Is the primarily text-based interface of IF an advantage or a disadvantage?
In terms of conveying information, a purely text-based interface seems
inferior to a graphical one. The advantage lies primarily in the ability to
create puzzles with creative solutions. Whereas a graphical game is
typically constrained by clickable hotspots and a standard set of verbs, a
text based game allows the user to intuit new verbs or discover 3rd level
objects that don't appear in the room description. On the other hand,
whereas books are often praised (as opposed to movies) for stimulating the
reader's imagination, the players's own visualization of a room/object may
hinder his ability to solve certain types of puzzles if it differs too much
from the author's intent.
I guess what I dislike about mechanical puzzles is that their difficulty
derives mainly from this limitation of the medium -- an inability to
properly visualize the device/system that is being manipulated. Oftentimes,
a mechanical puzzle would be fairly trivial if it weren't for this
limitation, thus the puzzle aspect seems forced. The planetarium in The
Dreamhold was a recent example of this, but there are countless more. The
times where I have enjoyed a mechanical puzzle have been cases where the
solution involved thinking out of the box (e.g. the machine room in The
Yes. There are lots of ways in which text is more versatile than film
or animation. As well as describing how things appear, it can also
tell you what the characters think about them. A writer can change the
mood of a piece by altering the register that he uses. Text can move
from the concrete to the abstract without the need for clumsy
symbolism, so that descriptions of tangible objects give way to
meditations on ideas associated with them. It's also possible to
describe different time periods within a sentence without jarring the
reader in the way that a film flashback is jarring, or to tell in a
page the events of many years (the whole of One Hundred Years of
Solitude is narrated at that rate).
But one thing language is notoriously bad at is accurately describing
physical objects, especially where spatial relations are important.
You don't really evoke indescribable feelings in the reader with
passages like "The yellow plug is in the blue socket; the second,
fourth and fifth yellow lights are shining; the large lever is down."
This doesn't mean there can't be puzzles in interactive fiction, of
course. It all comes down to fuzziness. The most satisfying puzzles
involve understanding and then manipulating a complex but logical
world, but that complexity can be more or less fuzzy. Text lends
itself to a world's politics, history, motives of characters, poetic or
mystical or sentimental properties of objects. Those things are all
fuzzy. Mechanics can make for interesting interactive fiction puzzles
when it effects the actual landscape, as in For a Change, because
giving the feeling of a landscape is also fuzzy. But a text
description of a brass machine with a room of its own is a hangover
from the days before graphic adventures, when the only option for
implementing such a thing was in words. The less fuzzy such a
contraption is, the better. The best medium for it is a Heath Robinson
Tha advantages and disadvantages of text input are a separate argument,
I think. As you say, the main upside is the feeling of extra freedom
for the player that comes from not having all the possibilities set out
from the start. Puzzles like Rematch, the maze in Photopia or the end
of Spider and Web leave the player feeling as though he has won through
inspiration rather than menu selection. Other games suffer because
they are obviously menus. Lock and Key would have been more effective
if it had been controlled using a mouse and some icons rather than text.
I don't mind mechanical puzzles in text games, although I agree that
there are a lot of really bad ones. I too was frustrated by the
planetarium, although I think my frustration stemmed mostly from the
Tutorial Voice constantly nagging me to sit back and watch instead of
fiddling with it. Were the Voice absent, I would still have been
frustrated (though somewhat less so) by the overly terse descriptions
that failed to provide a proper sense of scale and position. But that
was the real problem, IMO: that the descriptions of the machine were too
terse to give you any real mental picture of the position of the pieces.
Compare the machine to the cistern puzzle in the same game. Also a
mechanical puzzle, but a brilliantly discoverable one. The main
difference was that the relationships between the parts of the cistern
were described clearly, while descriptions of the relationships between
the machine parts ranged from vague to nonexistant. I think a lot of the
frustration associated with the planetarium would have been mitigated by
less aggressive hinting and more aggressive exposition.
Something I've always enjoyed about text games is the way you can
encounter an object that would be immediately recognizable to you or I,
but which the character in the story has no familiarity with. Those "Oh,
*that's* what that is!" moments are a great deal of fun to me, but they
can only occur in text games--graphics allow you to recognize the item
immediately, and then the PC's failure to understand its purpose becomes
merely an irritating reminder of the PC's lack of perception.
