IF Cruelty Ratings diagram - feedback request

72 views
Skip to first unread message

Jeremy Douglass

unread,
Sep 5, 2007, 4:25:55 PM9/5/07
to
I've created an IF Cruelty Ratings diagram, and would love feedback.
The diagram is here:

http://jeremydouglass.com/images/forums/2007/if-cruelty-redux.png

My goal was to make IF Cruelty explicable both to people rating games
(who need a logical way of evaluating them) and to people playing
games (who want a description of what situations are being warned
against and how to cope).

This based in part on the cruelty rating proposal of Zarf with
reframing by Andrew and others as discussed on ifWiki and in past r*if
threads. http://ifwiki.org/index.php/Cruelty_scale. However it may
not perfectly correspond to these past examples, as my goals prompted
me to substitute consistent vocabulary - the short definition of each
rating (in bold) is formed by reading back the flow chart decisions.

ptw...@gmail.com

unread,
Sep 5, 2007, 6:50:02 PM9/5/07
to
On Sep 5, 3:25 pm, Jeremy Douglass <jeremydougl...@gmail.com> wrote:
> I've created an IF Cruelty Ratings diagram, and would love feedback.
> The diagram is here:
>
> http://jeremydouglass.com/images/forums/2007/if-cruelty-redux.png
>

Yup, this is how I understand it. -Paul

Jeremy Douglass

unread,
Sep 5, 2007, 7:29:06 PM9/5/07
to
After a great discussion on ifMUD#craft, I've added extra information
to the diagram, which is still at
http://jeremydouglass.com/images/forums/2007/if-cruelty-redux.png

(while the old one is here: http://jeremydouglass.com/images/forums/2007/if-cruelty-redux-1.png
)

I may try reworking this entirely, as many suggestions focus around
simplifying categories and de-emphasizing undo, although that might
make it a different scale.

I also have a few observations and questions that came out of this
process, some of which have doubtless been made elsewhere:

1. Is Cruelty essentially a list of prohibitions, like MPAA ratings?
Just as a momentarily nude body may be enough to change the rating of
a 3-hour film from PG to R, is one missing warning enough to change
the rating of a massive work from Polite to Tough?

2. As compared to the amorphous "difficulty" Cruelty is a (relatively)
objective property of the system whose only ambiguity is whether or
not warnings and notifications were clear. It may be harder to
administer however because a casual reviewer to know whether they
bypassed Tough or Cruel situations before ascribing a rating.

3. There are design decisions which don't fit well in into these
scales ratings. One example is games which silently put you into an
unwinnable state which you can undo (e.g. with infinite undo) - for
practical purposes this seems cruel, but they aren't irrevocable so
they don't fit the definition as given. Another example is games
which kill you without warning, but always allow you to undo.
Currently the diagram would rate these Polite (they can't be Tough, as
they allow undo) but polite strongly implies warnings about death, not
just irrevocable death. What this really calls for is a spur-category
called "Rude" - unwarned but undoable death-fests (e.g. the first turn
of Rematch is rude, Shrapnel appears (but isn't) rude, etc.). But then
it isn't a scale anymore, but a tree.

Cumberland Games & Diversions

unread,
Sep 6, 2007, 12:46:49 AM9/6/07
to
To the (perhaps minimal) extent that I understand Cruelty, this chart
agrees with my understanding of Cruelty.

I do have one question (with a couple of facets): The examples in the
diagram seem to imply that a "warning" is something that happens at or
near the critical decision-point. What is the Cruelty impact of (A)
warnings that appear much earlier (B) warnings of a general rather
than specific nature provided as part of the game's documentation
[online or in the feelies], (C) warnings that are deliberately cryptic
and/or themselves rewards for solving a puzzle?

Passing a brass key, you attempt to cross a rickety bridge ...

(A) In another region, in another scene, five to seven play-hours ago,
an NPC remarked "Be warned, traveler ... some bridges, once crossed,
offer no hope of return."
(B) A note in the documentation indicates that some areas are
unescapable traps and that some items can be permanently lost or
destroyed. (B1) and explicitly mentions making the game unwinnable as
a result (B2) leaves it to the reader to infer that this might make
the game unwinnable.
(C) If the player solves the hokey color-coded anagram plate-stacking
folksong puzzle next door to the bridge (while wearing the correct
underwear stolen from a Grue), the player earns a scroll that lists a
secret government list of "Bridges that are structurally unsound and
could strand or even kill wandering adventurers."

