Immersion.

9 views
Skip to first unread message

Dave Howell

unread,
Nov 2, 2010, 4:06:22 PM11/2/10
to seatt...@googlegroups.com
I pulled down a couple IF works (sadly, I don't have many already on my computer) looking for URLs to click, and found myself unable to not start playing one of them, the 2009 IF competition winner "Rover's Day Out." 

I haven't yet had nearly as negative a reaction to it as I did to "Violet," but my first few lines helped to clarify something that I'm beginning to see as fairly endemic; the engine or the script or both are compromising my immersion. 

[I assume nearly everybody knows what I mean by that. If so, skip to next paragraph. If not: when reading a book, generally the reader isn't thinking about the words on the page, or the process of turning the page; these processes are so simple that they fade into the background, and the reader is thinking about the story itself. They are 'immersed' in the story.]


The opening lines from "Rover":

Living Room (on the futon)
Where the heavy drapes meet, a sliver of sunlight shines into the otherwise dark living room. A framed picture hangs on the living room wall in the small area illuminated by the shaft of light.

>open curtains
Six impossible things before breakfast is one thing, but you have to at least
get out of bed.

>get out of bed
You can't see any such thing.

What? I exactly duplicated what the text told me to do! Fine, let's see, what else can I say . . . 

>get up
You get off the futon.

Later, I'm in the kitchen. I've found a bag of dog chow. I've taken the bag.

>put chow in bowl
Which do you mean, the water bowl, or the food bowl?

>food
Rover tears around the corner excitedly, arriving in the kitchen.

Rover notices the lack of food in his soi-disant food bowl. He paws pathetically at the food bowl, and the hollow, tin ring of the pitiful bowl echoes through the kitchen.

What? Why isn't there food in the bowl? I guess "Which do you mean..." was a rhetorical question, and the program ignored my answer? But if so, then why didn't I get some kind of response when I said "food"?

>put chow in food bowl
(first taking the food bowl)
You  empty the bag of dog chow into Rover's bowl, giving him a generous portion that should more than carry him through a day's worth of Rover antics.

Rover watches with interest as you fill his food bowl.

On the plus side, the story filled in "first taking the food bowl," instead of insisting that I type that myself. I don't know why it didn't do the same when I first tried to open the curtains, especially since it didn't even understand my first attempt to get out of bed. On the other hand . . . 

Rover stares longingly at the bowl in your hand.

>put food bowl on floor
Rover makes short work of the food in his bowl.

*I* didn't pick the stupid bowl *up*, so why do I now have to put the bowl *down?* When I feed my cats, their food bowl never leaves the floor. Obviously the authors handle serving food to pets differently than I, but it doesn't seem like feeding Rover is supposed to be a challenging puzzle. If this is the case, then why on earth can't I just "Feed dog?" 

He eyes the empty water bowl thirstily.

>fill water bowl with water
(first taking the water bowl)
Rover would not appreciate it if you put anything aside from water in his water bowl.

Well, I can sympathize with his point of view. Why did the program say this?

Rover outright gapes at the bowl you carry, then gently sniffs it in hope that you'll give it to him.

>put water bowl on floor

Rover eyes the empty water bowl thirstily.

Oh, I see. It had no idea what "with water" meant. So how do I get water in the water bowl?

>turn on faucet
You run the water momentarily, and then turn off the tap to conserve water.

What? How can I put the water in the bowl if the water appears and disappears before I can type another command? 

>put bowl under faucet
I didn't understand that sentence.

>hold bowl under faucet
You can't see any such thing.

>pour water into bowl
(the water bowl into the water bowl)
Rover would not appreciate it if you put anything aside from water in his water bowl.

>turn faucet
It is fixed in place.

>run water
That's not something you can enter.

>turn on water and put water in bowl
You can't use multiple objects with that verb.

>put water in bowl
(the sink in the water bowl)
You run the sink briefly to make sure the water is cold, and then fill Rover's bowl about halfway up.

Well, that's a serious crash-and-burn. I don't even understand what "
(the sink in the water bowl)" 
means. But up to this point, I've been asked to be fairly literal. I was half expecting to be required to "open bag of food" before I could feed Rover, but now I can (and must) 'put water in bowl' even though there is no existing "water" in my environment. 

I'm still intrigued by the story that's unfolding, but I'm also (as usual) intensely aware of the interface, which means I can't maintain immersion in the story. 

I think maybe this can be summed up by the question, am I reading this work, or playing this work? I want to read it, but I suspect I'm going to end up playing it. 
Even then, I don't expect to achieve immersion. 

I wish I had more time to study this question; I find myself wondering to what degree the broader IF community is grappling with this issue, and in what ways.


Ben Cressey

unread,
Nov 2, 2010, 7:32:02 PM11/2/10
to seatt...@googlegroups.com
On Tue, Nov 2, 2010 at 1:06 PM, Dave Howell <groups...@grandfenwick.net> wrote:

>put chow in bowl
Which do you mean, the water bowl, or the food bowl?

>food
Rover tears around the corner excitedly, arriving in the kitchen.

Rover notices the lack of food in his soi-disant food bowl. He paws pathetically at the food bowl, and the hollow, tin ring of the pitiful bowl echoes through the kitchen.

What? Why isn't there food in the bowl? I guess "Which do you mean..." was a rhetorical question, and the program ignored my answer? But if so, then why didn't I get some kind of response when I said "food"?

In this game, "food" is a one-word command that appears to mean: "Rover! Time to get your food!"

It's a weird, non-standard choice that messes up disambiguation here.

 
>put water in bowl
(the sink in the water bowl)
You run the sink briefly to make sure the water is cold, and then fill Rover's bowl about halfway up.

Well, that's a serious crash-and-burn. I don't even understand what "
(the sink in the water bowl)" 
means. But up to this point, I've been asked to be fairly literal. I was half expecting to be required to "open bag of food" before I could feed Rover, but now I can (and must) 'put water in bowl' even though there is no existing "water" in my environment. 

The source code tried to support this syntax with the phrase Understand "water" as the kitchen sink. Which makes the parser treat "water" and "sink" as synonyms, arguably a poor idea. At least their hearts were in the right place.

I would recommend using the most specific standard verb you can. So "put water in bowl" becomes "fill bowl" - or in this case "fill water bowl". Obviously it helps to know what those standard verbs are - all I can say is you internalize them with practice.

