Two-Click Interface - which way round?!

[Thaumaturge]

I think that one thing is being missed regarding my suggested dynamic verb menu UI: that UI only shows up when there are multiple actions available on a given object. In cases in which only one action is available, it falls back to a one-click system, albeit one that displays an icon for the action that will be taken (which should, I think, at the least reduce the ambiguity that has been mentioned as present in some one-click systems).

First, what happens if you click on something closer to the edges of the screen?

The simplest solution, I think, would be to place limits on the position of the interface to prevent buttons from being placed off-screen; it seems as though it would be simple enough to prevent this from causing the buttons from overlapping the object, perhaps by placing the UI next to (or above or below, depending on which screen edge was involved) the object in that case (with the animation again helping to maintain the connection between the two).

Second, if I understand you correctly, you have to click inside the button in order to activate the action? So that requires a large mouse movement to hit a small target. According to Fitt's Law, this will be a slow action, and it requires both physical effort and concentration. That delay and that effort adds up for every action, and makes the interface tedious to use over time.

First, as mentioned above, the radial menu appears only for objects that have multiple actions, and which -- going with what you said earlier -- haven't been deemed to be better served by a different interface (such as your example of turning a wheel). Thus it (hopefully) wouldn't become tiresome. I also think that you overestimate the slowness, effort and concentration involved -- although I'll admit that it may be wiser to risk overestimating such things than underestimating them.

As to the size of the buttons, their hot-spots could perhaps be made bigger than their apparent size to allow for imprecise clicks on the part of the player.

A relatively minor, fixable problem is that there's no indication of what you clicked on ...

True, but that's an oversight in my example, rather than a flaw in the interface, I think: I was perhaps concentrating on portraying the radial menu, at the expense of thinking about other elements of the interface.

What I'm arguing is that the object-verb order of commands in the verbcoin paradigm tends to put more attention on the list of commands that are available, and especially so if the commands are dynamic. ...

Hmm... I think that this may be a matter of individual response, because I think that my own experience in playing adventure games has been to the contrary: to me, dynamic verbs make the world feel more reactive, more open to varied interactions, than with other interfaces.

A good example, I think, might be the interface used by Gabriel Knight 3 (which used a list rather than a radial menu, and didn't include animations). It remains one of my favourite interfaces, I don't recall it ever becoming tedious, don't think that I ever felt that the buttons were too small and feel that the inclusion of dynamic verbs actually made the game world seem more open than most games as a result of providing options not usually available.

You don't need a verb coin for this.

Smell: Look at food (right click)
Eat: Use fork on food
Pick up: Use doggy bag on food

Hmm... To my mind, the inventory item solution feels clunky: it clutters my inventory with objects that may have no other use, and means adding yet another for each such action. If there's a bowl of soup as well we add a spoon, and so on as other actions on other objects become available.

To me the dynamic list feels more elegant in this situation.

Further, how would you apply that concept to allowing the player the options both to sit in a chair (so as to do some writing, for example) and to move the chair (to get something from a high shelf, perhaps)? I suppose that you could make the action dependant on the time, or game events (such as having the character move the chair when it becomes relevant, and then not again), but that doesn't seem to be easily communicated to the player, or you could introduce another object to do the work of the chair when moved, but then you're bending the game around the interface.

If you mean having a separate hot-spot for the fork, then I think that it's all too easy to miss that they're separate hot-spots (even with tooltips, a quick mouse-movement can easily jump over the "plate" bit unless the developer has bothered to include a close-up of the plate), making it easy to miss that both actions are available.

Or alternatively (depending on how the food is going to be used in the game)
Eat: Click on food
Pick up: Click on plate

As a player, I'm not a huge fan of this, myself: as mentioned above, it can be all too easy to miss a small, nearby but separate hot-spot, I feel.

LucasArts had particular letters for particular verbs. From memory:
...

Wow, to my eye those LucasArts keys look horribly unintuitive.

