P&C vs. Parser Interface

[KatieHal]

The conversation has clearly gone away from Laura Bow, so, topic split!

[Blackthorne]

You know, it's not that I like either Parser or Point and Click better than each other.  I just think people that don't or won't play a Sierra game that uses parser are doing a major disservice to themselves.

Some amazing, fun and beautiful games were made using the parser.  Yes, it may be a little more work to type a lot - but it pays off.  I'm just saying don't deny yourself the experience because parser intimidates or frustrates you.

Personally, I think the best system would be a Point, Click and Parser system.  The combo of all three would make me explode in my pants.


Bt

[MusicallyInspired]

The problem is, people who are used to P&C don't want to have to do more work with a parser. All the opportunities it provides over P&C's limitations seem like chores and more work instead so they find no joy in it. And all the limitations of P&C become "more accessible" elements of game design. If one were to give them a chance, however, and get used to it they'd enjoy it at least as much as P&C.

[Lambonius]

TSL would have been much better with a parser interface.  Especially if the player had to match the length and verbal diarrhea of the narrations for each parser input.   

[KatieHal]

We thought about it, but we wanted to make sure you'd be able to play it, Lamb. 

[Blackthorne]

Hoo, BURN!!


Bt

[Lambonius]

I just want to be able to type things like:

> assuage personal guilt over unsuccessful attempts to locate kidnapped infant son by asking melancholy prince about suppressed childhood traumas

and

> mentally equate turbulent weather conditions to lifetime of personal struggles with dichotomy of good and evil


Although, I suppose TALK MAN and LOOK TREE would work just as well. 

[Big C from Cauney island]

Parser interface was cool growing up.  It definitely made you think, albeit sometimes frustrating.  As someone else mentioned, KQ3 was a great example of a game that wouldn't have worked without parser.  I'm not knocking it.  But I think that games as a whole have to evolve, its only natural.  I wouldn't want to play the same game over and over again. 

Think about it. Atari 2600 had 1 button. Commodore 64 was capable of typing, but almost every game I played was with a simple joystick.  Then more parser focus. Then point and click. Then all sorts of motion control devices, and entirely new game concepts period. A lot has changed, and I don't think it's all bad. What would be bad is no innovation.  The only gripe I have is that I loved the story focus on space and kings quest. It was great to follow the characters with each release.  And who knows? there might be games out now that do the same thing, or at least get close. I'm not a huge game player. Assassins creed did a great job combining elements, though I still was more connected to the old stuff. This could have been partially due to my age and interests at the time. 

Bottom line, some like parser some like p and c.  At this point, it doesn't matter to me as long as there is a great story to follow.  The F.E.A.R. series had an ok story, I just didn't like fear 3. Fear 2 was awesome.  Also, Mafia 2 has an awesome story, really great game and you can get it cheap on steam. Another great example of an action game with cool story to follow. Neither of which even have p and c.

Evolution is only natural, and as I posted earlier I think things will come full circle once the new technology stuff like kinekt and ps move become more common place.  But as for how long that will be, who knows.

[Lambonius]

Evolution is only natural, and as I posted earlier I think things will come full circle once the new technology stuff like kinekt and ps move become more common place.  But as for how long that will be, who knows.

Off topic, but I really have to say that I think motion control devices like kinect and PS move are a passing fad.  Wii has been out for a full generation and they STILL have yet to make a game where motion controls work better than a traditional controller.

[MusicallyInspired]

All this talk of parser reminded me of a recent Telltale blog that was posted where a guy was talking about working on a pseudo-parser speech system for future games.

[wilco64256]

All this talk of parser reminded me of a recent Telltale blog that was posted where a guy was talking about working on a pseudo-parser speech system for future games.

Nothing would make me happier than to stand in front of a Kinect and shout, "Throw baby!"

[MikPal]

As a person who doesn't speak english as mother tongue, I have to say that I feel P&C-games work much better. Even though I have no problem with playing games with a parser, there are just some moments where the language barrier just stops you cold. Got stuck with "call dispatch for warrant" in Police Quest 2 for five years.