I don't suffer the conceit that text has to be "moving" or
"inspirational" or whatever to be worth reading. Paperback adventure
novels are just as deserving of the name "literature" as some stuffy
philosophical piece that shoves the author's self-important theories of
life down the reader's throat. IMO, literature is good (in the sense of
being well-written) if the author has good understand of language and
accomplishes what was intended, even if that intent is simply to
entertain the reader without pretense. I judge text games the same way:
not by how "meaningfull" they are, but by whether the author handled the
medium well. This criteria trancends any particular story or puzzle
genre, and places responsibility solely on the author for all successes
and failures. Bad authors produce bad puzzle, good authors produce good
puzzles, and whether the puzzle is mechanical or otherwise has nothing
to do with it.
>I've never been particularly fond of mechanical puzzles. They take a long
>time to solve, and I generally find them frustrating. I've been wondering
>about this recently, and I think my problem with them is that they don't
>seem very suited to the medium.
Actually, mechanical puzzles themselves aren't a problem in text
adventures. The issue is with the implementation, where there is no proper
emphasis on what needs to be done - it is visible, but is generally a bit
too much unnecessairy description that makes other bits of the machine look
It's more of a problem with writing - placing the "most important" stuff in
the middle of a paragraph can and will cause the user to skip over the
important detail - especially if the paragraph is stuffed with description.
The same applies the other way around - if there isn't enough description,
the player feels as if the item is basically just scenery.
>I guess what I dislike about mechanical puzzles is that their difficulty
>derives mainly from this limitation of the medium -- an inability to
>properly visualize the device/system that is being manipulated.
I can envision things clearly - just that the text isn't always written
properly or omits a few details or doesn't properly describe an object.
(E.g. "Examine car" doesn't tell you about the trunk that was referenced in
the room description.)
In fact, there are some games where I resorted to "examine all" in order to
detect every item within a room. In some cases, I wasn't able to find a
clear reference path directly to the object, or to other objects it is
related to. "Happy Hunting" as they say.
You've touched on something I find fundamental here. To me, the
defining characteristic of IF is not the text that the player reads, but
the fact that the player communicates his intention to the games in
plain English (or Italian, or...). I've played many graphic adventures
over the years, and thoroughly enjoyed some of them, but I can never
escape the feeling that I am severely constrained by the interface. I'm
not inhabiting the game's world and doing what I might logically do if I
am really there in a graphics adventure. Rather, I am fiddeling with
the interface, trying to determine what arbitrary pre-scripted action
clicking in such and such a place will allow my character to perform.
It's horribly constraining.
I agree that trying to textually describe a complex mechanical apparatus
is a daunting task, and I was at times also rather frustrated by the
Dreamhold, but I wouldn't say the solution is to just throw out all
mechanical puzzles. The solution is... just show me a picture or
diagram. You might even allow me to rotate the viewpoint and so on with
the mouse, if the device is extremely elaborate. I don't have a problem
with interacting with a machine using the mouse, as long as it is
obvious what each click will do. Just continue to allow me to
communicate using the parser as well.
When textual description works... use it. When graphics work better...
use them. With Glulx and HTML TADS, all this and more is possible.
Sound effects, sound puzzles, even voice actors are all possibilities.
Look to Future Boy! for a taste of what I'm envisioning.
Another thing, for this IF player, anyway, is that I have to be
motivated to solve a complicated puzzle. While I have encountered
puzzles on ocassion that are just so cool in and of themselves that I
can't wait to pull out my notepad, roll up my sleeves, and dig in, they
are the exception for me. I need to know why I am solving a particular
puzzle, and I have to be motivated enough by the plot to want to do so.
This was my major problem with The Dreamhold. I found the fantasy
setting uninteresting, the old amnesia plotline is not something I was
particulary dying to see again, and I saw the big plot twist coming from
miles away. If I had felt more motivated by plot and character to solve
its puzzles, I would not have given up in boredom and turned to the
hints so quickly. (Please don't take the above as a troll or slam
against Zarf. I have the utmost respect for his accomplishments and
contributions, and am thoroughly in awe of Spider and Web. The
Dreamhold just didn't do it for me, though...)