Nevertheless, thank you -- while on reflection I'm familiar with such shortcuts as "tab" for the inventory, and do now remember a few games in which I've used shortcuts for one or two other things, I don't think that I knew about either the Sierra or LucasArts shortcuts that you mentioned.

(Still regarding shortcuts)

You should definitely consider them. ...

You make some good points; I'll hopefully try to keep shortcuts in mind -- thank you. ^_^

I don't think that I agree that a verb coin necessarily obscures a significant portion of the object being interacted with, since I think that the backing "coin" can be left out -- and that doing so may actually improve the interface. Something like this:

Technically true, but that doesn't solve the other issues with verb coins, i.e. that extra actions are needed to perform a simple task. The player now has to (1) click, (2) wait for the animation, (3) move the mouse to the right icon, and (4) click again. Frankly, an interface like this would make me quit the game within minutes.

Really? To my mind it seems less onerous than most other UIs, excluding only one-click (which limits the available number of actions or calls for additional UIs or potentially-shoehorned inventory items or conversations. Bear in mind that the animation needn't be long -- 0.1 seconds, perhaps 0.2, I would imagine. However, this could easily be a matter of personal preference.

Again, the interface used by Gabriel Knight 3 (as described above) stands out in my memory as one of my favourite adventure game interfaces, as mentioned above. Similarly, I recall very much enjoying the radial menus used in Neverwinter Nights. (In fact, as I recall, those two games were my inspirations for this UI.)

[Vince Twelve]

If the game designer tries to make various responses/animations to most hotspots for almost every verb, adds hotkeys to each verb (walk, talk, interact, inventory)...then is a Sierra-Style interface "acceptable"?

Yes.  They're all acceptable.  But is it the best choice?  What I advocate is that the "various responses/animations" should be meaningful in order to justify the interface. 

If I click "Talk to" + "Brick wall", it doesn't matter to me if the player character says "I can't talk to that." or if he walks over to the wall and attempts, humorously, to engage in conversation with it while gesturing wildly.  Neither of those responses are meaningful (though one is more entertaining than the other) and they don't justify the interface. 

To justify the multi-verb interface you must satisfy at least one of the following two conditions on at least one puzzle. Obviously, doing these on many puzzles is preferable.

• At least two of the verbs must have different meaningful responses on a single item.  For example you can "Push" a heavy chest to block a doorway when someone is chasing you, but you can also "Open" the heavy chest to find a useful item inside.
• A verb must be used in an unexpected or non-intuitive situation.  For example, you have to "Talk to" that brick wall because it's actually a magical brick wall that can talk, and figuring that out is part of the puzzle. "Talking to" NPCs or "Interacting with" light switches does not satisfy this condition.

I think that if your game doesn't do at least one of these two things, and maybe even if it does these things but doesn't do them enough, you probably would have been better off with a simpler interface.

(Exceptions allowed for games that just have really funny responses to multiple verbs, even if they're not actually meaningful.)

[Radiant]

First, as mentioned above, the radial menu appears only for objects that have multiple actions, and which -- going with what you said earlier -- haven't been deemed to be better served by a different interface (such as your example of turning a wheel). Thus it (hopefully) wouldn't become tiresome. I also think that you overestimate the slowness, effort and concentration involved -- although I'll admit that it may be wiser to risk overestimating such things than underestimating them.

I think this underlines the fault in your reasoning. An interface is not there to look cool or be pretty. An interface is there to seamlessly let the user interact with the game. The question of how much concentration is needed, the issue is that any concentration is needed in the first place. The player should be concentrating on the story you're trying to tell, not on getting the interface to work.

If in this thread here, five or six people suggest not to use something (in this case a verbcoin) you'd better believe there's five or six thousand people outside this thread but that also won't play your game if it has a verbcoin, so using a verbcoin would just cut a huge chunk away from your target audience. After all, there are tens of thousands of free games on the web, so if your game has an interface that they don't like or don't understand, they'll go play the next one. Sorry if that sounds harsh, but that's the way it works.

[Trapezoid]

To justify the multi-verb interface you must satisfy at least one of the following two conditions on at least one puzzle. Obviously, doing these on many puzzles is preferable.