Nothing would make me happier than to stand in front of a Kinect and shout, "Throw baby!"

I think Project Rub for the DS makes you say "I love you" to the mike. Luckily Last Window only required me to blow on it, but I'm pretty sure the other folk in the train thought I was just making out with the DS.

[Damar]

Personally, I do prefer parsers.  It does give the illusion that you can do more in the game.  But I do think it is an illusion, because for a great many things the answer is going to be "You can't do that."  The freedom of the parser is only superficial.  And while I would agree that the point and click was a bit of a dumbing down, it wasn't fatal.  For example, in a game like SQ4, I really can't think of anything that I could have done with the parser that wasn't already thought out by the icon options.  I never felt, "Ooh, I wish I could type 'do such and such.'"  The point and click interface was designed well enough that I didn't feel I was missing out.  And in some cases, like SQ6, the point and click actually seemed more immersive and complex than parser had been in the past.  Rooms like the Delta Burkselon lab and the elevator had a huge selection with the point and click icons.  The point is that in its execution, point and click can be a complex interface.  Think about it, there's only so many things you can do anyway.  You can look, touch, smell, and taste.  Whether those options are available in icon or parser form, it doesn't change how many actions are doable.  Like I said, parser just gives the illusion that you have more to do.  I might be able to type "Get freaky with Valanice" in the tower of KQII, but that doesn't it's really an option.  I'll just get a "I don't understand freaky" message.

Of course the exact opposite is true when dealing with games with a simplistic interface like KQ7 or MOE.  If point and click is technically dumbed down, those games are full on lobotomies.  That did affect the gameplay and there was a ton of stuff I would have loved to look at and explore in KQ7 that I just couldn't.  Because the KQ7 interface, to borrow a phrase I read somewhere, sucked the sweat off of Lamb's nuts.

[MusicallyInspired]

For example, in a game like SQ4, I really can't think of anything that I could have done with the parser that wasn't already thought out by the icon options.  I never felt, "Ooh, I wish I could type 'do such and such.'"

Yes but you have to remember, that game was designed specifically for the P&C interface. If there were a parser involved (which Scott wanted) there would definitely be many more interactions, ways of doing things, and even more or alternate puzzle designs. Actually the story probably would have been completely different because Ken Williams forcing the P&C interface after promising to leave the decision up to Scott played a part in the many events that forced SQ4 to be so dark.

Whether those options are available in icon or parser form, it doesn't change how many actions are doable.  Like I said, parser just gives the illusion that you have more to do.  I might be able to type "Get freaky with Valanice" in the tower of KQII, but that doesn't it's really an option.  I'll just get a "I don't understand freaky" message.

I beg to differ here. We aren't just talking about utilizing each of the 5 senses when it comes to interaction. In fact there are many different ways to interact with objects with each of the senses in themselves. Take the hand icon. How do you want to use it with the object? Stroke it to feel its texture? Try to pick it up? Attempt to push or pull it? Jump on it? Kick it? Throw it? There are many ways to "use" an object and that's just for the hand icon. What about talking? Do you want to speak to it? Whistle to it? Blow on it? If a game were designed around that kind of freedom in interactions then a whole new world of puzzle design opportunities (some more complex than simply "use") open up to the designer.

Take this example of a parser interaction puzzle for instance: Picture a rock face on a giant boulder. Looking at it yields no valuable information other than it looks like rock face. "Using" it would not be enough to solve the puzzle ("What do you want to do with it?"). "Pushing" it also doesn't work. Then you decide to "feel" the rock face. Upon doing so you feel the texture of the rock face and sense a depression that was otherwise undetectable to the eye (maybe it was covered in dust or something which just made it blend in). Now you can push on the depression which triggers a button or a latch or something. Now when you "push" the rock face it opens up like a door.

