any
sign of the notorious AGT parser problems, perhaps because there was
little opportunity to encounter them.

Basically a single joke, quickly tossed off. You might imagine
mailing this to a friend just for five minutes' amusement (that's
about how long it is). It would probably be funnier in such a
context, too.

"The Obscene Quest of Dr. Aardvarkbarf" (TADS)

You're a student running around a mildly fantastic college
campus trying to deliver a letter to Dr. Aardvarkbarf, whose name
is a good indication of the kind of humor involved. It's not very
obscene--the most risque detail is taking off your clothes, and since
there is no one around to notice, that's not too exciting. It's not
very funny or very interesting, except for some of the building
descriptions. The Administration Building has some catchy, if rather
overstated, text which could have helped support a more
interesting game, something like _Bureaucracy_. The outsides of
buildings in general suggest that a real campus was the model, whereas
the insides are very stock.

The parser has trouble with anything more than two-word commands,
and has been given very scant vocabulary to work with. I was
baffled at getting the lockpicks out of the box, since the game
does not know "lock pick" nor "lockpick", until I hit on TAKE ALL
FROM BOX. The game is also pretty buggy. At one point you retrieve
an object by searching a desk: you can do this repeatedly, causing
the object to teleport back from wherever you left it.

You need to collect sandwiches so you won't starve to death, or at
least I got that impression; unfortunately your inventory is small
and the number of objects fairly large, with annoying consequences.
A backpack would help here, but really I'd kill the food puzzle: it
is tired out and annoying.

Mainly, though, I just ran out of enthusiasm after going through many
dull halls and looking in many dull desk drawers. It turns out I
was only a couple of moves from the end when I quit, but I was not
motivated to go back.

"CASK" (Inform)

A tiny game, two puzzles in four locations. The author notes that it
is a translation from another language into Inform, and it's not a
very successful one. There is one bit of nice description (when
you are inside the cask being crushed) but the game is very buggy,
arbitrarily lethal, and (as the author notes in his own walkthrough)
solvable only because an Inform error gives you a hint as to where a
key object might be. It could be a nice coding exercise--it's awfully
short to be called a game--with some revision.

"Symetry" (Inform)

A tiny horror game about a cursed mirror and the creature within it.

What a frustrating game! It's about as buggy as they come, with a
single puzzle whose solution strikes me as completely unguessable (who'd
have thought the game had that word in its vocabulary?) And yet
the situation is compelling and the prose, luridly overwritten though
it is, really caught my attention. The "lose" ending is better than
the "win" one, which resorts to cliche. Somewhere in this mess is a
miniature gem, but it will take a lot of polishing to reveal. The
current incarnation gives the impression of having been written in spite
of the Inform library, not in collaboration with it--many ordinary
actions are apparently special-case code, such as getting into bed,
turning off lights, and so forth. This would be okay if the special-
case code actually worked, but is amazingly frustrating as it stands.


Rating 1 games

"The Tempest" (Inform)

Shakespeare's famous play, translated very literally into IF (it
is almost like reading the text of the play with occasional
breaks for the player to insert an action). While this is a novel
idea, it seems on the face of it to be doomed to excruciating linearity.
This implementation is certainly that, compounded by an odd reluctance
to use synonyms. There is only one puzzle to tackle at a time, and
you must phrase it just so: "blow storm" and not "cause storm" or
"make storm" or "blow wind" or "fill sails" or, as far as I could tell,
anything else.

I got stuck immediately, wrote to the author for help, was able with
his/her hint to get about two steps further, and gave up after a total
of 45 minutes of play. I just got hopelessly tired of "That
instruction, that verb, doth elude me" as a response to almost every
action I attempted. Several weeks later, armed with a walkthrough,
I went back to it and was still frustrated. I could never visualize
my surroundings well enough to act intelligently; I got tired of
paging through dialog, even Shakespeare's; and I still typed a dozen
wrong commands for every one that worked.

The Shakespearean library responses were cute, but the game didn't
work for me as a game at all. There's just not enough scope for
actual play.

"Congratulations!" (Inform)

I like the premise--dealing with the demands of a brand-new baby. But
there is no detail in the coding, the puzzles are totally rudimentary,
and the tone is offputting. (The only apparent way to lose is putting
your baby in the blender.) It's a very short game, unless you get hung
up on guess-the-verb. No consideration is given to alternative ways
of doing things: for example, though you've just given birth and are
therefore presumably the mother, you can't try breastfeeding. There's
just nothing very interesting here.