• At least two of the verbs must have different meaningful responses on a single item.  For example you can "Push" a heavy chest to block a doorway when someone is chasing you, but you can also "Open" the heavy chest to find a useful item inside.
• A verb must be used in an unexpected or non-intuitive situation.  For example, you have to "Talk to" that brick wall because it's actually a magical brick wall that can talk, and figuring that out is part of the puzzle. "Talking to" NPCs or "Interacting with" light switches does not satisfy this condition.

Yes. It's hard to steer away from game design on this topic, because the biggest difference between one-click and multi-verb is the potential for interactivity. One-click tends to become a "click on everything" game. Multi-verb doesn't automatically avoid this problem, either, but if you flesh out the game environment well enough, every action/hotspot combination becomes a Schroedinger's Cat of meaningfulness. You want to encourage the player to think "what if I..." instead of merely clicking everything. Maybe I was getting a little crazy with my suggestion of making an animation for every action, but my point was that having multiple verbs is only preferable when they don't feel extraneous.

I think DOTT is a great example of this. First thing you do in the game is open the clock. In order to discover the secret passage, you have to wonder if the clock is openable. A little specificity helps put you in the game. That game's loaded with objects you need to push and pull and open and close. (And, incidentally, most of the time when you try to push/pull/open/close something, the protagonist actually walks up to it and *touches it* before saying "I can't move it." This simple gesture helps reinforce that there ARE things that can be moved, and that the protagonist won't automatically shoot down all of your ideas.)

[qptain Nemo]

It sounds like you've completely misunderstood what I was saying, as demonstrated by the fact that I value and appreciate all those games you list, and personally really enjoy several of them. (I've praised The Vacuum as one of my favorite AGS games numerous times.)

I'm sorry, I stand corrected then. Rather merrily, in fact, because I'm glad to hear you appreciate those games and The Vacuum in particular. Let's clarify it then!

But Trapezoid wasn't talking about alternative gameplay elements. He was talking of interactivity that doesn't tie in to the gameplay at all, that doesn't accomplish anything:

If the player wants to push a Coke machine, why not have a generic shoving animation, even if it doesn't accomplish anything? [...]
I'm not saying you need to program an animation for every possible action, but the game needs to reward actions whether or not they're part of the solution, at least sometimes.

Fair enough. However, I'd like to note in the part you're quoting he says "reward actions whether or not they're part of the solution". I'm not sure it's fair to equate "reward" to "not acomplishing anything".

I'm saying I think the qualities of "open world" games are largely incompatible with the qualities of adventure games, and that you need to strike a (fairly conservative) balance when it comes to "pointless interactivity": while a little bit is great, adding more and more random, non-goal-oriented crap to do in an adventure game pretty quickly just dilutes the actual game, making it worse.

Can't say I agree at all. Again, you perceive it as "non-goal-oriented crap", and I perceive it as precious non-goal-oriented bits. I don't remember ever being dissatisfied with an adventure games for having too much interactivity. Not in Legend entertaiment's games, not in Edna and Harvey. You call puzzle-solving "the actual game" and it makes me wonder why would interactivity for the sake of interactivity or for the sake of immersion, or even just for fun not be a part of the "actual game". Let alone that it can as well actually severely affect the game's world and the characters and the story, and that many open-world games are actually terribly goal-oriented (there's nothing really to do in Skyrim other than just to perform quests or randomly kill everything you see). The only reason I can find is viewing games and adventure games in particular through challenge-centric priorities, where challenges are virtually the only thing that really matters.

And I can understand different priorities, and how they make certain things less appealing, but I don't really see how optional interactivity can ruin the experience for somebody who doesn't care for it much. It will if it makes puzzles much harder simply merely by its presence, by increasing the numbers of things you can try, but that in turn will mean that the puzzles are illogical and rely too much on guessing correctly and bruteforcing through all available actions, and that's bad design anyway in my book. And if you ask me, that kind of thing plagues adventure games design way more than excessive interactivity.

