f13.net

Atul Varma from Humanized (http://www.humanized.com) posted some good thoughts on the Humanized weblog (http://www.humanized.com/weblog/) about interface design in games. The post (http://www.humanized.com/weblog/2007/07/13/forging_the_seal_when_beating_the_game_is_beating/) discusses a recent play session of World of Warcraft, but can be applied to other MMOGs as well. He sums up his thoughts saying,


Good food for thought. Hell, even the buttons for the formatting of posts on this site require the use of tooltips to decipher at times. Of course, depending on the browser you use, the tooltip might just read, due to lack of alt tags, I guess. In any case, I thought it was an interesting read.


Permalink to article (http://www.humanized.com/weblog/2007/07/13/forging_the_seal_when_beating_the_game_is_beating/)

Easy to say UI suck.  Not so easy to fix.  But, of course, that's what they want you to pay them for.

I've been pondering about possible solutions to the UI problem. In MMOGs, the need for commands in the game is so vast that it is nigh on impossible to provide an interface that is not cryptic in some sense. Now, the aforementioned problem with determining which icon to click may have been an oversimplification of the problem for the sake of making a point, with a total disregard for the fact that there is a learning curve with any user interface. How steep that learning curve is is another issue, but there you are.

A user who frequently works within a given application will have a good working knowledge of the interface and will know which buttons to click and which menus store certain tasks and will probably even have their own hotkeys and macros created which is what brings me to my next point. Beyond developing an easy to learn and use interface, which is what Atul Varma is getting at, we need to design with the user in mind and provide a wide array of customization, particularly in complex applications, such as MMOGs.

So I was thinking back to my old MUD days and my current system administration days. Aliases. Macros. Hell, commands. There are many MMOGs that have already implemented this idea, but I think it is worth mentioning that providing your users the ability to execute all game commands in text format and giving the ability to create their own macros, shortcuts, hotkeys, aliases, etc. is the way to go. Now, I am not sure how this could have been applied to the WoW quest discussed in the article, as I don't know if the hotkey used was a one-off spell, or scroll, or potion, or if it was some other game mechanism that may or may not be easily accessible as a text command.

While text commands do not make the user interface experience easier for new users, it certainly makes the user interface much better for experienced users. The ability to label your own buttons is particularly compelling. When games are not simple enough to use a directional pad and a couple of buttons labeled A and B, we must come up with more elaborate solutions. So what about mouse gestures? Here is an interface concept that is implemented in web browsers and other applications that could be leveraged for use in games. Not all games, perhaps, but some. Why not give users the ability to assign commands to mouse gestures? Sure, repetitive stress injuries might abound, but I'm just thinking out loud here.

In any case, I can remember my first MUD ever and being totally floored at the sheer number of commands available when first learning to play it. After administering my third MUD, I looked back and thought about how much I had learned or just plain absorbed from the use of the MUDs throughout the years. At that point, I could have rattled off every command from several different codebases, but that's not the point. The point is how to get from the new user to the seasoned user with minimal pain and suffering. How do we make the interfaces better? More clear? More concise? Have we reached a dead end? Or have we just become complacent because the new MMOG has roughly the same UI and commands and we're all just seasoned players?

I don't really have a problem with game UIs, as long as they let you have some control over it. EQ2's is fine imo. WoW's sucked but was good if you modded it (but then had to dick with updating the mod every patch, which sucked). I did put in EQ2Map, but that's not really a UI issue, for the main UI it's all stock for me.

Maybe we just need to dumb it down to 4 buttons on the face of a gamepad ;)

And that's his whole point, that WoW in his example needed a 3rd party mod installed to make the interface more extensible. The interface should be more extensible by default as is the case with EQ2. I'm in the same boat with you, Sky. I never installed an interface mod on EQ2. Perhaps it was just laziness. I don't know. And no, 4 gamepad buttons are not the answer, although even gamepads have gotten ridiculous with the number of buttons. I never played on the Wii, but it looks like it may have less buttons than others, thus going back to a simpler model of user interface design. Of course, the motion sensing stuff adds to what they can do, but any way, I haven't seen one so I can't speak to the number of buttons or the simplicity. It just "seems" that way from what I've read and heard. Back to the MMOGs, even EQ2 has the "cryptic icon" issue. Hell, Word has that problem if you've never used it before.

I'd like to see context menus more heavily used.  This would be especially useful if menus were customizable.  I'm speaking in EQ2 terms here, but it applies universally.  Imagine if, like your hotbars, you could set up your spells (abilities, whatever) in a context menu.  Casting a heal would be a right-click, downward mouse movement, left click.  As opposed to, see who is stealing agro, press corresponding function button, hit 1 (heal hotkey), find MT and press his / her function button.  In the current system, I've had to (riskily) take my target from the MT, and perform several steps to execute one spell for that idiot that wants his dps to be teh winnar.  Sorry, healer blues talking.