To me it feels like there are so many _different_ ways to warn a
player of certain risks that ... while most of these factors can be
defined objectively (unwinnable, irrevocable, etc), the idea of a game
"threatening" to become irrevocably un-won, or "warning" about the
possibility ... is a lot more rubbery, a lot more open to variations
not only in approach, but in designer intent, making "abrupt" more of
a judgment call.

Or, in BRIEF mode: The line between Tough and Nasty seems less rigid
than the other category divisions.

Bear in mind that I'm largely clueless.

Ryusui

unread,
Sep 6, 2007, 4:27:28 AM9/6/07
to
I personally like the chart. ^_^ Just a little typo warning: the
example for "Tough" says "with escape and no undo".

Victor Gijsbers

unread,
Sep 6, 2007, 5:33:03 AM9/6/07
to

Looks good, but change "they key" into "the key" in the cruel example. :)

Regards,
Victor

Jeremy Douglass

unread,
Sep 6, 2007, 2:29:17 PM9/6/07
to
On Sep 5, 9:46 pm, Cumberland Games & Diversions <sj...@io.com> wrote:
> I do have one question (with a couple of facets): The examples in the
> diagram seem to imply that a "warning" is something that happens at or
> near the critical decision-point. What is the Cruelty impact of (A)
> warnings that appear much earlier (B) warnings of a general rather
> than specific nature provided as part of the game's documentation
> [online or in the feelies], (C) warnings that are deliberately cryptic
> and/or themselves rewards for solving a puzzle?

This raises some really interesting design issues. My inclination is
to be conservative about "warning" - that warning is a direct
identification of risk, given clearly and at reasonable proximity in
time and space to the risk. Here is my logic:

A game might casually mention the phrase "rickety bridges" in its
opening infodump or supporting feelies, or it might leave a scrambled
phrase on a rock elsewhere that spells BEWARE THE BRIDGE, and all of
these affect gameplay. I don't think they should be primary
considerations in rating though, because the purpose of ratings are to
advise players on what treatment to expect and what play strategies to
adopt - should you save never, periodically, or paranoically? Any
warnings that aren't reasonably clear or near the risk, or that
require deductive or lateral thinking don't count as "you should have
known" - they are "you COULD have known."

Unclear, distant warnings are fine design ideas and can make for great
games. They don't count towards lowering a game rating from Nasty to
Tough, however, because you can't in good conscience tell a friend
"when it looks like you should save, do it" - instead you should tell
them "save every now and then just in case," because the warnings may
be present in theory, but your friend could very well miss them. I
think the same applies to notices of unwinnability given only after
the fact - there are many ways to vaguely hint that a game has now
become unwinnable, but if the hints are obscure, happen many turns
after the event, or require lateral thinking, I wouldn't lower a
rating to Nasty - it should still be Cruel. The point of ratings are
not whether the game contains some element that technically qualifies
as warning or notice - the point is whether the average player can
reasonably expect to experience the level of warning or notice
indicated by the given rating.

Of course, that is my thinking - I'd be interested to hear if others
agree or disagree.

Also, thanks to Ryusul, Victor and Jacqueline for typo help - they've
been fixed.

Cumberland Games & Diversions

unread,
Sep 6, 2007, 4:50:22 PM9/6/07
to
> [...] They don't count towards lowering a game rating from Nasty to

> Tough, however, because you can't in good conscience tell a friend
> "when it looks like you should save, do it" - instead you should tell
> them "save every now and then just in case," because the warnings may
> be present in theory, but your friend could very well miss them. [...] The point of ratings are

> not whether the game contains some element that technically qualifies
> as warning or notice - the point is whether the average player can
> reasonably expect to experience the level of warning or notice
> indicated by the given rating.
>
> Of course, that is my thinking - I'd be interested to hear if others
> agree or disagree.

Your reasoning looks great to me, and I appreciate the substantive
response. As a newb and an outsider, it's really great how this
newsgroup never makes me _feel_ like a newb and an outsider.

Poster

unread,
Sep 7, 2007, 6:31:19 PM9/7/07
to

A fantastic idea. Clear, concise, and unambiguous.