So, in other words, I can't relate/understand why you say it will make adventure games worse. I have noted that you're proposing balance not eradication though. Also I suppose you wouldn't treat interactivity that affects the world and the game state significantly even if it's not goal-oriented per se the same way as interactivity that is "pointless" and mostly superficial? Or? What about interactivity that has a point but its only point is to increase immersion? Again, I remember you said "balance" but it's hard to guess where you draw the line when you say "non-goal-oriented crap"...

I didn't label it pointless, Trapezoid did, by saying "it accomplishes nothing" and "isn't part of the solution."

First of all, "isn't part of the solution" isn't exactly the same as "pointless" unless puzzles are the only thing that matter at all in adventure games. But even then, you're still claiming it's not a part of the gameplay. I disagree. A lot of fun of playing well made text adventure games is being able to screw around. It can yield great fun. But if you're excluding this sort of thing out of the gameplay, you're making the enjoyment of that sound akin to playing the main menu of Doom.

I agree that puzzle-solving is one of the core components of adventure games and their gameplay. But I don't think that it's all there is to them. I love cozy utterly linear puzzle-centric adventure games, otherwise I wouldn't be praising the hell out of Revolution, Daedalic and Pendulo for excelling at exactly that. But I don't agree with labelling any deviations as (potentially) bad design.

Again, no one is saying that! If you think having certain actions in the game will be fun, by all means put them in! But then think carefully about the most effortless, least intrusive, least complicated and easiest to learn UI you can design to carry out those actions (I include in this concept of UI the possibility of accessing the actions through special inventory items, dialog options, etc.): and the answer will, in my opinion, almost never be a verb coin.

I have a very hard time imagining how the hell can you make selecting an item from your arbitrarily-sized inventory significantly easier to perform than selecting an additional action from whatever, assuming both implemented as reasonable as possible. And "Verb coin" is a dangerously ambiguous term I feel. Those things that you have to summon by holding the mouse button in are abominations. But if you just click on a thing and a small menu (possibly consisting of icons)appears nearby, is that a verb coin? Or is it just a context menu?

And actually come to think of it, why is selecting from an inventory/dialogue window somehow so drastically different than selecting from an instant verb coin/context menu? You choose an item from a selection of items. It's the same process?..

Ummm... why? It's not like there was any gameplay to it: you couldn't jump, you couldn't fall, it wasn't really interactive since you couldn't really do anything; mostly the levels were so linear you were simply walking forward from Point A to Point B (very little exploring), and apart from a couple of extremely poorly implemented chase/stealth sequences, there was no challenge or skill to it. It was only a chore you had to perform in order to play the game.

I remember one level as literally consisting solely of walking up a hill, without any opportunity for interaction.

Again, I'm not saying walking/moving can not be a fun thing, even in an adventure game. That's what Journey was all about, right? But it's something you have to design for. In Dreamfall, it seemed like they'd only thought "Oh, it's 3D so we have to do direct control. Hey, people love Tomb Raider so they'll love this, right?" without given any deeper thought to why moving around is fun in Tomb Raider.

Ah, and yet again we return to what "gameplay" is. Yes, I'd definitely prefer if Dreamfall was a billion times more interactive. However, it still had a really nicely realized world, very pretty and quite immersive. And since I was immersed, the mere act of walking through a nice world and looking at some things was a pleasant experience. If it allowed for more interactivity and exploration, walking in it would be tremendously more pleasant, as well as if it was enhanced some way as you say. But even as it was, it was pleasant to act on being immersed even by just walking. It's a chore, if you think about it as of a task that you have to accomplish, but if you think of it as walking through nice places, it's pleasant.

So to sum it up, how I see it, you really care about the distinction between "pointless" and "non-pointless" interactivity (and maybe further between "non-pointless" and "goal-oriented"?). And I, frankly, don't. I want as much interactivity as I can get. But of course the more consequences it bears, the sweeter it is. It doesn't mean it has to be in the form of adventure games, true, but I'm still to see a compelling argument as of why would it spoil and taint adventure games so badly to have the level of interactivity they once had in their finest examples (again, I point you to Legend entertainment, whose best games in my opinion dwarf most of Sierra's and Lucas' offerings in terms of quality of adventure gameplay as well as content) or maybe even more. And you still didn't say what fate do you propose for the "non-standard adventure games".