Yes. In fact some mechanical puzzles could work in text if there was an
accompanying image. (But there's still the question of whether they'd be
interesting.) And the simple act of manipulating objects is not necessarily
uninteresting. Manipulating objects is at the core of Metamorpheses, where
not only is it interesting, but the descriptions of the size & shape of
objects is quite ample. That kind of puzzle could probably work equally well
in text or graphics.
> from the start. Puzzles like Rematch, the maze in Photopia or the end
> of Spider and Web leave the player feeling as though he has won through
> inspiration rather than menu selection. Other games suffer because
Yes, Spider and Web in particular is a great example of a puzzle where I
can't imagine a way to do it in a graphical medium. Of course, I should also
point out that it was implemented using a custom Q&A style interface which
disabled certain standard IF commands.
I played the game in expert mode, and I was still annoyed by it. The basic
problem is that I already know generally what I want to do (jump from A to
B). In real life, it would be a matter of intuition and hand-eye
coordination; in a text game, it's a matter of observation and trial &
error. In short, the choice of medium has transformed the nature of the
puzzle, and not necessarily for the better.
Of course, I'm not really picking on The Dreamhold specifically. It's just
the most recent game I played with annoying mechanical puzzles. Other recent
examples are Typo!, Max Blaster & Doris DeLightning (multiple instances),
The Erudition Chamber, etc.
> fiddling with it. Were the Voice absent, I would still have been
> frustrated (though somewhat less so) by the overly terse descriptions
> that failed to provide a proper sense of scale and position. But that
> was the real problem, IMO: that the descriptions of the machine were too
> terse to give you any real mental picture of the position of the pieces.
I disagree. Actually, I had a pretty good concept of the general layout of
the device. But the difference between solving it or failing depended on
some very precise timing (e.g. is the red sphere at 72 degrees or 74 degrees
at the end of a turn). I find it silly that I have to execute a complex
maneouver with sub-second precision using a turn-based interface where a
turn appears to be longer than a second.
Describing the state of machine in greater depth might conceivably make it
easier to envision the aparatus, but then it would have to be unabashedly
revealed as a mundane mathematical problem to solve.
Also, if it already takes 8 lines to describe the state of a complex
machine, would it really be more helpful to see it described in 20?
> Compare the machine to the cistern puzzle in the same game. Also a
> mechanical puzzle, but a brilliantly discoverable one. The main
> difference was that the relationships between the parts of the cistern
> were described clearly, while descriptions of the relationships between
> the machine parts ranged from vague to nonexistant. I think a lot of the
I didn't think the cistern puzzle was brilliant, but it was definitely
comparitively easier to envision (and solve) it. To a certain extent, it's a
lot easier to describe a room (or series of rooms) than a complex object.
> medium well. This criteria trancends any particular story or puzzle
> genre, and places responsibility solely on the author for all successes
> and failures. Bad authors produce bad puzzle, good authors produce good
Any specific case can overrule the general case. My comment was that
attempts at mechanical puzzles seem to fail more often than not.
> The solution is... just show me a picture or
Andrew said the same, and I'm sure that would be fine. It won't always
be the answer, because adding even one illustration gives the thing a
different feel, in the same way that illustrating a novel does. Don't
take that as a value judgment on my part: like everybody else, I
thought The Curious Incident of the Dog in the Night-Time was a piece
> When textual description works... use it. When graphics work
> use them. With Glulx and HTML TADS, all this and more is possible.
> Sound effects, sound puzzles, even voice actors are all
> Look to Future Boy! for a taste of what I'm envisioning.
But I found the Creeeaaak! that happened every time I opened the door
quite distracting. Funny at first, but I turned the sound off pretty
That was just the sort of game I had in mind when I said that text
lends itself to "poetic or mystical or sentimental" properties of
objects. Emily Short works hard to keep you thinking about the
metaphysical implications of the transformations you're making, and I
don't think that would work as well if presented graphically.
I even briefly considered adding a little orrery diagram in the status
bar. But I decided that would add confusion, not alleviate it. (And,
indeed, you say your problem was not with *visualizing* the device.)
But I've seen both positive and negative comments about it. And note
that the entire problem of climbing the device is part of the
"optional" goal of the game. The "basic" goal only requires you to
As I've said, I wasn't shy about including optional puzzles in
Dreamhold that I expected a lot of players to *not solve*. Infocom had
some really bastardly -- but fair -- puzzles that took me weeks or
months to figure out, and I felt really good when I did so. Any such
puzzle will inevitably leave some players entirely behind.