Usually that works. If the game comes back and says, "What do you want to fill the bowl with?" I elaborate with a single noun ("water"). If that doesn't work, or if "fill" is obviously unimplemented, then I have to run through the litany:

"fill bowl with water"
"put water in bowl"
"pour water in bowl"
"put bowl in sink" / "turn on water"
"help"
"hint"
"walkthrough"

Anyhow, most of the objections you raise are legitimate failures in the game's design. One approach, if you get fed up with Rover, is to check out a later IF work by a more experienced author. Generally these will allow for a wider variation in syntax.

I've only been playing IF Comp games for a few years, but for what it's worth I don't think the contest results are a good starting point for new players. The winners are the cream of the crop, but the harvest includes a lot of lightly tested, poorly implemented games. Eruption might be the only game in the lot that I'd recommend to newcomers. The story is bland and the author's motivation rather insipid, but the execution is solid all the same.


Ron Hale-Evans

unread,
Nov 2, 2010, 9:21:12 PM11/2/10
to seatt...@googlegroups.com
I'm finding this thread on intfiction.org instructive:

http://www.intfiction.org/forum/viewtopic.php?f=6&t=1283

Aaron Reed's list of Inform extensions that create a more forgiving
parser for newbies seems especially useful. They would solve Dave's
GET OUT OF BED problem, for one thing.

As a player, I find it useful to think of the parser as something like
a Unix command line. It only looks like English. It is a computer,
after all! However, this may be hard to explain to people who aren't
programmers or who have never used a command line, Unix or otherwise.

Ron

--
Ron Hale-Evans ... rw...@ludism.org ... http://ron.ludism.org ... (206) 201-1768
    Mind Performance Hacks book: http://oreilly.com/catalog/9780596101534/

Dave Howell

unread,
Nov 3, 2010, 2:08:02 AM11/3/10
to seatt...@googlegroups.com

On Nov 2, 2010, at 18:21 , Ron Hale-Evans wrote:

It's a very interesting thread, although I find what *isn't* said as interesting as what is.


> Aaron Reed's list of Inform extensions that create a more forgiving
> parser for newbies seems especially useful. They would solve Dave's
> GET OUT OF BED problem, for one thing.

It would, but somebody pointed out that adding all the fancy modules turned 16 failed parses into 9 failed parses. (They also pointed out it wasn't really a fair test, but still...)


> As a player,

Which is *exactly* my earlier point. "As a player" of a text adventure, I'm OK with the idea that the input prompt is part of the puzzle.

But "as a *reader*" of interactive fiction, I want the command prompt to be utterly transparent.

I think some of the comments in that thread point to a key issue, one of the things not being said I mentioned above.

The overwhelming focus (or so it appears to me) to solve the problem of confused 'newbies' is focused on the parser, when the problem is at least as much, if not more, with the *narrative text.*

I think I'll leave it at that for now.

Dave Howell

unread,
Nov 3, 2010, 2:08:11 AM11/3/10
to seatt...@googlegroups.com
There was one post in the thread Ron linked to that almost had me leaping out of my chair....


> katz: One of the most marked things I've noticed is the tendency to try to "talk" by just entering text. For instance, when my husband was at a part where the player in surrounded by flames that can burn you, he would type "ouch." This puzzles me because, even if the parser understood it, I'm not sure what it would accomplish.

What? Argh! The husband is responding *in character!!* It's a sign of immersion! What does it accomplish? It makes the story richer and more alive!

Thankfully, the very next post was from somebody who gets it:

> Laroquod: It's a performance — it doesn't really matter whether the game can play along. It's a consequence, I believe, of being put in the second-person mindset. You are being placed in a role. There is a temptation (at least on my part) to think of the things I am typing as part of the story rather than as purely utilitarian.

I found the post after *that* to be rather tragic, unfortunately (emphasis mine)

> Ron Newcomb: You know I do the same thing sometimes, typing expressions as if in character. It just seems natural. I only do it if I'm really into the role, though, so _it's_not_often_.

Not often? I think if I'm dealing with a work of IF and I do not spend *at*least* 50% of the time feeling immersed, I'm going to consider that work a failure. If this seems unreasonably harsh, I feel I should point out that (1) I expect 100% immersion from literature. If I'm distracted by a clumsy phrase, a misspelled word, or pages stuck together, I'm annoyed. And (2), immersion is the *minimum* requirement for enjoying a story. After than, *then* I get to find out if the plot is compelling, the characters engaging, the setting intriguing, and so on.

Games *also* require immersion. I don't play StarCraft because I want to master the mysteries of grouping units under command keys, or the fun of trying to memorize which letter makes the drone build which building. I play to do battle with alien opponents, to crush them beneath my heel, or tentacle, or appendage.

Believe me, I appreciate that the IF/text adventure's interface, the text parser, is an incredibly difficult one. It's so much easier to constrain the options by editing down to point-and-click, or shift-and-poke, or whatever. But this thread, like the field in general, seemed to be overwhelmingly about problem-solving, and only minimally about art. Alas.

Emily Short

unread,
Nov 3, 2010, 2:52:24 AM11/3/10
to seatt...@googlegroups.com

On Nov 2, 2010, at 11:08 PM, Dave Howell wrote:

> There was one post in the thread Ron linked to that almost had me leaping out of my chair....
>
>
>> katz: One of the most marked things I've noticed is the tendency to try to "talk" by just entering text. For instance, when my husband was at a part where the player in surrounded by flames that can burn you, he would type "ouch." This puzzles me because, even if the parser understood it, I'm not sure what it would accomplish.
>
> What? Argh! The husband is responding *in character!!* It's a sign of immersion! What does it accomplish? It makes the story richer and more alive!

That's absolutely the value of *the husband typing that*. But what would be the immersion-supporting thing for the parser to do in response? What would you like to have happen here?

Ron Hale-Evans

unread,
Nov 3, 2010, 3:07:01 AM11/3/10
to seatt...@googlegroups.com
On Tue, Nov 2, 2010 at 11:08 PM, Dave Howell
<groups...@grandfenwick.net> wrote:
> I think if I'm dealing with a work of IF and I do not spend *at*least* 50% of the time feeling immersed, I'm going to consider that work a failure. If this seems unreasonably harsh, I feel I should point out that (1) I expect 100% immersion from literature.

I don't expect 100% of anything from anything. (And if you're "feeling
immersed", you're not immersed.)

> If I'm distracted by a clumsy phrase, a misspelled word, or pages stuck together, I'm annoyed.