Now take this same puzzle in a P&C designed game: There's a rock face on a giant boulder. Clicking the eye on it yields no information. Clicking the hand icon on it hoping that somehow interacting physically with the object will reveal something gets you a message saying "You run your hand along the rock face-" (my wasn't that helpful! I didn't even have to think of that!) "-and discover a depression hidden under sand." Then when you use it again (or maybe you don't even have to, maybe it just assumes your automatic next action would be to do just that) instead of running your hand across it like before you actually push into the depression and then proceed to push open the rock door.

See the difference? In one instance actual thought is needed to go into the whole process. If you can't think of the correct interaction then you're basically stuck. In the other instance, you don't even need to think. All you need are your powers of observation and the ability to click on anything that you deem interesting and the game works it all out for you. It's like your best friend trying to get you into adventure games and telling you how to do everything, only he's not needed. The game does it for you.

A bit of a stretch? Maybe, but the principle still holds. A parser definitely does require more thought to go into the process. Many P&C defenders would then say "No, it just turns into a game of discovering what exactly the developer wants you to type" and to that I say....uh, YEAH. That's the whole point. Discover how to correctly solve the puzzle the way the designer intended. Sometimes he doesn't think of all the possible ways you could type in the needed interaction (or sometimes it's literally impossible to program some exact phrases people think of, because of the way Sierra's parser systems worked) but they are few and far between and there was plenty of room for improvement in that regard that could have been done, had P&C not taken over.

The fact remains, much more is possible and can require much more actual thought rather than a mild experimental click fest to solve puzzles when you have a parser ready to type in any command necessary. In SQ4's instance, adding tongue and nose icons don't do anything for the game other than give you an excuse to find some funny narrator dialogue because they aren't needed AT ALL in the game (same goes for SQ1VGA). Neither provide any further access of interaction with objects. There's just the hand which is all-encompassing. And the same goes for all the other icons.

Of course the exact opposite is true when dealing with games with a simplistic interface like KQ7 or MOE.  If point and click is technically dumbed down, those games are full on lobotomies.  That did affect the gameplay and there was a ton of stuff I would have loved to look at and explore in KQ7 that I just couldn't.  Because the KQ7 interface, to borrow a phrase I read somewhere, sucked the sweat off of Lamb's nuts.

Given the example I just gave above, you could say the same for a 4-7 icon interface game when compared to a parser game.

A possible solution would be to just provide more interactions but this proves cumbersome. Take the early LA adventures. Eventually they just dumbed it down from 16 possible interactions to 9 to 5 to 4 to 3 to 1. Why? Because the players would see the entire list of interactions and feel the task is too daunting and too hard. To be fair, a parser in and of itself can be even more daunting to some people but at least it doesn't list a whole long line of interactions that can look even worse.

But there's problems with everything. Take Sierra's general VGA P&C interface. It doesn't really give away too much just by the list of interactions, but it doesn't make it terribly hard either. Just use everything you can and there you go. Inventory items help but that just prolongs it out and makes it a boring click-everything-on-everything-and-eachother fest in some instances. Telltale have removed the option of using inventory items on eachother altogether to remove that problem (except for TMI). Some games (like Return to Zork) offer more interactions than just "use" when right clicking on them, but the very reason that they all get displayed on the screen gives away everything you could possibly do for them. True, this removes the problem of trying to guess what the designer wants you to do with them, but it also makes it that much easier. Which is why I defend the "guess what the developer wants" design aspect for the simple reason that it makes you think. And that's the whole point of an adventure.

Like I said, I like them both, but parsers do have a lot going for them and provide much more freedom for experimentation than P&C does. I think with the way the industry was going (and still is) is that people don't want to experiment and take their time with games anymore. They just want to beat it so they can feel like they accomplished something. And that is a whole different topic altogether.

I do hate the term "Point & Click game", though, which everybody calls adventure games. Makes it sound like a baby's toy. Another thing the interface ruined for adventures.

