[bigjko]

Brilliant article, and, while reading it, spawned a few puzzle ideas in my head.

[Cellardoor]

Indeed, very nicely done. There is, in my opinion, a much better way to get rid of what you so cleverly call the Keyring Syndrome:

Spoiler:

Return to the text parser interface.

A nice addition; One of the reasons the VGA remakes of old Sierra games weren't very succesful is because the puzzles became way too easy with the new mouse interface. LSL1's "Tie rope around waist" puzzle comes to mind.

[Steve Ince]

The only way to bring back a text parser is by incorporating some kind of voice-recognition interface.

The key to making puzzles interesting is by layering in the complexity, not having an interface that creates complexity. In this day and age you wouldn't get a game published commercially if you didn't have a simple interface. That's not to say that the interface shouldn't be able to do lots of things, just that the way it's delivered needs to be simple and intuitive.

[Wormsie]

Originally Posted by Cellardoor

Indeed, very nicely done. There is, in my opinion, a much better way to get rid of what you so cleverly call the Keyring Syndrome:

Spoiler:

Return to the text parser interface.

No, the point of the article was that you don't need the tedious text parser interface, but that the introduction of point-and-click has gradually lead to worse puzzles. It's not a better way, because parsers are otherwise, to put it bluntly, lame.

[Cellardoor]

Originally Posted by Wormsie

No, the point of the article was that you don't need the tedious text parser interface,

I know it isn't - it's just my opinion.

The key to making puzzles interesting is by layering in the complexity, not having an interface that creates complexity.

What about an interface that reduces complexity? That's what the LSL1 example is about. The puzzle itself is complex because of the ingenuity of the puzzle, not because of the parser interface. When the game got updated to point and click, the puzzle became way simpler.

[sethsez]

Originally Posted by Cellardoor

That's what the LSL1 example is about. The puzzle itself is complex because of the ingenuity of the puzzle, not because of the parser interface. When the game got updated to point and click, the puzzle became way simpler.

But that's because the puzzle was designed with a parser in mind. A good puzzle designed with point and click in mind might suck if it was attempted with a parser.

[Jake]

Originally Posted by Cellardoor

What about an interface that reduces complexity? That's what the LSL1 example is about. The puzzle itself is complex because of the ingenuity of the puzzle, not because of the parser interface. When the game got updated to point and click, the puzzle became way simpler.

I think you're not even listening to what Ince is saying, or you missed the point. You basically just repeated (an incorrect version of) what he said.

[Ninja Dodo]

I agree with Steve, the interface needs to be simple and easy to use. Having lots of differents "verbs" to use for interaction is not necessarily a good thing. In my opinion it's much better to have a small but solid set of actions that you can do, but have them extremely flexible so that you can combine them however you want and anything you could logically do with them you actually can do.

[Nellie]

Am I the only one here who enjoys not having each possible interaction laid out for me on a plate, so I get to use my intuition to decide how to act? This can have very entertaining results.

Using Future Boy as an example, the very opening scene has you falling through the sky to your doom. The parser presents a blank cursor, and what to do is entirely up to me. So I type:

>fly
Gravity begs to differ.

Hmm... maybe I need to kickstart flying some other way?

>jump
And yet strangely you do the opposite: continue falling.

How about...

>up
Down is probably a better bet.

For a laugh, I type:

>down
Well obviously.

And for another laugh I try:

>plummet

With absolutely no expectation that the designers have written a response to this rather obscure command. The reply comes:

That you can do.

Which makes me laugh. Eventually I move on from having fun with commands and go on to solve the puzzle, with help from the hints that are now popping up with each action I try.


How is this a complex interface? It takes extremely simple grammar: open door; give pipe to Sherlock; ask Henry about wives; tie bungee cord to coyote - anybody with a basic grasp of the English language can pick up the required structures very quickly (more quickly than it took for me to learn to be comfortable using a joystick, or how to play a racing game without smashing into every corner, for example). The actual process of giving commands is not complex at all.

I think the real complaint with parsers is that the potential range of commands is virtually unlimited. This can give the impression that you'll never be able to work out what to do because there's so many options - zeroing in on the 'correct' one seems an insane task. This is certainly true for badly designed text adventures (or unapologetically difficult ones), but is not an absolute rule for every text adventure in existence, any more than stupid puzzles and pixel hunting in one graphic adventure doesn't condemn all graphic adventures to including the same problems. Good design is key. A good designer will anticipate as many of the player's actions as possible, leading to logical responses for even relatively obscure commands like 'plummet' in the Future Boy example. A good designer will also make sure than none of the commands required are too obscure for the player to figure out - much as a good graphic adventure designer does with their puzzles. With good design, the intimidation of the vast range of commands available to you is merely an illusion. The game is won through with simple intuition.