Immersion requires cooperation from the author and the reader. Fairly
often, immersion fails to happen because of something on the reader's
end. Bad lighting, to take one small example. And of course, pages
stuck together are _almost always_ the reader's fault... >;)

> And (2), immersion is the *minimum* requirement for enjoying a story. After than, *then* I get to find out if the plot is compelling, the characters engaging, the setting intriguing, and so on.

Couldn't agree less. I can't begin to get immersed in a work if plot,
setting, or characters are broken in some significant way. I can laugh
at it, though.

>But this thread, like the field in general, seemed to be overwhelmingly about problem-solving, and only minimally about art. Alas.

Problem-solving is an important part of art. When I hear you talk
about your own art, Dave, such as the Hugos you made, I hear you talk
a lot about such details as figuring out how to cast material X, or
how you had to find a substitute for material Y, and it worked out
better anyway. I seldom hear you talk about such high-level concepts
as meaning or artistic intent. Does that mean you're not an artist? I
don't think it does.

Someone once said that when amateur writers get together, they talk
about art, but when professional writers get together, they talk about
money. I think it's kind of like that.

Ben Cressey

unread,
Nov 3, 2010, 1:28:31 PM11/3/10
to seatt...@googlegroups.com
On Tue, Nov 2, 2010 at 11:08 PM, Dave Howell <groups...@grandfenwick.net> wrote:
Not often? I think if I'm dealing with a work of IF and I do not spend *at*least* 50% of the time feeling immersed, I'm going to consider that work a failure. If this seems unreasonably harsh, I feel I should point out that (1) I expect 100% immersion from literature. If I'm distracted by a clumsy phrase, a misspelled word, or pages stuck together, I'm annoyed. And (2), immersion is the *minimum* requirement for enjoying a story. After than, *then* I get to find out if the plot is compelling, the characters engaging, the setting intriguing, and so on.

Most authors of static fiction do not take reader preferences into account. They usually don't encourage you to read the chapters in a different order (as Gene Wolfe does in _The Sorceror's House_). Nor do they provide opportunities for bored readers to discover key plot elements that they may have missed while skimming. Does that mean those stories are failures? For the hypothetical reader who doesn't know the standard way to read a book, maybe it does.

Like tabletop RPGs, IF is part performance. Do you expect total immersion from every session with a tabletop RPG? Do you fault the game design for players who break character, or for GMs who fudge rolls or flub an accent? Or do you accept that an otherwise compelling game may be marred by lackluster performance?

IF has a similar learning curve. Once you gain enough experience that your part in the performance is not the weakest link, it becomes a lot more interesting. Until then you're kind of like the guy at the table who wants to hit on the NPC barmaid all night. You're not "wrong" as such, and the better GMs can channel your efforts into more productive areas, but you're still swimming against the current.

 
Games *also* require immersion. I don't play StarCraft because I want to master the mysteries of grouping units under command keys, or the fun of trying to memorize which letter makes the drone build which building. I play to do battle with alien opponents, to crush them beneath my heel, or tentacle, or appendage.

Starcraft is very "meta" to me. It's about pwning noobs and grumbling about lame strats with college buddies. But even in the single player campaign, it's hard not to fault the lack of immersion. What could be more wrenching than forcing you to rebuild the same workers and early tech buildings every 20 minutes when the next mission starts? If that doesn't break immersion, it's only because you don't want it to. There is a tendency to decry the weaknesses of IF - parser, control scheme, opacity to newcomers - without recognizing that extremely popular FPS / RTS / RPG games have equally potent flaws.

 
Believe me, I appreciate that the IF/text adventure's interface, the text parser, is an incredibly difficult one. It's so much easier to constrain the options by editing down to point-and-click, or shift-and-poke, or whatever.  But this thread, like the field in general, seemed to be overwhelmingly about problem-solving, and only minimally about art. Alas.

The art of IF comes not from the writing or the typing, but the interacting. Solving interaction problems is very much about art.
 

Lucian Smith

unread,
Nov 3, 2010, 2:12:38 PM11/3/10
to seatt...@googlegroups.com

I think 'not give you an error message'.

This is something I saw in 'Worlds Apart' and have always wanted to see in
more games. In that game, you could say or telepathically think things,
and the game would respond "You say, "<thing you said>"" or
"<italics>thing you thought</italics>". And that was it. It didn't
pretend to understand what you said or thought, it just let you do it. At
several places in the story, I used those commands just because I wanted
the protagonist to do that; to roleplay a little.

Obviously, that can't work with '>OUCH' but it could work with '>SAY
OUCH!' or even '>"OUCH!' like in a MUD. If you could train the player to
do that, I think many people would enjoy it, and that it would facilitate
immersion in the game.

-Lucian

Andrew Plotkin

unread,
Nov 3, 2010, 4:37:56 PM11/3/10
to seatt...@googlegroups.com
On Wed, 3 Nov 2010, Lucian Smith wrote:

> * Emily Short <ems...@mindspring.com> [2010-11-03 06:53] writes:
>>
>> That's absolutely the value of *the husband typing that*. But what would be the immersion-supporting thing for the parser to do in response? What would you like to have happen here?
>
> I think 'not give you an error message'.
>
> This is something I saw in 'Worlds Apart' and have always wanted to see in
> more games. In that game, you could say or telepathically think things,
> and the game would respond "You say, "<thing you said>"" or
> "<italics>thing you thought</italics>". And that was it. It didn't
> pretend to understand what you said or thought, it just let you do it. At
> several places in the story, I used those commands just because I wanted
> the protagonist to do that; to roleplay a little.

Good point.

> Obviously, that can't work with '>OUCH'

For words that are clearly (and only) exclamations, it might well work
out. This is the bucket of the standard library "DRAT" => "Quite."
interaction, and I've always liked that one.

I can easily see adding "OW/OUCH" to the swearing action. And then a game
could customize the reaction for scenes when the player has really been
hurt.

--Z

--
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*

Dave Howell

unread,
Nov 3, 2010, 7:39:32 PM11/3/10
to seatt...@googlegroups.com
By the way, after using "Rover's Day Out" in an earlier post as an example of some de-immersifying experiences, I went on to finish the game/story, and am very pleased to report that it has won a gold star for >50% immersion.


Note: I will try to be somewhat vague, but there will almost certainly be a few spoilers in the rest of this message...