[Thaumaturge]

Second, if I understand you correctly, you have to click inside the button in order to activate the action? So that requires a large mouse movement to hit a small target. According to Fitt's Law, this will be a slow action, and it requires both physical effort and concentration. That delay and that effort adds up for every action, and makes the interface tedious to use over time.

On reflection, I think that you may actually have a point here. I think that I see a way to fix the problem, but I fear that it won't work well near the screen edges. (I do think that it would work well in first-person adventure games, however.)

So, what about a linear list of dynamic verbs, likely as icons with tooltips? It seems to have worked for Gabriel Knight 3: for all that's been written about that game, for or against, I don't seem to recall ever seeing criticism of the verb list. It should also be very easy to prevent the list from either covering the hot-spot or placing buttons outside of the screen, and in terms of hitting the buttons one need only navigate to them once, after which moving along the list should be fairly simple.

I think this underlines the fault in your reasoning. An interface is not there to look cool or be pretty. An interface is there to seamlessly let the user interact with the game. The question of how much concentration is needed, the issue is that any concentration is needed in the first place. The player should be concentrating on the story you're trying to tell, not on getting the interface to work.

I'm not convinced that there's likely to be any interface that requires no concentration; it's then just a matter of how much is called for. Even a single-click interface calls for enough concentration to hit the (sometimes relatively small) hot-spots in the world, which I believe that Snarky pointed out as a potential problem for the radial menu.

If in this thread here, five or six people suggest not to use something (in this case a verbcoin) you'd better believe there's five or six thousand people outside this thread but that also won't play your game if it has a verbcoin ...

On what do you base this? I sincerely doubt that the posters in this thread make a terribly representative sample (it seems to me that it's likely skewed by a handful of elements), and as a sample it's far too small to likely be useful; if you're suggesting that I defer to expertise then the number of people responding doesn't seem terribly relevant. The numbers that you posted may seem scary, but I don't currently see a basis for them.

Those arguments aside, are you opposed to the radial-menu format specifically, or any form of dynamic verb list? If the former, see my new response to Snarky above: I've conceded that a radial menu may well be poorly-suited to a third-person adventure.

[Radiant]

Overly small hotspots are also a flaw in game design, yes.

Anyway, I object to this kind of radial menu. I'm not principally opposed to dynamic verb lists, but I do think verb coins of any kind are not a suitable interface by the standards of this decade.

[TheBitPriest]

Technically true, but that doesn't solve the other issues with verb coins, i.e. that extra actions are needed to perform a simple task. The player now has to (1) click, (2) wait for the animation, (3) move the mouse to the right icon, and (4) click again. Frankly, an interface like this would make me quit the game within minutes.

I think there could be an alternative:
- Short click: default verb (open door, pickup item, look at for RMB);
- Long click: display additional verbs.
If long click is hard to use (to determine a time period needed), it may be "hold and drag to the side" - which opens menu at the place defined by drag direction. Or just double-click.

I like this.  This option allows the convenience of giving the player a default interaction on the one button/one touch/LMB, and it also allows for more complex contemplative interaction.  While the verb-coin circle (mentioned in other posts above) is creative, it's been done many times before, and I don't think it's ever taken off for many of the reasons cited.  My vote:  It just gets in the way.  To implement Crimson's long-click idea, I think I'd make a pop-up/slide-in verb coin bar on one of the the edges of the screen with a limited amount of useful (and used!) verbs.   

[Radiant]

I don't like it; having a difference between short-clicking and long-clicking is very unintuitive. It's basically a verbcoin that only displays if you hold the mouse button down long enough.

[Babar]

Okay, with all these restrictions created to keep in mind the touchscreen mobile user, I'm curious...aside from single button (with not even mouse-hover text possible, it seems?), what would be an appropriate mobile device interface? I've never had one, so I'm a bit in the dark here. Andail mentioned something about two-finger touch simulating right-click?

