The "Look at" paradox.

[LUniqueDan]

Yeah, I surrender. I implemented a "look at" button instead of "read" to my main gui. And I never felt myself so boringly redundant : There's a desk in the screen. Over the desk, the mouse will tell the player that this is a desk, and if the player look at the desk what I'm I suppose to say?

- "It's a desk - just open it for further info"?
- "It's a dictionnary definition of a desk
- put a joke about desk
- Make an historical background of the desk, even if it's useless for the story and kill the beat?



I know, this is a very common feature nowaday. Not to mention, a lot of recent games who have more than 1 option, feature olny 2 cursors/clicks/verbs :

Look at / Do everything else in the world

Back in time, in text adventure, this was necessary. But now? I understand that some sleuth stories need it. Or game sets in totally weird gameworld like The Dig. But is it a little superfactory ?

But here's the point : is it a little paradoxal that graphical adventure games all have "look at" feature? Mostly when it's linked to descriptive label telling what's under the mouse?

Discuss.

[Radiant]

Well, this is why I tend to use only one action in most of my games. There isn't necessarily a need for a separate "look" and "use" descriptions for every single hotspot.

I find this very elegant:
* Left-click on ground: walk there
* Left-click on hotspot: interact with it
* Left-click on character: talk to it
* Left-click on yourself: open inventory

[LUniqueDan]

 This is very elegant indeed.

That means that you specially design the game in order to need olny 1 (+use inventory at) interaction by object/thingie/gizmo.

D'you have any insight of where you need to overturn that to avoid a problem? And how d'you solve the situation.

[Crimson Wizard]

You could script a function that makes player say "I see nothing special." and call that function for every self-explanatory item in your game.

Other than that - it may be a stupidly mannered line, something like "the desk's shape mysteriously looms in the dimmed light of the room..."

[Crimson Wizard]

Sorry for double post.

I do not know if you may consider changing your controls on current point of development; but if there's actually a problem of some verbs not much used, I may suggest this: swap the order of commands - make object go first and verb second. I.e. player selects an object and a Gui appears with only those verbs shown that are truly useful. This will also allow you to introduce specific verbs only usable at limited number of occasions.
But, ofcourse, that will require more difficult scripting.

[LUniqueDan]

[AGSer pride] Don't worry, Crimson, my unhandled events are sets. [/AGSer pride]  But even with random "Nothing important about that" variation it makes me sad. Maybe I'm too stuck in the Zak McKracken / Maniac Mansion pattern where if the verb apply, it must be working.



In other hand, I'm trying very hard to avoid   this stupid post by PetuniaCon1413 where she wants Zak to answer "Use Bed: This is no time for sleeping!  It's the middle of the day!" while in the original version the player can actually USE the bed and back to sleep.

[Crimson Wizard]

Heh, posted the same time...
There was another suggestion I made in the second post above, but I now see that you are making the Maniac Mansion style game? Then, unfortunately, it cannot be used. Unless you rework the game style

[Trapezoid]

Even relatively mundane 'look at' responses add detail to the environment and the protagonist. "Looks like mahogany." or "Reminds me of the one my grandfather had."

One of the fine things about adventure games is the act of exploring every nook and cranny of the gameworld. Use your imagination and give everything a base level of detail and interaction, and it will go a long way to creating a rich experience for the player.

[Radiant]

D'you have any insight of where you need to overturn that to avoid a problem? And how d'you solve the situation.

Stock responses.

The only reason LucasArts can get away with so many verbs is to use stock responses. For example, trying "Look At <foo>" gives the stock response "Nice <foo>." for over half of the hotspots in Monkey Island II. In Maniac Mansion, most hotspots respond with e.g. "There's nothing to read on it", "It doesn't seem to open", and so forth.

To cut your development to a reasonable timeframe, you'll need one of (1) few hotspots, (2) few verbs, (3) many stock responses, or (4) more spare time than most people.

[GarageGothic]

I can think of a couple of different solutions, depending on the situation and object in question:

• Reduce the number of hotspots. If it's not relevant enough for the player to even comment on, should it be a hotspot in the first place? To me it seems that the classic silly responses of the Look-At command are remnants of an era (namely the early Sierra point-n-click games) before cursor highlighting became standard, and pretty much anything on the screen could be clicked on.
  
  
• If it's possible to interact with the object, it should be possible to examine it first - very few games have a high enough resolution, or descriptive enough hotspot names, to make it perfectly clear what you're interacting with and why. The Look-at text should at least provide the motivation for the player to want to interact with the item, and possible even hint at its use.
  
  
• Make Look-At a useful and necessary feature, expect that the player uses it. Some times it could lead to close-ups of items, but always offer useful information - e.g. add details to the story or a sub-plot, help establish the game world, offer some insight into the NPC who owns the item, or express the personality of the player character. I thought Gemini Rue was a brilliant example of how to build atmosphere and depth through Look-At comments.