There was a point about half-way through the game where my immersion was utterly shattered, but eventually I figured out it was about 75% my fault and 25% the authors. Having successfully fed the dog previously, in a later loop, I went into the kitchen and found the cabinet where the dog chow is kept already open. I couldn't figure out how to feed the dog! Where was the food! I eventually resorted to reading through a walk-through, but it was of no help at all; it just seemed to assume I would feed the dog the same way I had before. I tried feeding rover the edible floss; the game told me only chow went in the food bowl; vet's orders. (Nice!) I finally realized that this IF cabinet was acting like an IF cabinet and not a real cabinet: even though the cabinet was open, the chow was nevertheless invisible until I looked IN the cabinet. Now I happen to think if I [L]ook around the kitchen, and the cabinet is open, that I ought to have 'seen' "bag of dog chow (in cabinet)", but that's not how IF usually works, so I'll take most of the blame for that one.

On the other hand, a bit later, the boarding sequence was extremely successful. The story made it very clear that things weren't going well, but if I managed to just slow the boarders down, then there might be some hope of surviving. I found myself racking my brains; "How do I stop them? What should I do?" rereading the text for clues, and feeling quite panicky. It's a turn-based game, for crying out loud! I had as much time as I wanted, but I *felt* like events were careening out of control, and then had a huge sense of relief when rescue arrived. I was thinking much more about myself-as-character than player-at-keyboard. Wonderful! Wheee! I definitely felt like I could have or should have been more effective (and the game told me at the end that I'd only found one of eight ways to impede the boarders), but I felt like my failure to be more effective was because I couldn't figure out what to do with what I had, and NOT that my solutions would have worked perfectly if the game weren't so dumb and didn't understand my commands.

I imagine that if I'd been much cleverer, and figured out most of the solutions, that I would only started to feel panicky before being rescued, and would mostly have been feeling smug and clever. That would have been a very different 'read' of that scene, but both paths led me to the next part of the story. In fact, I don't really know if there's a way to NOT move on to the next scene from there, but who cares? I am totally down with less-clever me reading a different version of the story than more-clever me, as long as both of me stay immersed.

There's kind of an info-dump at the end, which seemed kind of clunky and forced, but it wasn't nearly bad enough to compromise immersion, and it would be been pretty hard to re-write the story to avoid it, I think. Immersion failed at the very end of the story, in part because at this point I was really engaged with the characters, and I hit the 'bad' ending with very little time to avoid it and absolutely no idea what I was supposed to do. I glared at the screen in high dudgeon, and checked the walk-through. "What? And how was I supposed to know that THAT would work?!"

If my response on reading a solution is "Oh, duh!" then I know the fault was mine. In this case, it's much less clear. In retrospect, there were a few really subtle hints near the middle of the game. I think there were definitely things that could have been included in the narrative that could have made the final solution far more clear.

I think the plot, in the end, was pretty thin, although as a literary work, "Rover's Day Out" is really really short, so it would be hard to add all that much more to the plot. Some of the writing (esp. the cue about the traitor) seemed pretty heavy-handed. On the other hand, apparently I hit most of the parser quirks right at the beginning; through most of the game, parsing was excellent, and the descriptive prose was also extremely good; I had very clear images of most of the locations in my head without having to read page after page of text to get that. Finally, the clever duality of cottage/spaceship was used to excellent effect. What I though at first was just a fun sub-layer turns into a fundamental plot mechanism. Bravo!


Dave Howell

unread,
Nov 3, 2010, 7:39:48 PM11/3/10
to seatt...@googlegroups.com

On Nov 3, 2010, at 13:37 , Andrew Plotkin wrote:

> On Wed, 3 Nov 2010, Lucian Smith wrote:
>
>> * Emily Short <ems...@mindspring.com> [2010-11-03 06:53] writes:
>>>
>>> That's absolutely the value of *the husband typing that*. But what would be the immersion-supporting thing for the parser to do in response? What would you like to have happen here?
>>
>> I think 'not give you an error message'.

Exactly.

>>
>> This is something I saw in 'Worlds Apart' and have always wanted to see in
>> more games. In that game, you could say or telepathically think things,
>> and the game would respond "You say, "<thing you said>"" or
>> "<italics>thing you thought</italics>". And that was it. It didn't
>> pretend to understand what you said or thought, it just let you do it.

> For words that are clearly (and only) exclamations, it might well work out. This is the bucket of the standard library "DRAT" => "Quite." interaction, and I've always liked that one.


>
> I can easily see adding "OW/OUCH" to the swearing action. And then a game could customize the reaction for scenes when the player has really been hurt.

And how many times in real life have you banged an elbow or stubbed a tow and said "Ow! Dang!" (or similar) and had somebody say "Careful!"
Like THAT's any help! {chuckle}

It seems like one could safely assume (most of the time) that a single word that ended in an exclamation point was some kind of expletive.

[My mind wandered a bit; I was thinking about how I'm not the only person who sometimes types expletives when the game frustrates me.
>F**k!
Error: You need to provide a noun.
and how that leads to a very obvious noun. Probably not a functional game command, but it would make me feel better. But I digress.]

Dave Howell

unread,
Nov 3, 2010, 7:40:31 PM11/3/10
to seatt...@googlegroups.com

On Nov 3, 2010, at 10:28 , Ben Cressey wrote:

> Most authors of static fiction do not take reader preferences into account.

I think this is actually quite incorrect. They often do not *consciously* take reader preference (more accurately, the reader's capabilities) into account. I'll address this in a later paragraph.

> They usually don't encourage you to read the chapters in a different order (as Gene Wolfe does in _The Sorceror's House_). Nor do they provide opportunities for bored readers to discover key plot elements that they may have missed while skimming. Does that mean those stories are failures? For the hypothetical reader who doesn't know the standard way to read a book, maybe it does.

I would consider them failures *IF* they claimed to be interactive. The term I coined almost fifteen years ago to refer to 'stuff people usually read for fun' is 'linear narrative text,' so that I could include autobiographies and other kinds of nonfiction-as-story as well as the wide field of 'traditional' fiction. "Linear" would be the key adjective in this case.

> Like tabletop RPGs, IF is part performance. Do you expect total immersion from every session with a tabletop RPG?

Expect it? No. Desire it? Yes. And I very explicitly noted that I do NOT expect 'total immersion' from IF. I'm willing to give a gold star to a work that can pull off 50%.

> Do you fault the game design for players who break character, or for GMs who fudge rolls or flub an accent? Or do you accept that an otherwise compelling game may be marred by lackluster performance?