This is probably a personal preference, but it applies even going back to my EQ days:  I hate icons.  Icons hide the true meaning of what that particular button is supposed to do.  This is why I would make macros for any spell that I memorized.  Its much easier for me when the shit hits the fan to find the button with the big yellow HEAL on it than some rinky-dink heart icon with a pink potion and a teddy bear on it.  No matter how many thousands of times I've healed, my brain still pauses for a nanosecond and asks, "Is that really the heal spell?  And why is it so pink?"  That nanosecond could kill you.

I've always been tempted to set up some kind of voice recognition for macros, but then I'd have to admit that I'm a complete dork.  Even teamspeak makes me nervous because it reveals that Mr. Shadowknight really sounds like Steve Urkle, and so does that hot DE that I've been hitting on for the past hour.

The best UI of any game that I have played so far has been original SWG.  Movement keyed with the right mouse button, actions accessible from the left mouse button, F keys for special activation instead of the number keys, a stout macro system, the combat queue, automatic camera/view movement with mouse movement, and system messages that showed up on your screen as well as your chat box...I could expand my character's and targets status bars to a big enough size that I could watch it with my peripheral vision and still watch the action in the game. 

It blended with the game better than any I've seen so far.  I didn't notice the UI, which is why I liked it.  I hate hate hate hate hate hate hate hate hate watching timers on special icons, trying to watch buffs / debuff icons on my HAM bar, etc. 

The last time I played any MMO, much less SWG NGE or WoW or Vanguard or  :nda:, I found myself watching rotating timers, the HAM bar, and such rather than watching the GAME.

Something is wrong there....

Bullshit.  I have a character at 70, and I never used any mods.  You might need them to raid effectively (I haven't done any of that), but that's not something you're going to be running into "a few days" after rolling your avatar, unless you're catassing to an inhuman degree.

Which is not to say that I don't think the hotkeys are out of freaking control, but that's more a problem with the "Ding/Grats" Diku system, where you must get some new fob every level or the game dies.  My Shaman has (with the default UI) something like eight different toolbars, each with twelve slots, and every one of them is full with some dongle or other (and there's still a few more items I'd like to stick to it if I had room).  That's the way the game is set up, and there is no interface short of jamming a plug into the base of my skull that is going to let me hit any one of ninety-six hotkeys whenever I need to without just slapping down an ugly pile of buttons somewhere.

UI is evolutionary (whoa, deep thoughts...). WoW's UI is actually damned good, when compared to old EQ1 or, worse, old UO. iWe learned those UI's fine, but they were alienating to anyone without a great deal of desire to simply be part of this then-new experience.

I would guess that it's mostly MMO veterans that modded WoW. Like EQ2s UI, WoW's is perfectly usable out of the box, unless you come to a new MMO with preferences set up by prior experiences.

Of course, the game mechanic itself is mostly to blame. There's just so much information that needs to be managed. But I do think that we're going to start seeing that change. In reviewing the UIs from the upcoming games we know about, it seems as though a lot is changing. Maybe it's not for the better, nor the worse. But at least people are thinking beyond the hotkey bar and target lock.

Different people have different requirements for UIs. Using WoW as an example, a new player needs easy access to help, tutorials, and not an overwhelming amount of options on screen. A raider requires a heads-up display resembling a fighter aircraft, preferably with the interface itself changing and morphing depending on the situation. Intermediate players look for interfaces that show quest completion, help with combat efficiency, and keep your cyb0ring IMs seperate from your guild chat.

I disagree with the original author that operating systems are bad examples for UIs. I disagree mainly because users are, by necessity, familiar with operating system conventions. They know to look for right click menus, they look for window widgets in the appropriate places, they expect to get help when hitting F1, and expect a menu of some sort somewhere. Ignoring the training your users come to the door with is probably a bad idea; at least it's a wasteful one.

My main pet peeve with game UIs are ones that are ornate at the expense of functionality. Your user interface is not a good place to show off how arty you can be. It's supposed to be usable, first and foremost. One of my first reactions upon trying Everquest 1, back when it first released, was how hideous the interface was. It may have looked fantasy, but to my eyes it looked like something cooked up in Visual Basic by someone who liiked marble textures.