[Wyz]

Yes I feel that the 'Look at' verb is indeed a paradox. You already see the image on the screen, and the mouse over will tell you what it is. In fact your mouse cursor can be seen as the focal point of the character, this is what he will look at.

So, lets get rid of the lookat verb?
I say no!

In my perception it actually means 'Examine'. When looking at said desk you might just see it's basic structure, it has drawers, a desktop, just a desk but at closer inspection you might notice there is a small blood stain on it, or there is a small compartment that is hidden to the naked eye. If the desk is simply a desk and nothing more well then you could do one of three things: make a joke, explicitly describe it or use a stock answer.
Some would say, if it adds nothing to the game just leave the hotspots out but personally I'm a fan of a lot of hotspots. It gives my curiosity way, and I really enjoy reading the descriptions. Besides, it is not entirely without influence on the gameplay:

• Sometimes a certain object in a room draws attention but when examining it it turns out it is nothing special, this is a sort of micro red haring; it wakes you and makes you think that what you're looking for might in fact be in a less obvious place. It is mind games like this that make descriptions awesome.
  
• In other occasions it might tell us more about the setting of the game or reveal something about characters, even the playing character. In this case the descriptions can be totally optionally, and people who want to spend a lot of time to get really deep into the story can read them, whilst other can simple skip over it.
  
  Obviously the two approaches do not mix in one game; that's where games usually fail and people get frustrated. People who don't want to read all descriptions but have to because there might be a hint needed in one of them. Well if the game does this consistently I think it might actually work just fine. Just change the verb from 'Look at' to 'Examine'
  
  The same goes for inventory, you don't snug something in your pocket without having a look, but the 'Examine'verb will do a closer inspection: hidden compartments, actually read something, notice certain details that give a hint or notice that something is not what it seems to be.
  
  A bit much, but I've actually put a lot of thought to the subject lately so it needed out.

[Eggie]

For the game I'm working on I've basically taken it to mean 'be inspired to spout off a hint by' or 'make a tortured joke about'.

A bit of a leap, maybe. But it sure keeps things more fun than a dude saying 'there's nothing special here' or 'It's a desk!'.

Honestly I'm slowly coming around to the views in the VinceTwelve article that less can be more in terms of verbs and actually kind of have been since the very first Telltale Sam and Max episode turned out not to suck.

[theo]

Frankly, Point n click games in general could do with less forms of interaction. Generally there are way too many ways to interact with items in point and click games, which in my opinion truly breaks immersion and only serves the purpose to frustrate and confuse the player. I see now in the vincetwelve article that me and vince seem to share the same thoughts here. Though perhaps I would argue that it can be taken one step further. To me, once I opened up the box of "left click does almost everything", I figured I might as well go all the way and call it "left click does everything". It does wonders for flow and immersion.

That said, I do find the right click = "examine" a nice feature in many games, but only where it actually serves a real purpose and is used consistently. Generally 95% of hotspots won't benefit from this feature and thus as a whole the feature adds more confusion and badness (as the feedback will likely be "I see nothing special about it") than goodness to the overall game experience.

The desk is a good example of interaction redundancy. If the desk where in JD and I wanted the player to get a clue from its looks AND be able to open a drawer to look for contents, I'd either have two hotspots, one for the drawer and one for the desk surface, and have bwana open the drawer when clicking the draer, and talk about the surface when clicking the surface, or I'd go for my old JD trick of simply having a Game.DoOnceOnly on one single "desk" hotspot. This is a great way of making the first click be the desired "action" and the following clicks be an "examine", where this for gameplay or story reasons is necessary. I do it all over the place in JD and it works beautifully. This second solution should be used with caution though, specifically if the hotspot has an obvious "verb" use to it, but the character talks about it instead of doing the desired action - this will frustrate the hell out of the player.

In short: In general I'd rather have more hotspots than mouse modes. It's less obscuring and hence, better.

[Sughly]

Yeah this is a pretty interesting topic, and I like what everyone has to say about it. I particularly agree with Wyz and Trapezoid - I like the idea of not only having a lot of hotspots for numerous reasons, such as creating a fuller world and more points of exploration which in turn can help disguise the otherwise obvious puzzle design (ie this needs to be clicked to solve something), but also as Trapezoid said, just using basic interactions to try and tell something about the character or story. That was a poorly written sentence, too lazy to rewrite though 8)

Now with that being said, I do feel a bit torn since I also completely agree with where Theo is coming from. I found myself loving the one click format when playing games like Journey Down and Ben304's numerous iterations of it. The one advantage that only really comes to mind with verb choices is that, since they are mostly redundant, youre able to surprise the player with original uses of it, such as using the talk icon in Monkey Island to chew the gum or eat the biscuit. Not sure if thats much of an argument to make for it though

EDIT: man I dont even know what I'm talking about, really shouldnt respond to things like this when I'm tired. I talked about two different things there - hotspots to begin with and then mouse modes. *shrugs*

[cat]

I'm not sure I like the "many hotspots" on one thing (e.g. desk) approach. It can lead to serious pixel hunting. How should I know I have to move the mouse above different parts of the desk?