-- Poster

www.intaligo.com Building, INFORM, Seasons (upcoming!)

Faust

unread,
Sep 9, 2007, 7:51:38 PM9/9/07
to
This pretty much exactly mirrors my own personal thoughts on cruelty
of individual puzzles -- however, I think it's tricky to apply labels
to entire games based on it. For example, if an otherwise mild game
had a moment in the final showdown where you could die, I wouldn't
want that to color anyone's enjoyment of the other 95% of the game.

Maybe this should be a puzzle cruelty scale. Then you could say, the
game has four cruel puzzles or whatever. If a game had more than a
handful of cruel moments, then maybe it'd be appropriate to label the
game as a whole. Just a thought about scale of application, though --
nice chart!

Andrew Plotkin

unread,
Sep 9, 2007, 11:46:37 PM9/9/07
to
Here, Faust <guenth...@gmail.com> wrote:
> This pretty much exactly mirrors my own personal thoughts on cruelty
> of individual puzzles -- however, I think it's tricky to apply labels
> to entire games based on it. For example, if an otherwise mild game
> had a moment in the final showdown where you could die, I wouldn't
> want that to color anyone's enjoyment of the other 95% of the game.

The original "scale" idea is really more about what the player
expects, and whether the game violates his expectations.

So the example you describe is actually the *worst* arrangement. The
game structure leads the player to expect safety, and then yanks it
away -- perhaps leaving him without a saved game. That's exactly the
kind of wrench that led to me using the term "cruel" in the first
place.

(Unless, of course, the game actually comes forward and changes the
player's expectation in some way. Such as the classic "Oh no! Do you
want to save before doing that?" :)

(In contrast, a game which offers constant violent death right from the
beginning will give the player a clear idea what to expect.)

In a broader sense, of course you're right -- any work can be
inconsistent, and it's important to be able to say "Well, this part
has this problem, but this other part doesn't." But you might as well
express the definitions simply -- "A game is X if..." -- and worry
about the grey shading when you talk about particular examples.

--Z

--
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*
If the Bush administration hasn't thrown you in military prison without trial,
it's for one reason: they don't feel like it. Not because you're an American.

Faust

unread,
Sep 10, 2007, 3:46:40 AM9/10/07
to
On Sep 10, 10:46 am, Andrew Plotkin <erkyr...@eblong.com> wrote:

>
> So the example you describe is actually the *worst* arrangement. The
> game structure leads the player to expect safety, and then yanks it
> away -- perhaps leaving him without a saved game. That's exactly the
> kind of wrench that led to me using the term "cruel" in the first
> place.
>
> (Unless, of course, the game actually comes forward and changes the
> player's expectation in some way. Such as the classic "Oh no! Do you
> want to save before doing that?" :)
>
> (In contrast, a game which offers constant violent death right from the
> beginning will give the player a clear idea what to expect.)
>
> In a broader sense, of course you're right -- any work can be
> inconsistent, and it's important to be able to say "Well, this part
> has this problem, but this other part doesn't." But you might as well
> express the definitions simply -- "A game is X if..." -- and worry
> about the grey shading when you talk about particular examples.
>
> --Z

You're right, a game where the concept of death is introduced without
warning into the endgame isn't the best example... I was rushing when
I posted that. I was thinking more of something like "Anchorhead",
where particular puzzles are astonishingly cruel and/or obscure (the
loony bin sequence), but where a lot of the time, while she's
strolling around, the PC is pretty safe from harm. I just don't know
how I'd rate Anchorhead's cruelty, or if reading someone else's
holistic rating of its cruelty would help me that much.

I'm not saying there's no merit in the holistic rating concept. I use
two holistic ratings myself. To me, adventure game cruelty is binary.
either a game is "LucasArts-type", or it isn't. If it isn't, I have to
save all the time, my playing is overshadowed by a vague sense of
impending doom, and I get pissed off more often. If it is "LucasArts-
type", I kick back and enjoy exploring the game universe. This
flowchart is only slightly more complex than my system, so overall I
approve.

Cumberland Games & Diversions

unread,
Sep 10, 2007, 12:45:16 PM9/10/07
to
> I'm not saying there's no merit in the holistic rating concept. I use
> two holistic ratings myself. To me, adventure game cruelty is binary.
> either a game is "LucasArts-type", or it isn't. If it isn't, I have to
> save all the time, my playing is overshadowed by a vague sense of
> impending doom, and I get pissed off more often.