This is an interesting article, because the author has no understanding of the medium and feels that his experience in utility design should naturally transfer to MMOs. He believes that MMOs are directly analagous to, for example, CD burning utilities, and that developers should pick the best defaults to meet the needs of 90% of the users and customization options should be disabled, hidden, or non-discoverable. That is entirely wrong. MMO players are all power-users, and power users love customization. They don't download UI mods because they're necessary to play, they download UI mods because they're power users who live and breathe the game. WoW very correctly feeds complexity to new players over the course of many hours and is a sterling example of great MMO UI.

That's not to say that players should be forced to fight the interface. His example of possessing the dragon was perfectly apt; of course presenting a player with four unlabelled icons and no feedback or direction was bad design. But extrapolating that one encounter to condemn the UI as a whole is foolish.

QFT.

Developers who seem to think the UI should stand between the player and the things the player is supposed to care about (characters, objects and settings in the actual game world), or who seem to believe that the UI is part of the game world, irritate me beyond comprehension.

The only reason your UI needs some level of game-related flavour, is so that it becomes less visible and jarring when alongside the game world objects. Everything is about bringing the player closer to the game world, not standing between player and game full of self importance and bad textutures.

The current fiasco on mtgo v3 is another perfect example of how to do everything wrong.


He's not only wrong about MMOGs, he's also wrong about CD burning utilities for people who operate CD buring utilities for hours at a time, most days of the week.

A better comparison to MMOGs is something like graphic design or CAD software, or other professional/pro-sumer tools. If people are going to play a single game as a hobby in itself, design needs to account for their growth as a player.

The autor's example in a mmog, putting you in a new role with new icons and not telling you what they do, is something 'happy-friendly-flavour-filledy' GUI designers get wrong often. But even if those four abilities had been text options, the quest design meant you'd still be guessing which one works, the key problem was quest design, rather than overuse of flavour in a UI, or too many options.

I disagree. The vast majority of MMO players are MMO hobbiests like we are. They are previously a gamer that finds an MMO they like and gets sucked into playing it for months and months at a time. That's not a power-user. That's someone who comes from playing games designed for 10-20 hours of play staying in a genre designed for 100-400 hours of play.

Maybe the user will eventually want to customize their UI, but this is by no means a foregone conclusion. I know plenty of endgamers who kicked and screamed all the way into CT_Raid for example, and getting the CT stuff is about as turnkey as they come.

This is corrollary to discussions about how many people are deeply invested in maximizing their total MMO experience. Think about how many people are in a guild versus how many are bothering to hit the guild forums. The greater the percentage, the more hardcore/niche the guild is, in my mind. Most times you're lucky to get 10% of your members to care that much.

People approach new games expecting to play it right out of the box. Modern MMOs are like this. So their GUIs should reflect this. I think they largely do, based on the type of games modern MMOs are. So a fundamental shift in GUI needs a fundamental shift in the total game play experience, as we've seen in the demos of TR and AoC.

When you dedicate hundreds of hours to a single activity, you can't be described as casual. Hobbyists are anything but casual; that's the wrong word to describe it. They spend hundreds of hours organizing their stamp collections.

Those players may have resisted installing ct_raid, but that was an example of bad  design, since the mod was absolutely required to progress through the endgame. It's also an obsolete example, since a raid interface is now included in the game.

You're assuming that UI complexity and customization necessarily implies picking and configuring dozens of mods just so. That's not what the author was talking about. He was talking about the default UI complexity; with multiple button bars, hitting shift-2 followed by 4 to summon a hippopotamus, monitoring the status of 4 other players as well as your own health and power while simultaneously trying to situate yourself to attack the monster's backside, stuff like that. He looked over his nephew's shoulder playing his 60 in WoW, and wrote an article based on that experience rather than playing himself. While there's a great deal of attention and decision-making required in a simple diku-style group battle, it's all introduced slowly and iteratively over dozens of hours. That's not bad design.

There's a much better and more in-depth blog linked in the comments to the above article.

http://www.unwesen.de/articles/wow_design_improvements_introduction (http://www.unwesen.de/articles/wow_design_improvements_introduction)  There's the link to the guy's intro article.  Some of you folks (Darniaq) will have all kinds of fun with his views on social-ties and expanding the UI as his views parallel yours.  I, however, would find most of his desired additions merely distractions and nuisances at best and damned inconvenient invasion of my personal space at worse.

 Game vs Lifestyle/ World rears its head again.

When I did the base SWG UI my intent was to make something that was more understandable and usable by non-MMO players. Based on the stats we gathered and the polls we did, it did OK at that -- it was greatly preferred to the other UI configs by folks who had never played an MMO before.

But we had so many MMO vets that wanted, well, EQ's UI -- which I have always considered to be terrible -- that we had to put it in as an option, and this crowd persistently pushed for it to be made the default, which eventually happened. Whereupon the other players started clamoring for it to go back the other way... you get the idea.