Also, a description can add a lot to the atmosphere. You can get details not only about the thing itself, but also about the PCs relation to this object and thus about the PC himself.

I think it was in 5DAS where you could look at the doors and you would get a different answer every time (my favourite being something like "I once dated a girl that had the same door")

[Shane 'ProgZmax' Stevens]

I like giving the player plenty of choice, but most authors don't really consider whether a particular action will have enough uses to justify it.  Take the smell and taste actions in the Space Quest games.  They were almost entirely worthless aside from some novelty/joke value, and these days something like that will just confuse the average player.  I do think there's room for separate actions like talk and interact, since you might want to have interact do something different from talk when clicking on an npc (like have you perform some action appropriate to the conditions, like punching or helping someone up).  I think context is more important than limiting your actions.  For instance, you might look at a the cover of a book while it's on a desk, but in your inventory you read it with the same action, so a better way to label the verb would be 'examine' since that covers the bases of both a casual look and a more detailed study.  If you use the right images/names for your actions you can get away with them doing more.  If you really push it, talk, interact, and examine could be used for everything you need while still giving you a good range of actions that could be performed on each entity.  You could talk to a person, interact with them (based on context or inventory) or examine them for a description.  With a book you could examine it for a description, interact with it to pick it up, or talk to it (pointless). 

Sometimes less is more, but sometimes it can be downright confusing, too.  It's important to keep that in mind when trying to boil down the actions/verbs to the basics.  I don't personally mind a game that uses multiple actions so long as they are not one-off uses and cannot be logically reduced without losing some key functionality (open and close can be easily combined into interact, but something like 'kick' or 'shoot' would need special distinction since they perform very specifically).

[Eggie]

Oh man, Progz just reminded me of something that's in direct conflict with my crumbling resistance to one-click.

EXAMINE and READ being two separate verbs is one of my favourite things about text adventures.

READ sign
"Trespassers will be prosecuted"

EXAMINE SIGN
"It's decorated with flowers, little smiley faces and a winking rabbit."

EXAMINE BOOK
"It's leather spine looks on the verge of crumbling to dust but some malevolent force seems to hold it together."

READ BOOK
"Ooh! Cupcake recipes!"

I also love OPENING and CLOSING books to switch between reading the cover and the contents. Omnomnomnomnomnominteraction.

[Radiant]

EXAMINE and READ being two separate verbs is one of my favourite things about text adventures.

Well, that usually works in graphic adventures, too. Eye icon = examine, hand icon = read.

[hedgefield]

Great thread.

I'm a big fan of simple control schemes, I abandoned the seperate walk/talk/fondle cursors a long time ago, but I'm not sure I'll give up on the look option yet, mainly for character development reasons. For every stupidly obvious desk interaction there is a wealth of LookAt interactions to be had with other things, like characters. If any click on a character initiates Talk, you miss out on the opportunity to have the player comment of that character. The way the protagonist describes him/her might be indicative of how they feel about that person. Plus if the purpose of a metal sign hanging somewhere is to be picked up so you can use it to wedge a door open, clicking on it will never tell you what the sign said (except with Theo's DoOnlyOnce method, which does sound very ingenious, though it relies on the fact that the player will think to click on the sign again). Besides, though probably few and far between there might be games that do not have a cursor overlay that already tells you what you're poking at.

I suppose it depends on what you are going for with your game. If it's a simple game, a one-button scheme might suffice. I remember I found it a breath of fresh air that talking to characters in Death on Stage involved no branching dialogue of any kind. So straight-forward clicking might make things easier to process. I did an action-based adventure one time with no dialogues or LookAts, and it didn't feel out of place. But at the same time I found some TellTale games very empty and bland because there were only two things to poke at in every room and that was it.

[EnterTheStory (aka tolworthy)]

you'll need (4) more spare time than most people.

That's the killer. I absolutely love the freedom of the old games like Zak and Maniac Mansion. But to do it for anything beyond the simplest game (or without a reasonable sized team of people) is just not practical.

I found that with my first four games. I allowed any character to interact with any object in any game (the games linked together). This meant millions of possible interactions. The only way to do it was to have very generic dialog, spend very little time on it, and ignore times when cutting corners gave odd results. The games suffered as a result: they generally looked unpolished. I've learned my lesson with the latest game (Monte Cristo): it has ONE hero and ONE kind of click. As a result I can spend far more time on the story, and produce a much better game.