For flubbing an accent, I'd fault the GM. Fudging rolls? Depends; why did they think they needed to? It could easily be because there's a point (for example, the oh-so-notorious idol's mouth in D&D module S1: Lair of the Lich King), which is a major screw-up by the module author, and needs to be fixed by the GM.

Did the player break character because they're hung over, or because they can't figure out what's going on or can't decrypt the rules?

> IF has a similar learning curve. Once you gain enough experience that your part in the performance is not the weakest link, it becomes a lot more interesting. Until then you're kind of like the guy at the table who wants to hit on the NPC barmaid all night. You're not "wrong" as such, and the better GMs can channel your efforts into more productive areas, but you're still swimming against the current.

A question that keeps re-appearing has to do with 'how do we more people interested in IF?'

If I'm going to write a story for people to read, I will take for granted that the potential audience has to bring a certain minimum competence to the table: e.g. they have to be able to read. My facebook posts are very different from my LiveJournal essays are very different from my IF-list emails, because I'm writing for different audiences and I *am* taking into account what they can and/or will (i.e. prefer) to read.

The page I created to describe how I made the 2009 Hugo bases was written with shorter sentences, more pictures, and in a friendly style, whereas the essays I have elsewhere on the site tend to be more verbose and more provocative. For an essay, I'm willing to sacrifice some readers in order to make the essay stronger for the people who are left. For the Hugo page, I wanted as many people as possible to read it, because I wanted their vote.

Nevertheless, in both cases, I put a lot of effort into making the barriers (like the 'learning curve') as low as possible. I don't use garish colors, or blinking text, or dancing meatloaf, or put some annoying sound file on the page. I don't have gray text on a beige background.

Yes, like reading itself, interactive fiction has a learning curve. However, I am trying to make a case for the idea that the current state of the art is presenting a curve that is needlessly steep; that there are a lot of traditions, conventions, and habits that need to be scrapped, or at least recognized for what they are.

>
>
> Games *also* require immersion. I don't play StarCraft because I want to master the mysteries of grouping units under command keys, or the fun of trying to memorize which letter makes the drone build which building. I play to do battle with alien opponents, to crush them beneath my heel, or tentacle, or appendage.
>
> Starcraft is very "meta" to me. It's about pwning noobs and grumbling about lame strats with college buddies. But even in the single player campaign, it's hard not to fault the lack of immersion. What could be more wrenching than forcing you to rebuild the same workers and early tech buildings every 20 minutes when the next mission starts? If that doesn't break immersion, it's only because you don't want it to. There is a tendency to decry the weaknesses of IF - parser, control scheme, opacity to newcomers - without recognizing that extremely popular FPS / RTS / RPG games have equally potent flaws.
>
>
> Believe me, I appreciate that the IF/text adventure's interface, the text parser, is an incredibly difficult one. It's so much easier to constrain the options by editing down to point-and-click, or shift-and-poke, or whatever. But this thread, like the field in general, seemed to be overwhelmingly about problem-solving, and only minimally about art. Alas.
>
> The art of IF comes not from the writing or the typing, but the interacting. Solving interaction problems is very much about art.
>
>

> --
> You received this message because you are subscribed to the Google Groups "Seattle IF" group.
> To post to this group, send email to seatt...@googlegroups.com.
> To unsubscribe from this group, send email to seattle-if+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/seattle-if?hl=en.

Dave Howell

unread,
Nov 3, 2010, 7:40:44 PM11/3/10
to seatt...@googlegroups.com
I hope most of you had already figured this out, but just in case....

I'm taking advantage of the small size of the seattle-if mail list, and the fact that I've met some of you in person, to take a pretty provocative approach to this discussion. Normally I would advance these ideas more gently; I'd propose one at a time, pad them with more supporting material, and just generally go more slowly. But I think this group is atypically savvy, and I'm really excited by near-term potential I see in interactive fiction, so I've disengaged a couple of the safety protocols. If anybody is annoyed by my tone, I beg your forgiveness. Please feel free to drop me a private message with your concerns or frustrations, and I will do my very best to respond politely and gracefully. Thanks.

Emily Short

unread,
Nov 3, 2010, 8:08:16 PM11/3/10
to seatt...@googlegroups.com

On Nov 3, 2010, at 4:39 PM, Dave Howell wrote:

>
> On Nov 3, 2010, at 13:37 , Andrew Plotkin wrote:
>
>> On Wed, 3 Nov 2010, Lucian Smith wrote:
>>
>>> * Emily Short <ems...@mindspring.com> [2010-11-03 06:53] writes:
>>>>
>>>> That's absolutely the value of *the husband typing that*. But what would be the immersion-supporting thing for the parser to do in response? What would you like to have happen here?
>>>
>>> I think 'not give you an error message'.
>
> Exactly.

The core problem here is a pretty fundamental one (and Lucian acknowledges it later on, but you snipped that part). If the parser assumes that all player input is valid roleplaying, then it will also accept (rather than helpfully correcting) input like

>SQUISH THE ORANGE

so you'd get

>SQUISH THE ORANGE
You say, "squish the orange."

...Which, well, isn't what I probably had in mind, and it would be a lot more helpful to the player to give a fuller error message. (One game that does try something like this is Masquerade, by Kathleen Fisher: an older work, but it would reply "You mutter something incomprehensible." if you entered something it didn't recognize. But I think there were some unfortunate effects on playability from doing this, even though I appreciated what she was trying to do.)

So, if you're not going to do that, you then have to go back to a deal where you have the author hand-check the input for verbs that seem to qualify as valid roleplaying. Inform has in fact had at least a reply to common swear words (and some less-common British ones) for the past fifteen years. [Tangent: This leads to its own problems, because some authors want their game to play dumb about those things -- e.g. when writing for schools. Which is a pity, because ten year olds are probably the prime audience for a game that has funny responses to "SHIT" and "FUCK". But there you go.]

But if you're doing that, then it gets back to the author to try to account for a good range of inputs. Some authors do. Lost Pig and Tale of The Kissing Bandit are two that surprised and amused me with unexpectedly implemented actions; the first scene of Act of Misdirection is fun this way too. And I've gotten a small amount of positive feedback from players who liked that some of my games let you SMILE, FROWN, etc., even if those gestures do nothing.

As a design goal, though, accounting for roleplaying is basically a bottomless pit, because there are always way more possible roleplaying actions/comments/exclamations that the player could make than that you can anticipate and account for properly.