[Knox]

I think there could be an alternative:
- Short click: default verb (open door, pickup item, look at for RMB);
- Long click: display additional verbs.
If long click is hard to use (to determine a time period needed), it may be "hold and drag to the side" - which opens menu at the place defined by drag direction. Or just double-click.

I like this idea too. However I'd let the player choose how they want the additional verbs to be displayed in the options (double-click, long-click for "x" amount of time, etc). If you double-click on a hotspot or hold down of "x" amount of time, the additional verbs could be displayed as a semi-transparent list or just alpha text labels, as small as possible so it wont bug the player too much.

Man, now Im thinking of going back and offering 3 modes to choose from: Sierra classic, verb coin and 2-click.

[qptain Nemo]

Anyway, I object to this kind of radial menu. I'm not principally opposed to dynamic verb lists, but I do think verb coins of any kind are not a suitable interface by the standards of this decade.

Why is a dynamic verb menu appearing nearby the point of interaction so much worse than a dynamic menu appearing elsewhere in a fixed place? I sort of prefer the latter too, but still.

[Radiant]

Why is a dynamic verb menu appearing nearby the point of interaction so much worse than a dynamic menu appearing elsewhere in a fixed place? I sort of prefer the latter too, but still.

It's not about where the menu appears, but about how many actions you must take in sequence in order to activate the verb.

[qptain Nemo]

But I don't see how it's possible to decrease the said amount of actions for dynamic verbs to less than 2. You need to select the subject and then you need to select the verb. You could like make it so it's enough to hover over the hotspot and then you can select the verbs with WASD or the mousewheel, and I like that idea, but it doesn't really appear all that radically different from a verb coin that acts like a graphical context menu.

[Radiant]

Well, for both a one-verb game and a "left to use, right to look", it is one. And for a verbcoin it's three actions, which is more than necessary.

[qptain Nemo]

Well, for both a one-verb game and a "left to use, right to look", it is one. And for a verbcoin it's three actions, which is more than necessary.

What? So basically you say you're not against dynamic verbs, just against the verbcoin, and then I ask you well, how would you implement the dynamic verbs, and you respond by saying well, you just don't? Errr...

[Radiant]

I don't see you asking how to implement dynamic verbs, I see you asking why a verbcoin is worse than a menu in a fixed place, and then asking how a game can require less than two user actions to select a verb. Hence my response.

[qptain Nemo]

Ok. So let me clarify: do you see any way to make selecting a verb from a dynamic list any easier than with two actions or my "one and a half action" proposal?
And to be fair, I did specify I was talking about dynamic verbs. I said "I don't see how it's possible to decrease the said amount of actions for dynamic verbs to less than 2".

[kaput]

Wouldn't verb coins work pretty well on a tablet? 'Click on hotspot to open verb coin and click again to choose your action'. I get the impression it wouldn't be so bad. I do like the BASS style system, myself, but I really don't see all that much flaws with a verb coin.

I get the whole 'let's make things easier for the player thing', but honestly, I'd hate to say the day where adventure games lose the sense of an open world and all the interactive possibilities become almost 'verboten'. I get the impression that many feel that the player should just look at the screen and it say "Well done, completed the game!".

Whilst I can see casual players preferring the two click system, I can also see the traditionalists liking more possibilities. I guess it's just finding a common ground?

On the case of verbs in general, I do find 'verb lists' a little more fidgety.

[qptain Nemo]

I guess it's just finding a common ground?

I'm not sure it's what we need. There are clearly (at least) two schools of thought here: many possible actions vs "left to use, right to look". And both have their merits but they are hardly combinable. I think it'd only enrich the genre and gaming in general if both would remain and prosper, independent of each other and yet united.

On the case of verbs in general, I do find 'verb lists' a little more fidgety.

But what is the difference between a dynamic verb coin and a dynamic verb list of the same size as you see it? I feel it's essentially the same thing.