Yeah. For me, the pleasure of an adventure game depends a lot on the
sense of character perspective ... and when my most common "strategy"
is the SAVE command, that's about me, not the character.

I like it best when SAVE is just that thing I type when it's time to
turn the computer off and go to sleep, and my most common strategy is
GET ALL :)

That said, I do think the Cruelty scale is nifty and useful.

steve....@gmail.com

unread,
Sep 10, 2007, 3:00:17 PM9/10/07
to
Faust wrote:
> I'm not saying there's no merit in the holistic rating concept.

Yes, but this would be an interesting critical thought, for exactly
the reason you initially expressed. The purpose of the graph is not to
theorize in the abstract (as Andrew suggests), but to characterize
specific games in broad terms. The problem with broad characterization
is that it assumes that the game is approximately-equally cruel all-
the-way-through, which is either a spoiler insofar as we have to
explain the exceptions, or gives the player a set of expectations
which brashly override and level-out the more subtle scenarios and
clues given -- with much more care and attention -- by the game
designer himself.

It's not a problem for games which are entirely standard and moderate
in the cruelty respect, but it's a problem for games which are not.
The assumption of typicality is not an assumption which advances the
genre. In a word, it's reductionist.

Michael Martin

unread,
Sep 10, 2007, 4:43:11 PM9/10/07
to
On Sep 10, 12:00 pm, steve.bres...@gmail.com wrote:
> The problem with broad characterization
> is that it assumes that the game is approximately-equally cruel all-
> the-way-through, which is either a spoiler insofar as we have to
> explain the exceptions, or gives the player a set of expectations
> which brashly override and level-out the more subtle scenarios and
> clues given -- with much more care and attention -- by the game
> designer himself.

I started noticing around the 2006 comp a tendency towards having an
out-of-band marker that indicates that one is entering a lethal
endgame. I'm open to the claim that this counts as a spoiler, but I'm
not sure it's more of one than any other kind of command prompting.

One example that didn't really work for me was For A Change, in which
you get an "OK, the game isn't winnable, but this doesn't kill you"
message in-band and essentially without warning (cuing UNDO). A
stranger example is Spider and Web, in which you are warned at the
beginning that the game will go lethal at a certain point that you
will recognize upon reaching it (cuing SAVE). It's right (about the
recognizability, and the lethality), but I'm not sure it generalizes.

--Michael

steve....@gmail.com

unread,
Sep 10, 2007, 7:34:37 PM9/10/07
to
Jeremy, the chart looks quite right -- and it's nice to hear that
ifMUD was harnessed towards a discussion useful to your interests.
Generally they enthusiastically mundanify between spasms of
personality-occultism and off-topic casual/social-leveraging. (Has
anything useful come from them? Oh yes, the Xyzzy mudsturbation
awards.... >APPLAUD)

I would only suggest that this cruelty-designation scheme is probably
a useful framework against which the IF writer can consider his
various efforts (one of many considerations, right?). -- But this is
probably too reductionist/simplistic for an evaluation of IF works,
such evaluation being an entirely different kind of operation.

Cumberland Games & Diversions

unread,
Sep 11, 2007, 1:00:02 AM9/11/07
to

> I would only suggest that this cruelty-designation scheme is probably
> a useful framework against which the IF writer can consider his
> various efforts (one of many considerations, right?).

That's why I like Cruelty most of all ... it crystallizes some design
decisions that might otherwise have remained nebulous ... That said,
while I have kept the Cruelty concept in mind, I'm not inclined to be
"consistent" within a single design ... For example, I enjoy the
structure of Plundered Hearts, where the game begins with a timed life-
or-death puzzle set (and it's very possible to get the game into a
state where only RESTART will do the trick), but once you save the
ship and get past the opening crisis, the game becomes much more
exploratory and no longer rushed, since you've won the right to relax
and go puzzling :) While that particular game as a whole might count
as "Tough" (by degrees), it's very easy for me to imagine a successful
use of that structure in a new game where the opening crisis is
"Tough" and the post-crisis exploratory phase remains "Polite" (and
that is, in fact, the structure of the game I'm currently working on).

Reply all
Reply to author
Forward
0 new messages