In my experience it's easiest to get a good effect with this on games where the protagonist is strongly characterized, where it's both more possible to anticipate the player's attempts at role-playing and gives a stronger effect to the narrative if you have some response ready.

Other approaches, less explored so far:

-- section off spaces where the player is encouraged to enter freeform input with the explicit understanding that this input won't be parsed the same way. Blue Lacuna does this; it's worth playing through the first chapter even if you go no further than that, just to see how it approaches some of this.

-- go to a crowdsourcing approach, as with Subservient Chicken. Put your game online, comb transcripts for player responses you haven't anticipated, and continuously upgrade the game as you get more samples. Eventually converge on something more polished. Drawback: this is even more work than IF typically is, and also rules out the game's being sold or used as a comp entry ever afterward because it will already have had a public release. But we're getting closer to having the technology to do that, at least, with browser-based terps and the ability to store and send in transcripts automatically. (At least, I believe Parchment at least is working towards that.)

*

My own angle on this is that immersion is best served when the player is most able to act and least conscious of the parser as a barrier -- but that doesn't necessarily mean you can type whatever you like. It might just mean feeling extremely confident about what you can and can't do in the game. There are times when roleplaying is great and appropriate (Lost Pig again) and others where the form of engagement I have with the game is a little different.

But "it should understand whatever I do!" is not only a hopeless goal, it's actually counterproductive as a design desideratum. The constraints in the UI don't *just* keep the player from feeling like part of the world. They also do a lot to communicate what kind of game this is and how we might successfully interact with the elements of it. While I don't think you were quite asking for that, what you *are* asking for -- recognition and support for roleplaying -- becomes in the extreme case almost the same thing.

-- Emily

Ben Cressey

unread,
Nov 4, 2010, 3:12:47 PM11/4/10
to seatt...@googlegroups.com
On Wed, Nov 3, 2010 at 4:40 PM, Dave Howell <groups...@grandfenwick.net> wrote:

I would consider them failures *IF* they claimed to be interactive. The term I coined almost fifteen years ago to refer to 'stuff people usually read for fun' is 'linear narrative text,' so that I could include autobiographies and other kinds of nonfiction-as-story as well as the wide field of 'traditional' fiction. "Linear" would be the key adjective in this case.

At the risk of venturing too far afield, I would say that the linear aspect is both a convention that authors rely upon and increasingly at odds with the way most people interact with text. Stories and novels are certainly interactive: they depend on the reader to turn the page. Take away that interaction and they cannot succeed. However, they also depend on the reader to obey the author's notion of when to turn the page.

The reader can't simply skip a boring chapter and re-enter the narrative at a more compelling point, because doing so will obscure passages vital to the comprehension of subsequent events. Yet the only other choices are to slog through the tedious bits or to abandon the story altogether. George R. R. Martin is a wonderful storyteller but every "Sansa" chapter in _A Song of Ice and Fire_ is a chore. I don't like the character, I don't find her dilemmas interesting, and yet important sequences are told from her perspective. They are redeemed only by the unlikely prospect of her sudden, violent demise.

More people read more stuff today than at any other point in history, yet the pool of readers for fiction in particular has not grown in proportion. Arguably this is evidence of failure, not on the part of any one author but owing to the brittleness of traditional techniques such as foreshadowing when confronted with an audience whose attention span can be measured in seconds. It is only through the forbearance and goodwill of readers that literary devices can work at all. We play by the rules because we choose to be entertained in this way, not because the iron laws of writing compel immersion.

 
Expect it? No. Desire it? Yes. And I very explicitly noted that I do NOT expect 'total immersion' from IF. I'm willing to give a gold star to a work that can pull off 50%.

I took your earlier remarks to mean that anything less than perfect immersion would constitute a failure, and my point was that you can't expect that level of immersion until you've become so accustomed to the parser that it fades into the background. I agree that 50% is a reasonable figure. That's about what I expect, outside of IF Comp games.

 
A question that keeps re-appearing has to do with 'how do we more people interested in IF?'

Personally, I am more interested in attracting new authors and motivating the ones we have. A larger audience is one way of accomplishing that, especially if sustained commercial sales become a reality. But the IF community has other strengths which are not often discussed. People are generally willing to take a chance on a new author's work, by testing it, playing it, or dissecting its flaws and virtues in a review. It's easier to find an audience as an unknown, something that we have in common with the larger indie game developer scene.

 
Yes, like reading itself, interactive fiction has a learning curve. However, I am trying to make a case for the idea that the current state of the art is presenting a curve that is needlessly steep; that there are a lot of traditions, conventions, and habits that need to be scrapped, or at least recognized for what they are.

I feel like this point gets made more often than perhaps you recognize, and my response is that it does new authors a disservice. It suggests that fundamental concerns such as interface design are largely unsettled and in need of radical, visionary transformation. In fact the opposite is true: a modern work of IF would feel quite familiar to anyone who encountered the field in its infancy. It is all very well to ask authors to become talented writers, proficient game designers, and expert programmers, but obligating them to become innovative UX engineers strikes me as carrying it rather too far.

There is certainly some value in lowering the bar for the model player. It is set discouragingly high at the moment, and IF will not prosper through elitism. However, we do need to recognize that the conceit of a model player will always operate at some level, just as authors depend on a model reader who at a minimum is willing to read sentences in a sequential order. The parser is not so much a liability as an easy mark for criticism. Players who find it frustrating because they don't know what to type are missing the point. Players who resent it for rejecting input when they do know what to type should know enough to fault the author instead.

Dave Howell

unread,
Nov 4, 2010, 8:40:57 PM11/4/10
to seatt...@googlegroups.com

On Nov 4, 2010, at 12:12 , Ben Cressey wrote:

> At the risk of venturing too far afield,

Risk? Venture away! After all, *I* have rewritten the subject of this thread a couple of times for that very reason. {smile}

> I would say that the linear aspect is both a convention that authors rely upon and increasingly at odds with the way most people interact with text.

This seems to me to imply that people will treat all text the same. I think, rather, that people are learning to interact with text in a wider variety of ways. However, I don't think just because somebody's an avid Tweeter that they will necessarily find themselves 'at odds' with the text of a novel, for instance.

> Stories and novels are certainly interactive: they depend on the reader to turn the page. Take away that interaction and they cannot succeed. However, they also depend on the reader to obey the author's notion of when to turn the page.
>
> The reader can't simply skip a boring chapter and re-enter the narrative at a more compelling point, because doing so will obscure passages vital to the comprehension of subsequent events.