The design premises for the original SWG UI were pretty simple:

- move the commonest activities to the most basic interface
- chatting is more common than hotkeys
- hotkeys should be on keys for, well, hotkeys
- make everything have a default action that you reach by dbl-click and make extended options available in other ways
- adhere to GUI conventions if you can
- make everything back into a text command
- permit text-based macroing for high-end customization

but we ran aground of the fact that mouselook swallows your mouse, and so to have both mouselook and a cursor will *always* require a toggle key, and a toggle key is *always* obscure. We also ran into the basic divide between those who chatted more, and those who needed a billion hotkeys accessible quickly for rapid combat.

That said, during beta, when users were polled, the default SWG UI won handily in terms of user preference, over the three other default mappings we provided...

I think he really hypes up that addons are needed, and seems to think (as some of you do) that allowing addons is either cheap (letting players do your work for you) or bad for usability.  I think neither.  The fact is, 8 million players can come up with better ideas for everything than your 50-man dev team.  Why not let them come up with UI ideas and cherry-pick the best? 

In terms of usability, I think it's genius that you can access additional usability only by knowing the current UI inside and out.  You can almost make the argument that some mods are required for raiding, but raiding by definition is meant to be a bleeding-edge activity.  To even access it you have to have driven the UI for 70 levels.  If you don't know it by then, I have no idea how to help you.

His example of Mario is false too, IMO.  You can't tell me that you knew which button, the helpfully labeled "A" and "B", threw a fireball the first time you played?  "Fireball" has both the letters "a" and "b" in it, no help there!!  However, the first time you did it, you learned it and never had to think about it again.  Likewise, I suspect the second time you do the quest he mentioned, you know how to manipulate the pet bar.

UI improvements will always be worth striving for, but short of a new control contraption for each game, you will never have zero learning curve.

I tend to agree with just about every point you make here, as well as all the intentions that Raph makes.

Personally, I think game developers of all sorts should extend the ability to modify GUI's to as big an audience as possible. Guild Wars had at least a basic ability to move your GUI controls around with an editor, and I think that this should be the basis for future GUI development--a visually based editor that allows for any user to modify several different "themes" (not the usual sense of themes being purely graphical, but information presentation styles) of UI quickly and easily.

UI Mods don't relieve the developers from making good UI's, but they do allow customers to adjust their UI to their personal play style and needs, but currently the games that allow for maximum flexibility in information presentation also require you to be an XML artiste...and that limits the general benefit of UI modding.

One of the issues in UI modding is that the data people want to view is buried in the network stream with other data that they don't need to/shouldn't be able to view directly, and most user moddable UI's provide some form of tagging system (a meta-UI if you will) to present this data, and then an XML layer to organize the data. I think this should both be formalized (would be great if it was cross-developer, but at least should be consistent within a developer's releases), and an editor provided that allows for easy manipulation of both what data is presented, and how/where it is presented.

Finally, development groups that allow UI moddability should absolutely support that moddability. It is endlessly frustrating when a patch comes out that literally causes a modded UI to ctd, and that flat out shouldn't happen...and doesn't need to.

It's been so long I don't remember the options now.  I think it was the base interface, but modal so I could have additional binds.  No matter what I used, I liked that players had the option to use different styles built into the game.

What I would be interested in seeing in future games is a set-up screen, before getting into the world, which let us play with different interfaces to see which we liked best.  Being able to do so before we ever contact the server gives bonus points.  Yet more points for making it possible to save the interface into a loadable file on your own machine.

And then, have those ideas show up in other games.

I'm speaking specifically of the H.U.D. quest progress lists that were modded into WoW, nicked by Blizzard and put into the default UI, and then showed up in the default LotRO UI.
But I'm sure there are more examples :)

So nobody agrees with the article, then?

Group hug!

Well, it's fun to talk about, so as food for thought it works.  But most of his conclusions are overblown or Captain Obvious ("usability is HARD"), IMO.

They should be pleasant and good looking though. That awesome fountain in before City Hall? - players will look at it once, twice if you're lucky. Healthbars? They will have the bloody thing before their eyes one hundred percent of time.

WoW interface is one of the worst out there, partially due to WASD movement choice. WASD has no place outside of FPS. In WoW it takes TWO hands just to move your character... what you suppose to use for macros?

Eh?  There is a "click to move" toggle in the interface menu, in addition to the default key layout which lets you move around fairly comfortably with the mouse alone (you just can't strafe).  Even if there wasn't, WASD only needs one hand to work.  I can't think of a movement scheme would work better in WoW.