I find it impossible to work that sort of thing into an IFComp-sized
game. Players these days (and I include myself) want the game to go by
pretty smoothly -- and, for IFComp games, in under two hours. I took
Dreamhold as an opportunity to exercise the bastard, as it were :),
while still having a game which was winnable by newcomers.
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
I'm still thinking about what to put in this space.
I didn't find the machine at all frustrating. I'm a newbie and was happy to
sit back and watch the machine. So perhaps part of the problem is a
difference in expectations between a novice and an expert?
>I disagree. Actually, I had a pretty good concept of the general layout of
>the device. But the difference between solving it or failing depended on
>some very precise timing (e.g. is the red sphere at 72 degrees or 74 degrees
>at the end of a turn). I find it silly that I have to execute a complex
>maneouver with sub-second precision using a turn-based interface where a
>turn appears to be longer than a second.
If solving the puzzle means getting the mask (which is all I managed to do
with the machine), then I found it fairly easy and non-frustrating.
Just in case, some spoiler space.
I had to try to "take mask" and the first two lack-of-success messages led me
to wait for a certain wheel to come around again. If there's more to it than
that, then I missed it.
"Yo' ideas need to be thinked befo' they are say'd" - Ian Lamb, age 3.5
http://www.cs.queensu.ca/~dalamb/ qucis->cs to reply (it's a long story...)
I see from Andrew's message that there was more to that puzzle, so my comment
is mostly irrelevant -- except to point out that the part Andrew expected
beginners to solve was fairly easy to solve.
There's more to that puzzle; also, in the expert mode even the basic
part (getting the mask) is easier.
For myself, I solved it not through observation but through sheer
repetition. Instead of figuring out what the disc's periods were,
and waiting until that, I just typed "grab [spoiler]" -- unsuccessfully --
and then typed "g" until it worked. I'm just as glad that I didn't
have a Tutorial Voice looking over my shoulder telling me not to do
that (although I did *try* it in novice mode and got a bit amused
at the Voice's annoyance).
David Goldfarb |"Hey, mister! Are you about to drag our brother off
gold...@ocf.berkeley.edu | to a bleak nether realm of despair where the
gold...@csua.berkeley.edu | future is nothing but an endless sea of anguish
| and horrible misery?"
|"We wanna go tooooo!" -- Animaniacs
Harder. (Stupid brain.)
I personally love text games for the *lack* of information that the
player has; it adds to the challenge, and the freedom to visualize a
room as the player sees fit, combined with the extra mental aerobics
required to solve puzzles of a mechanical nature, is something that is
lost in graphical games, as good as they may be. Though certainly some
games, such as the Myst series, are better off as graphical games, but
I think that's because they are well-designed games to fit the
graphical medium; a good graphical game won't work well in text, and a
good text game won't work well in graphics.
Maybe I was wrong to say that Metamopheses could work *equally* well using
text or graphics, but I think it could still work as a graphical game.
Having a little bit of mystery in the size and shape of things could make
some puzzles a little more subtle (and interesting). It occurs to me, for
example, that converting LGoP2 from a graphical game to text might have
improved it from dreadful to merely awful.
Like I said before, I think that including images could clarify certain
types of puzzles where the layout of the room/system is hard to visualize
but important. But isn't there a reason why we avoid putting graphics in our
games? I, for one, would be happy to include static images to accompany the
text *if* I could make them look good. (But there are rather severe limits
to my artistic ability.)
Well, actually I tried to solve it through repetition as well, but somehow
failed. I tried typing "d" at opportune times for more than the cycle of the
machine (at least 20 times), and it never worked. When I finally read
someone's solution on the newsgroup, I immediately said "I swear I tried
As for the suitability of such puzzles, I guess my comment would be that
when I played these games in the 80s there were far fewer options available.
If you solved a game in a few hours, you felt a bit ripped off. Now, it
doesn't seem like such a bad thing. (But actually, I never liked the
mechanical puzzles much back then either. IIRC, I used the walkthroughs
quite a bit.)
Hm...I typed "grab [other spoiler]", repetition of which worked.