[example redacted]

Alas, this isn't always true, although I think it should be. Well, there shouldn't *be* any boring chapters. A boring chapter is a fault or failure. Assign blame as ye see fit.

Personally, I skipped over fairly large chunks of "The Lord of the Rings," either reading just the first line of paragraphs or (especially those dreadful lumps of epic poetry) skipping them entirely, and didn't feel like I'd missed a thing. {chuckle}

> More people read more stuff today than at any other point in history, yet the pool of readers for fiction in particular has not grown in proportion.

This implies you feel that it should have? I don't think that's true. About 1.5 to 2% of the US population reads more than 3 books a year. 90% of the books sold by Barnes&Noble are purchased by <10% of their customers. The people who liked reading linear narrative text were already doing it. The increase in overall reading is because people who *could* have read fiction, but *didn't*, are now reading things that they couldn't read before because those other things (twitter, RSS feeds, blah blah blah) are new.

>> Expect it? No. Desire it? Yes. And I very explicitly noted that I do NOT expect 'total immersion' from IF. I'm willing to give a gold star to a work that can pull off 50%.
>>
> I took your earlier remarks to mean that anything less than perfect immersion would constitute a failure, and my point was that you can't expect that level of immersion until you've become so accustomed to the parser that it fades into the background. I agree that 50% is a reasonable figure. That's about what I expect, outside of IF Comp games.

Ah. I regret my earlier failure to communicate clearly.

> A question that keeps re-appearing has to do with 'how do we more people interested in IF?'
>
> Personally, I am more interested in attracting new authors and motivating the ones we have.

I think that's an excellent thing to be interested in. :)


>> Yes, like reading itself, interactive fiction has a learning curve. However, I am trying to make a case for the idea that the current state of the art is presenting a curve that is needlessly steep; that there are a lot of traditions, conventions, and habits that need to be scrapped, or at least recognized for what they are.
>>
> I feel like this point gets made more often than perhaps you recognize, and my response is that it does new authors a disservice. It suggests that fundamental concerns such as interface design are largely unsettled and in need of radical, visionary transformation. In fact the opposite is true: a modern work of IF would feel quite familiar to anyone who encountered the field in its infancy. It is all very well to ask authors to become talented writers, proficient game designers, and expert programmers, but obligating them to become innovative UX engineers strikes me as carrying it rather too far.

Hmm. It looks like I'm still not communicating as clearly as I would like.

Um, first of all, what's a "UX engineer?"

Second, I can see why you feel that "scrap the conventions!" would suggest the need for visionary transformation of the interface. However, one of the points I've been trying to make is that I think that there is still too much emphasis on the interface, and it's distracting people from other areas where I think there is far greater room for improvement.

I am not *intending* to ask authors to become expert programmers or innovative UX engineers. I *am* asking them to NOT be "game designers" at all. I think if we had some traditionally-trained writers creating IF as literature, with absolutely no "game", no "puzzle," that it would re-establish the bounds of IF and give many other authors a whole new perspective on what IF can be.

I am NOT saying that IF-as-game is bad. Quite the contrary. And I think the best works of IF will almost certainly have puzzle-solving or situation-managing aspects as a critical component of the overall work.

I *do* want to ask authors to become talented writers. That is exactly what I want. However, there are some serious fundamental differences between writing for IF and writing for linear narrative text. Nearly all LNT is first or third person, but almost all IF is second person, to name just one little quirk.

There are soooo many texts available to help beginning writers learn the craft. We don't have anything like that material for beginning IF authors. So I'm not only asking the authors to get better at writing IF, I'm asking the rest of us to become better critics, and editors, and reviewers.


> However, we do need to recognize that the conceit of a model player will always operate at some level, just as authors depend on a model reader who at a minimum is willing to read sentences in a sequential order. The parser is not so much a liability as an easy mark for criticism.

I absolutely agree. And an IF author should not feel like they are required to present a low bar. It is perfectly reasonable for somebody to say "No work is perfect. With the work I intend to create, I am going to focus on this, that, and some other thing, and this may mean that my work is harder to play/read than other works of IF. That might even be a deliberate 'feature' of my work, just like some authors will write a thick accent into the text to heighten the quality of the setting at the expense of readability."

BUT, assuming that an author would like a work to be more accessible rather than less, *all other things being equal*, then the IF community should be trying to identify and ameliorate as many barriers as possible. I think there are some pretty easy targets still standing, but not in regards to the parser.

> Players who find it frustrating because they don't know what to type are missing the point.

But why do they not know what to type? My experience has often (not always, but often) been that I get stuck with something that I *needn't* have gotten stuck on *if the writing or design had been better.*

The very first example I gave from "Rover's Day Out" was when the game said "You can't open the curtains until you get out of bed," so I typed "get out of bed" and it failed. That's not really the parser's fault, although this failure could have been fixed at that point. A better solution would have been to "(get up from bed)" or whatever the automatic assumption would come out as. A third would have been to rewrite the opening text . . . oh. How very strange.

I am now completely unable to get "Rover's Day Off" to show me the opening text that I quoted before. WTF? {keeps trying} {restarts game in Zoom instead of Gargoyle} {keeps trying}

Well, I am now utterly unable to figure out how I ever managed to see "Living Room (on the futon)" the first time I played the game.

Gah! Even when I get up, then get *back* into bed, I still can't get it to complain to me that it can't open the curtains. Now it's working perfectly.

{throws hands in air}

So, er, yea. If "Rover's Day Out" had behaved THEN the way it's behaving NOW, then I would have been much happier.

Dave Howell

unread,
Nov 4, 2010, 8:53:02 PM11/4/10
to seatt...@googlegroups.com

On Nov 3, 2010, at 17:08 , Emily Short wrote:
>
> The core problem here is a pretty fundamental one (and Lucian acknowledges it later on, but you snipped that part). If the parser assumes that all player input is valid roleplaying,

I probably snipped it because nobody's ever said that the parser should assume all player input is valid roleplaying. I believe, actually, that I suggested at one point that /[A-Za-z]+!/ might work as a way to identify expletives, assuming it didn't match a more specific entry in the dictionary.