Things that suck about the  WoW UI and design:

* Bars on the corners of the screen that must be constantly watched when the action is happening in the middle
* Being able to do better in combat by watching moving bars than the actual 3d landscape or item that I'm fighting
* Dozens of 'action' icons that all must be hotkeyed and immediately available for my character to be effective
* Actions that cannot, by default, be activated while maneuvering due to keyboard/hand distance restrictions

I have to admit, one of the reasons I like CoH so much is because, like a power user, I was able to reassign every game function to a place on my keyboard where it worked better for me.

--GF

Have you tried click to move in WoW? "You are out of range", "You are facing wrong way" and that just in PvE... Shadowbane had working click to move, WoW for some reason decided to sabotage it and not have auto-close and auto-rotate enabled with it and as a result its useless.

Wow, especially in PvP, requires circle strafing and overall super touchy on positioning/direction. For example you can completely shut down newbie caster by simply jumping through while spell is in progress, if you don't turn with mouse you will get 'facing wrong way' every time.

UI, almost by definition, breaks immersion. It's all game and zero world, and no amount of theme and set-dressing will make it feel more authentic. It should be intuitive, unobtrusive, and located appropriately to draw the player's eyes to where the character's eyes would be. Anything that isn't in use should disappear... figuratively if necessary, literally if possible. In a perfect world inventory would be visible objects carried, damage and status effects would be represented by textures and animations, abilities wouldn't have cooldown timers, only cast times, and players would all use voice chat rather than typing.

Not bloody likely. This would need to be the number one focus of the entire game design. The MMOG is too young a medium.

Until then I agree that what we want is wildly flexible and customizable UIs that let the players create what they need. They'll make gamey stuff, but the stuff that sticks will be the best there is.

That...may be a perfect world to you, but it's a game I wouldn't touch with a ten foot pole, especially with the requirement to communicate through voice chat.  If damage and status effects are represented by textures and animations, then I don't know instantly and completely what is happening to me.  If I get afflicted with Slow, I know what happened.  If I get afflicted by a purple thing, I have to differentiate it from several other similar but uniquely different purple things, and then hope I remember exactly what that particular purple thing does.  It's only if the entire game is going for maximum realism that we should have an interface like that, and that wouldn't be a mass-market game, it'd be one on the level of ATITD or Shadowbane's number of users (if that many).

The main problem of interface to me at the moment is the fact that - and I'm sure developers are starting to consider this more and more often - the number of abilities in the game can easily outnumber the amount of easily accessible buttons, and more importantly, easily rememberable buttons.  I have a Nostromo N52, so I have four hotkey banks of fifteen keys + a D-Pad to work with, and my main problem is not reaching my buttons quickly or pressing the right combination of keys - it's remembering all the abilities my character might have and which button or button combination they're mapped to.  We want more abilities - but at the same time we need to somehow be able to use them all effectively in a fight.  No amount of UI streamlining will make me effectively able to choose in a matter of seconds or fractions of a second between over a hundred abilities, and be able to find them and select them quickly enough, unless I have practiced this for months, essentially memorizing an entire method of control more through ingrained knowledge and reflexive reaction than anything else.  Even if introduced slowly, that's a lot of dedication from your players.

Meh, quoted myself instead of editing.

/agreed with Koyasha.  Let us not mistake realism for good UI.  Let's also keep in mind that combat in real life is chaotic, confusing, hurts, and is not a lot of fun.

edited to respond to Sinij:

I have never tried it before today, but I tested it out in response to this post.  I don't see what you mean.. it does have auto rotate, and in fact unlike SB puts a marker on the ground to show you where you clicked.  I clicked behind me, my character turned and ran there.  I can adjust (click somewhere else) mid run. I found it to be more usable.  What's auto-close?

Auto-close moves you into range for whatever ability you're using.

The type of interface sinij is looking for does a lot of the work for you, so you're just pressing combat buttons in the right order.  Almost no movement is necessary on your part.  I much prefer the wsad stuff to all that bull.  (Oh and SB did mark where you were moving.. it just went away after you clicked, I guess that's what you meant?)

And therein lies yet another problem.  The movement interface alone can make or break a game for some folks.  There is NO universal UI anymore than there is a universal set of game mechanics.   So much of this conversation is opinion and personal preference, but the original author (and the article I linked) treated it as if it was some sort of cosmic law, and couldn't fathom WHY such things weren't implemented differently.

CoX is one of my favorite UIs as far as remappability and function.  I rebound everything.  Their keymapping interface is a lot better now though, making manual binds not quite as necessary.