[Cez]

I beg to differ here. We aren't just talking about utilizing each of the 5 senses when it comes to interaction. In fact there are many different ways to interact with objects with each of the senses in themselves. Take the hand icon. How do you want to use it with the object? Stroke it to feel its texture? Try to pick it up? Attempt to push or pull it? Jump on it? Kick it? Throw it? There are many ways to "use" an object and that's just for the hand icon. What about talking? Do you want to speak to it? Whistle to it? Blow on it? If a game were designed around that kind of freedom in interactions then a whole new world of puzzle design opportunities (some more complex than simply "use") open up to the designer.

And then, I might just play scramble

[MusicallyInspired]

I assume you meant Scrabble?   I rest my case. Unless you're talking about the 1981 horizontal scrolling shoot-em-up?

[Cez]

No, I really meant scrabble --it was late last night.

Funny, I didn't get a "I don't understand scramble" or "what do you want do with it?"

[Damar]

While the example of the boulder is good, and shows how a parser is more subtle, I don't think the parser was ever used to that extreme.  Particularly with the earlier games many verbs would be good enough to accomplish what you wanted.  For example, using the boulder with the hidden latch, the ideal command would be "feel boulder" which would reveal it.  However it's just as likely, particularly with earlier games that any number of "hand" commands would give the same result.  "Feel boulder," "Touch boulder," "Push boulder," even "Climb boulder," or "Kick boulder" could have the same result because the writer of the game doesn't want to code a bunch of different responses and figures any physical act on the boulder is enough to justify a "you find a hidden latch" response.  Or as a more specific example, in KQIII, the commands "put hand in hole" and "put head in hole" both yield the same result, namely a rope ladder falling from above.  Those are completely different commands with completely different implications (careful exploration versus brazen action) yet they'll result in the same action.  And that's what I meant about the illusion of being able to do more.  Sure you can type in all those things, but in the end, the variety still gives way to the basic senses, like the icons do.  And that's if the game even understands what you want to say.  More often than not you get a "I don't understand" message.  And that's the issue.  A lot of the expansiveness of the parser system that's being discussed just never seemed to exist in that way.  Potentially it could have, if it had been expanded instead of abandoned for and icon based point and click interface, but I've never played a game with a completely expansive, ask whatever you want, parser.  Anytime you try to play that way, you'll get a myriad of "I don't understand" messages or clusters of commands that will do the exact same thing (whether you meant them to or not.)

Like I said, in the end, I do prefer parsers.  That said, I don't think that a well done point and click interface completely dumbs things down and it doesn't make me feel (again, if done well) that I'm missing out by not being able to type in whatever I want.

[MusicallyInspired]

Yes, I agree P&C doesn't dumb things down incredibly and I do like them, but it does dumb it down nevertheless. And while not all developers like to program in responses for everything you could think of to type, many do. Scott did. As did Josh Mandel. Actually to be fair, Josh's shtick was just writing the funny responses so that's not really from a puzzle design standpoint, but more of the smell/taste icon territory which are useless to solving puzzles.

Either way, good points. The whole point of my post was not necessarily to say that parser is better, but to say that there is more potential in designing a game with a parser than there ever will be with a P&C system.

[DawsonJ]

Personally, I'm fine with Parser- or P&C-based in terfaces. However, parsers were often mated to HORRIBLE graphics, meaning that typing "Look" generically on a screen didn't always give an explanation of singular items. So, in reality, you see a 2 pixel by 2 pixel brown thing on the ground, sans description, and have to type in every noun you can think of. Eventually, you get frustrated and have to read a FAQ. You end up typing everything from "Pick up nut" to "Pick up turd" to no avail. Then, you read, in a FAQ, that it's a walnut and can't believe that the programmers only allow "Walnut", thus excluding the word "Nut"! That's the "Read the Programmers' Minds" crap that's been previously mentioned. Due to the limited graphics available in the early games, we as players paid dearly at times for lack of better screen descriptions and usable nouns in parser vocabulary.