Ron Hale-Evans

unread,
Nov 5, 2010, 3:32:29 AM11/5/10
to seatt...@googlegroups.com

I think what we are discussing here is a kind of IF literacy. There is
basic literacy that encompasses not only understanding written
language but also such concepts as the table of contents, the index,
and turning the page when you come to the end of it. There is computer
literacy, which among other things at this point in history means
knowing that when you drag a file over a folder and take your finger
off the mouse button, the file will be moved to that directory. There
is also a -- granted, somewhat more arcane -- IF literacy, a cross
between the basic and computer kinds, which includes understanding
certain limitations of the average parser, such as that it is not a
posthuman AI that comprehends written English better than you do, and
the differences between what the parser will accept and ordinary human
discourse.

I'm sorry if this is obvious. It's a debate that never seems to go away.

rwhe

Andrew Plotkin

unread,
Nov 5, 2010, 12:03:04 PM11/5/10
to seatt...@googlegroups.com
On Fri, 5 Nov 2010, Ron Hale-Evans wrote:

> There is also a -- granted, somewhat more arcane -- IF literacy, a cross
> between the basic and computer kinds, which includes understanding
> certain limitations of the average parser, such as that it is not a
> posthuman AI that comprehends written English better than you do, and
> the differences between what the parser will accept and ordinary human
> discourse.

Yes, this is certainly true. I even compare it to literacy in specific
literature genres. There are skills associated with science-fiction-
reading literacy -- when you read

"In five years the penis will be obsolete," said the salesman.

...you're supposed to deduce that there's some kind of advanced
biotechology, *and* it's commonplace enough to be a product you'd buy
casually, and social attitudes towards sex are laissez-faire enough that
the salesman doesn't get arrested, and... etc. (Advanced grade: things
*aren't* changing so fast as to *actually* obsolete the penis, because
salesman always exaggerate, right?)

But a reader not familiar with SF conventions may miss all of that.

Anyhow. This conversation has been making me think about how the IF parser
conventions change very slowly. And it's pretty much gated by how fast the
standard libraries change. I may put "OUCH" into one of my games, but I
probably won't carry it forward into *all* of them. We talk about
accepting punctuation ("HELP!") and newbie flailing ("WHAT NOW?", "I'M
LOST"), but even when those are turned into extensions, they don't make it
into enough games to shift the social balance.

(Contrariwise, I have a personal tic about "LOOK SWORD", and will mark it
as a mistake in my upcoming Inform game. Albeit with an explicit error
message.)

I'm not working up to a suggestion, here, just observations.

Ben Cressey

unread,
Nov 5, 2010, 2:04:58 PM11/5/10
to seatt...@googlegroups.com
On Thu, Nov 4, 2010 at 5:40 PM, Dave Howell <groups...@grandfenwick.net> wrote:

Um, first of all, what's a "UX engineer?"

Sorry, I work for a staffing company and sometimes the recruiter jargon infects my brain. UX is an acronym for user experience and it contrasts with UI (user interface) design in ways that I won't pretend to fully understand. But UX is more focused on the human side - what task are people performing and how does it fit into the overall workflow? UI is more about widgets and graphical toolkits - should we use a combo box or a list box to present this data? The line between the two disciplines is not stark, especially when it comes to mapping requirements and modeling potential interfaces. UX people are typically better paid, though.

Anyhow, most IF libraries don't offer widgets as such. You get to pick between fixed grids of text and flowing buffers of text. You can play sound effects, show pictures, and do some crude animations. On top of that you get the parser. So talking about user interface seems to miss the point; what authors are interested in is the player experience, and which combination of the available elements results in the most compelling presentation. But they can do pretty well just by sticking with the conventional model - status bar on top, buffer in the middle, parser on bottom - and shouldn't feel compelled to experiment (or await the results of others' experiments) unless they enjoy that sort of thing. It sounds like we are in agreement there.

I am not *intending* to ask authors to become expert programmers or innovative UX engineers. I *am* asking them to NOT be "game designers" at all. I think if we had some traditionally-trained writers creating IF as literature, with absolutely no "game", no "puzzle," that it would re-establish the bounds of IF and give many other authors a whole new perspective on what IF can be.

I am NOT saying that IF-as-game is bad. Quite the contrary. And I think the best works of IF will almost certainly have puzzle-solving or situation-managing aspects as a critical component of the overall work.

I do like zarf's definition of "puzzle" as any discrete unit of interaction that advances the story or enriches the player's understanding of the environment. (I am working from memory and perhaps mangling his intent in the process.) For the player, the parser is always a puzzle to be solved. Failed solutions are inevitable, but juicy rewards should not be few and far between.

The challenge for traditional writers is to figure out what the player is doing while you tell the story. But it really is essential to approach it from the other direction: come up with interesting stuff for the player to do and try to get the story in there somewhere. That interesting stuff? Puzzles.


There are soooo many texts available to help beginning writers learn the craft. We don't have anything like that material for beginning IF authors. So I'm not only asking the authors to get better at writing IF, I'm asking the rest of us to become better critics, and editors, and reviewers.

Well, there's David Fisher's IF Gems, Graham Nelson's Craft of Adventure, MJR's Tips on Designing, and virtually everything on Emily Short's blog. That's just off the top of my head; there's a lot of high quality material out there.

 
BUT, assuming that an author would like a work to be more accessible rather than less, *all other things being equal*, then the IF community should be trying to identify and ameliorate as many barriers as possible. I think there are some pretty easy targets still standing, but not in regards to the parser.

What easy targets do you see?

I assume you have more in mind that Rover's Day Out, considering that it was the work of a team of first-time authors. Even though it won the IF Comp, it's not really fair to hold it up as an example of the typical defects in an IF work. The XYZZY Awards are a more reliable guide to the best output in the field, especially for 2009.

Ben Cressey

unread,
Nov 5, 2010, 2:14:58 PM11/5/10
to seatt...@googlegroups.com
On Fri, Nov 5, 2010 at 9:03 AM, Andrew Plotkin <zgo...@eblong.com> wrote:
 
Anyhow. This conversation has been making me think about how the IF parser conventions change very slowly. And it's pretty much gated by how fast the standard libraries change. I may put "OUCH" into one of my games, but I probably won't carry it forward into *all* of them. We talk about accepting punctuation ("HELP!") and newbie flailing ("WHAT NOW?", "I'M LOST"), but even when those are turned into extensions, they don't make it into enough games to shift the social balance.

 Sarah Morayati made a similar point in her review of Aotearoa yesterday. (In the first four paragraphs, after which the review commences.)

It strikes me as a problem that will fade in time, once Inform leaves beta and extensions become more dependable.
Reply all
Reply to author
Forward
0 new messages