I love my Nostromo... and the SWG launch interface was the best I've seen yet. Why? It was simple, low-key, and unobtrusive. No gaudy riots of color. I was upset when they not only put in the nasty one that is in now, but even *removed* most of the mono-color simple line drawing icons that I liked. Hell, at one point I was serously tempted to create a printed layout that would lay or hang in my line of vision that contained the macro or button for the skill I wanted to use. Then I could run "fullscreen" so the GUI would not be seen at all.

That seems strange to hear from a PvPer.  I thought it was all about skill?

And yeah, there is a yellow circle on the ground where you clicked until you get there.  I find it particularly useful to know how far you're going, since the direction is obvious.

Really? I completely disagree. In general if I was not in a super-heated battle I would use the right click+left click. Otherwise, if it was WASD, it would be because I was punching in macrokeys on the g11. To each his or her own though.

Wow. You guys are way more picky than I am. I think WASD movement+mouselook is great, I guess I'm used to stretching my laft-hand fingers quickly to activate abilities from being a musician or something. I almost never mouse-click things if possible, just try to use the mouse for looking around.

I might be mistaken, but it seems to me more like looking for a chance to bash WoW at every turn.

I try to bash it once in a while so I don't feel like such a fanboi, but I have trouble finding faults!  And if you invent them like Sinij does, you look like an asstard.

There is nothing wrong with WASD and mouselook in WoW and the like. It gives the impression that a player's skill is part of the equation. Which, by the way, is true anyway. Anyone who thinks it's just about the numbers, particularly in PvP, hasn't done much PvP in WoW. Target-locking and stats only get you started.

The big issue in my opinion is how conversation happens. The default method of conversation for the genre is and will remain for some time text. There are a lot of reasons VoIP hasn't taken hold here, almost none of which apply to text. Regardless, text conversation means attention shifted away from action, which means downtime.

Maybe VoIP will become the default communication method someday. But I think we're years away from that. As an interim solution, I wish a company would license voice-to-text technology, allowing people to talk into a mic but for it to come out as text in a chatbox or speech bubble.

But I suspect I'm in the minority here for that wish :)

For the record, this is not actually true.  Atul is a friend of mine and I assure you he is not only a longtime WoW-player, but further, the two of us spent a summer at college working approximately 10 hours a week, spending the remainder of our time holed up in our dark, cave-like dorm room, eating nothing but ham salad sandwiches and playing EQ until dawn.

None of that is to say that your opinions on MMO UI's and whether or not they are crap are necessarily wrong, but regardless I can assure you this is not some random outsider's perspective you're reading.

I'm used to WASD - modified (movement ASD,WE,C targetting TR, activating combat FGB) so I prefer a UI scheme that supports WASD but also supports re-mapping each key to a different spot (which, BTW, WoW does support).

I notice how we're starting to talk about the control device (G15 keyboard) as being part of the UI, or at least having the UI be designed with it in mind.  That's fine, not everyone has a G15, but a lot of people do, so there should be some default option available to make use of it.

Personally, when I look at a game's UI, I break it down into two parts:  the control system, and the information system.  For the control I prefer WASD, like I said.  For the information system, the part of the UI that tells me what's going on in the game, in MMOG's you really only have 3 things:  stuff that's happening to you, whack-a-mole with the bars, and the graphical depiction of the world around you.  I think that as long as each of these aspects is treated as its own subcomponent, and optimized separately, the overall UI should end up being good.

For example, I like the pretty world, but I prefer a compact whack-a-mole area, and I prefer color changes to it rather than a sea of numbers or a sea of bars, or tiny debuff icons popping-up that I have to react to.  I'd rather have a compact grid, with no wasted space, with colors.  That's different from what I want for the stuff that's happening to me:  I want numbers for that, and color-coded text in a window, and many options to customize what's displayed and what's not.

So anyway, I think my point was sub-components of the UI, treat them as separate.

Nostromo FTW

I used "Click to Move" in WoW while I was travelling a lot. Right click on the horizon and the dude runs forever. I never had the problems Sinj did.

I'll propose the reverse. Maybe putting in OSX-like 'characters' (like the creepy guy who calls out the time on my mac) to do text-to-speech conversion. That way you could put in an easy-cheesy text filter to remove potty talk (or whatever else goes on your blacklist, like "Britney Spears" or "Rick Astley" or "LFG"), and have more appropriate voices. That hotzorz femelf actually sounds like a femelf and not a 43yr old office drone (I would so not hit that). With games like EQ2 where you pick a default voice set, it could base off that. Then you could turn on group/guild/raid/whatever voices as needed, or use all local voices (though that would be annoying in an Auction House, I guess).