As a (former) keen graphic adventure fan who, having played some sophisticated recent text adventures, has realised the potential of the parser beyond my memories of 8-bit text adventures, or Sierra's limited AGI parser (or even the horrendous conversation-bots of Starship Titanic, which some crazy people actually liked! ), it concerns me to see the parser summarily dismissed as a bad interface. A parser allows you to exercise your intuition, provides enormous density of interaction (copyright Steve Ince), and frequently entertains the player by responding to actions they expected no response from (Al Lowe has a good story about demoing LSL1 to a bunch of suits, and asking them to suggest commands. One of them shouted "masturbate!" The game duly responded: "The whole point was to stop doing that, Larry.")

All we are saying, is give Parse a chance.



For my next post: War and Peace: the extended edition.

[Steve Ince]

Point-and-click adventures have a hard time as it is. Imagine what a publisher would say to me if a game proposal of mine was to include a text parser.

[Tramboi]

Originally Posted by Steve Ince

The only way to bring back a text parser is by incorporating some kind of voice-recognition interface.

The key to making puzzles interesting is by layering in the complexity, not having an interface that creates complexity. In this day and age you wouldn't get a game published commercially if you didn't have a simple interface. That's not to say that the interface shouldn't be able to do lots of things, just that the way it's delivered needs to be simple and intuitive.

I both agree and disagree.
The "USE" concept has been the dawn of this keyring syndrome. It basically allow you to solve the puzzle by trying everything because of making the combinatorial explosion of possibilities manageable.
And writing clever lines is so inexpensive that you can reward the player often , even when he fails, when he tries something.
Then I've been bitten too much by parsers

[Tramboi]

Originally Posted by Steve Ince

Point-and-click adventures have a hard time as it is. Imagine what a publisher would say to me if a game proposal of mine was to include a text parser.

Maybe an hyperthreaded ps3.0 64bit with HDR and MRT support text parser could be sponsored by ATI or NVidia?

[Steve Ince]

Originally Posted by Tramboi

And writing clever lines is so inexpensive

That's not true. If a game is translated into five languages the cost of translation and recording can be 100,000 Euros or more. So the more clever lines you write the more it costs.

[Intrepid Homoludens]

Tramboi, I believe anything you add to a game's content increases the cost one way or another.

[Tramboi]

Originally Posted by Steve Ince

That's not true. If a game is translated into five languages the cost of translation and recording can be 100,000 Euros or more. So the more clever lines you write the more it costs.

I wasn't thinking of recording of this context, just plain text

[Tramboi]

Originally Posted by Intrepid Homoludens

Tramboi, I believe anything you add to a game's content increases the cost one way or another.

Of course but a professionnal line of text is still cheaper that a 512*512*32bpp professionnal texture, a professionnal 4000-polys mesh, or a professionnal 48kHz sound. I guess.

[Wormsie]

Originally Posted by Nellie

How is this a complex interface? It takes extremely simple grammar: open door; give pipe to Sherlock; ask Henry about wives; tie bungee cord to coyote - anybody with a basic grasp of the English language can pick up the required structures very quickly (more quickly than it took for me to learn to be comfortable using a joystick, or how to play a racing game without smashing into every corner, for example). The actual process of giving commands is not complex at all.

Wot do plummet mean?

That was a rhetorical question.

Is it any better to randomly start thinking up words or randomly click points on the screen? Well, the first option requires SOME thinking, at least.

[Tramboi]

Originally Posted by Wormsie

Wot do plummet mean?

That was a rhetorical question.

Ain't you supposed being busy becoming an adult, Wormsie?

[Fairygdmther]

Originally Posted by Tramboi

Ain't you supposed being busy becoming an adult, Wormsie?

Nope, you got it backwards - he wants to stay as a kid. And plummet is what ripe plums do off a tree!

FGM-Lyn

[Nellie]

Point-and-click adventures have a hard time as it is. Imagine what a publisher would say to me if a game proposal of mine was to include a text parser.

Oh, hell, they'd probably call the men in white suits.

Is it any better to randomly start thinking up words or randomly click points on the screen? Well, the first option requires SOME thinking, at least.

Be fair, now. If I was randomly thinking up words, I would have been typing commands like 'shoe', 'hit the crumpet with the weasel', or 'eat carpet'. My commands were intuitive reactions to the situation I was in (even if some were deliberately not very useful, where I just wanted to see what would happen). If they hadn't been intuitive, then they wouldn't have been anticipated by the designers and received the appropriate (and entertaining) responses that they did. The solution to the puzzle was also an action intuitively derived from the situation. Witness the power of the parser!

And plummet is what ripe plums do off a tree!

*ba-dum-tish!*