"Coming Home" (Inform)

Exploring a suburban house. Don't forget to bring provisions, since you
can easily starve to death: in fact I starved to death twice trying to
find my way around the outside from the front yard to the back (directional
indicators are lacking).

This wins, hands down, the competition award for "most annoying."
Not just the starvation (though I particularly hate this puzzle)
but sub-minimal room descriptions, illogical puzzles and events (why
can't you open the refrigerator? or get into the car, even though you
can drive it?) and a faintly unwholesome flavor (your mother exists
to do scutwork for you). After about forty minutes and three
deaths (did I mention self-locking doors?) I found myself in the
bathroom, trying to piss (as the hint file assured me I must) but unable
to guess the correct verb. At this point I gave up.

"Aunt Nancy's House" (Inform)

A competantly coded, though rather sparse, simulation of a suburban
house. That's it: no NPCs, no story, no puzzles, just a VCR that
plays tapes and some lamps you can turn on and off. The coding detail
does not extend to, say, taking the soda out of the fridge, and
at least one door is fictional. There is really nothing here to
hold the player's interest.

"Pintown" (Inform)

This game is about a small time rock musician wandering around his
neighborhood. According to the walkthrough, he's trying to please
his girlfriend, though I didn't get that far.

I'm of two minds about the writing. I rather like the neighborhood,
which is quirky and non-gridlike and feels like a real place. The
obligatory sordid bits (vomit, grime) just bored me, but on the whole
this was a game I was inclined to favor.

Unfortunately it is so buggy as to be unbearable. The van won't move
because "the engine is off" but can't be turned on because "it's already
on." Described objects don't exist (i.e the boys are stoning the cat,
but not only are there no stones, there are no boys). Logical
actions don't work. I could never find a verb that would let me use
the telephone. After about forty minutes and minimal progress I
gave up in complete frustration. I couldn't get any sense of the place
or character as real when every fifth command seemed to get a buggy

"Temple of the Orc Mage" (TADS)

A stock dungeon-crawl with bare rooms, bare objects ("It's an iron key"
as a response to Examine), and nothing in the early going to hold my
interest. I got stuck early (the walkthrough appears incorrect). I
couln't motivate myself to keep trying, since the responses I was
getting to all of my attempts were frustratingly unhelpful. For me at
least, a rich environment supports working at a puzzle much better
than a sparse one does. I tire very quickly of unmotivated "You can't
take that" and "You can't do that."

Of the puzzles I did solve, all were either search/find or get
key/use key with no elaboration.

The parser appears to have reverted to Scott Adams two-word syntax,
and rejects most complex commands. This is odd, since it's a TADS


I was really impressed by the top games, and I enjoyed the whole
judging process--thanks to everyone who participated, and Happy
New Year to you all!

Mary Kuhner

Andrew Plotkin

Jan 2, 1998, 3:00:00 AM1/2/98

Mary K. Kuhner ( wrote:
> Rating 9 games

> "Sunset Over Savannah" (TADS)

> An ambitious game about beachcombing, magic, and life decisions.
> It takes place in a small beachfront area of Savannah, Georgia,
> which is wonderfully realized: every location is detailed, none
> are perfunctory. I was amazed by the richness of actions which
> the game anticipated: I was almost never balked by "You can't do
> that" or a nonsense response. I climbed over railings, built
> sand castles, tasted assorted foods, chatted with the old man, and
> nothing broke the game illusion.

> The puzzles are elegant and mostly fair. Some are quite hard--
> Andrew Plotkin was listed as a playtester, and one wonders if he
> had a hand in a few of them--but extremely good use is made of
> contextual clues. When you fail, you often learn why you failed
> and get a hint as to what to do about it. One puzzle seemed
> unfair, and one contained a bug which led to reasonable actions
> making the game unwinnable.

There should be no unwinnable states; report them as bugs. (There are a
couple of places where you may think you've lost an object for good, but
you can actually get another one *if* you need it. The supply only dries
up after you've solved the puzzle.)

And no, I didn't come up with any puzzle ideas. I assaulted Ivan with the
usual slew of suggestions of alternate solutions, alternate syntax, false
solutions, and extra responses; but all the puzzle ideas are his.



"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the

Reply all
Reply to author
0 new messages