It's an interesting idea, but we read much faster than people can speak, and it's easier to have many different threads of information in text form. Text also maintains the anonymity, something that even the huge virtual-spacey games keep to. People like their text, relegating VoIP to core groups of RL friends or people who are really serious about the games.

My little idea there was /all about/ maintaining anonymity and not breaking the atmosphere. Not VoIP, but a text reader. You'd still have text, but you could turn on a text reader for flavor.

Now select that paragraph, and navigate Application Menu (Safari for me) > Services > Speech > Start Speaking Text. That's all I'm talking about. It would be kinda cool, I think. Not for normal communications, I'm just proposing a cool atmospheric gimmick using existing technology.

I like that idea. Probably only because I've become hooked on text-to-speech because my inner geek wants more ways to listen to book content while doing other mundane tasks. If I can't find an audio book, I just get the ebook and have my Mac read it to me. And with the ability to adjust the speed of the reading as well, it comes in rather handy.

Oh! As a matter of fact, you lead into a really good idea, Sky. EQII wanted to add to the immersion by including all the voice-over quests and such, right? Well, if they advanced text-to-speech a little bit more, they would no longer need huge audio files to accomplish the task, they would just need to set a text file for each quest that triggers the appropriate voice and invoke the text-to-speech engine to read it. Imagine the possibilities for games here. While I am sure this idea has come up in meetings with designers and developers already, devs take note! Even in the current incarnation, space based games could make use of it, albeit campy use.

When I was in the shared office I used to crank up my speakers, write some jokes in textedit and have the computer speak them while I went on break.

Last time I had to share an office, I ate herring straight out of the can, raw white onion, and white bread for lunch. For weeks. It only took two weeks.

Now I'm alone again, naturallllyyyyy!

In a way it is, but not desirable skill. Its very limiting on latency, where any bit of lag will drag down even seasoned PvP into mediocrity. Its also detracts from decision making, it becomes a lot about 'how' and very little about 'what'. I believe movement-based combat should stay in FPSs.

Examples to illustrate my point:

Movement-based combat: You attack other player, trying to position yourself behind in order to avoid attacks. Timing your movements with your auto-attack timer, so you are in range only during brief period it takes to attack. You wait for best opportunity to use your stun/snare and make sure that you are never in position where you could die in just one stun/snare. You try to jump through, use LOS obstructions or go out of range to counter enemy attack. Usually there is lot of running around. You don't get to kill anyone without limiting their mobility first. Kiting is major part of the game. Best defense is simply running away, that is if you get away. Movement speed is major combat attribute, alongside with hit points in importance.

Decision-based combat: You attack other player, first carefully probing to determine preexisting damage shields and damage reflect/mitigation buffs. You follow with debuff for your favorite opener combo. You time your combo and take notice of your opponent's actions. While opponent start with his opening you make quick decision between defending by counter-spelling, defending by interruption, defending by mitigation or pressing with your own attack. In the middle of your attack your opponent realizes what you are doing but doesn't have time to properly counter it. His own attack harmlessly redirected by your mitigation buff. Now your opponent has a choice of taking big chunk of damage, putting him 'one away' from death or blowing one of his long cooldown spells to stay well. Its 1:0 in your favor ether way but battle isn't over.

And you wouldn't even be able to play like you were if it weren't for mod creators pointing out the ridiculousness of the old base ui when it had no side bars. Or scrolling combat text, or half a dozen other things that have cross-pollinated from the original barely usable ui to what it is now. Oh, and there are 2 extra bars you can only gain access to via a mod, iirc.

Molten Core was entirely doable with the default UI of the time (I certainly made it work). Kharazan and up is probably doable with the default UI today, perhaps with the exception of threat meters. I'll give it a shot this week and see how it is.

I've played around with text-to-speech quite a lot ina previous life, particularly on the development side. We talk about the costs for including voice in an MMO where the aggregate datasize is basically only limited by the size of the 'net itself. Imagine that on a standalone consumer product though, where the amount of sound that can be stored is measured in single-digit minutes. :)

TTS still doesn't sound natural enough to be used for voiceover text if you're striving for immersion, as EQ2 was. However, like Soukyan, I think you've got a pretty good idea here for players who want to try and replace attention given to a textbox over to actually, like watching the action.

This should not be impossible to emulate either, but you'd need to probably get a WoW mod developer involved. This would also be easier to experiment with right away on a Mac because they include TTS in the core OS and they have many voices. Windows XP has "Sam" and that's it. On the Mac size, I always liked Zarvox, mostly because it sounded like a Cylon, but secondarily because it didn't clip the individual words as much.

Anyway, who's got an active WoW account, knows a Mod developer, and wants to head this up? We need:

A mod that will read the chatlog
The mod would look for chat lines that start with something like "<speak>"
The mod would then launch the TTS service and play it back

Now, if the mod could not do this, then we'd need an external program running in the background to read the chatlog and do the same thing as a separate process. I'm pretty sure that'd be fine by EULA/TOS standards because it's just the chatlog.

WoW mods can't, as a code limitation, interact with external executables as far as I know.

Well, I don't have an active WoW account and have not for some time. LoTRO on the other hand... Perhaps I can whip something up with it. The problem is that it requires using Windows. I suppose I could resub to WoW to test this out on the Mac. The benefit is that I don't care if the WoW account gets banned for EULA/TOS violation. ;)

I recall something similar. However, would a third-party program reading the chatlog in realtime be possible? That's about all we need here.

I was going to add, I don't think text-to-speech and speech-to-text and voice masking is as advanced and easy to implement as we'd like it to be.

'possible, but you'd probably get tagged by Warden and banned quickly.

Not that I know of.  You'd probably have to intercept the network stream, leading to Warden problems.

BTW, good summary Sinij.  Sounds a lot like UO pre-trammel PvP, which I liked, but I also like WoW PvP.  So it goes.

I couldn't get into WoW, but I find it impossible to hate on since I really gave it a thorough go.  It's not the sort of virtual worldish game I'm looking for, but it's not trying to be.  What it tries to do, it succeeds at very well.  It's the world's best apple.  Hating it for not being an orange doesn't seem fair.

I hope I make sense, I'm really drunk.

That's roughly how you write Addons in WoW (that's an oversimplification, but it'll do). The problem with something as mentioned in the article is that you have a moderately steep learning curve, and are under a huge amount of (time) pressure to get your thing done. Learning curves are fine, as far as I am concerned, but if there's a lot of pressure, work has to be done to make things easier on the player. Hell, just placing the buttons smack in the middle of the screen, with big icons and descriptive title would have been a good help.

I think - but absolutely can't prove, it's a gut feeling - that with UIs as complex as WoWs, which really get close to regular Desktop UIs in terms of what standard widgets are available and how you can expand on them, there's a feeling that you need Desktop-UI-like behaviour as well. Close windows with a little x in the corner, not a close button, and that sort of thing. Out of that feeling derives the need to place new buttons close to the button bar, so it's consistent with the look & feel of the game. And that's an understandable feeling, but only works for UIs you're likely to see & use a lot.

I don't know, I'm guessing... I'm just thinking that maybe the consistency idea was more in the designer's thoughts than the usability idea.

No, the chat and combat logs only get written to disk on a non-forced garbage collection which is very infrequent (think tens of minutes at minimum).

Hmm, ok. That's not good.

Per above, we know WoW mods can't call external programs. However, isn't the text-to-speech stuff built right into the Windows OS? Could a mod access the info in that way (what is that called, "making a call to the API" or somesuch?)

Aside from my WoW "hacking" meanderings and attempting to get my Mac to read a quest log to me, I found another article somewhat pertinent to the original topic, if only for the fact that it uses another game (Guild Wars) as an example in "humane" interface design.

Aza Raskin (http://www.alistapart.com/authors/r/azaraskin) writes an article titled Never Use a Warning When you Mean Undo (http://www.alistapart.com/articles/neveruseawarning) for A List Apart (http://www.alistapart.com/).

In the article, he discusses the ease with which users can accidentally click OK to an "are you sure?" warning and often times lose work that we meant to save or keep. He refers to a method that Guild Wars uses to attempt to regain the users attention to what could be an irreversible action.


But Aza goes on to state that the issue is not how to get the users attention in a more effective manner, but rather to provide the user with a method for recovering from an accidental loss/deletion. GMail is a good example of this that is given. When you move an item to the trash, there is no "are you sure?", but rather an "Undo" link provided to reverse the action should you realize you did not mean to do this.

I am not missing the fact that the entire Recycle Bin (Trash Bin) paradigm is meant to save people from their own mistakes, but that's not the point. Looking at game interfaces and character deletion, for example again, suppose a player clicks Delete and types the character name in Guild Wars to delete the character. Once complete, the character is gone. One could argue that the game company may be able to recover it after some phone calls and some trouble, but lets assume that is not an option or not one that the game company is going to entertain very often because, let's face it, they gave you a warning. Looking to another method used by the game Eve Online. If you delete a character there, there is a waiting period during which you can undelete that character. If I remember correctly, the time for character deletion is 24 hours, but I want to say that it may be on the order of days.

In any case, it's a compelling thought, and it is interesting to look at how games implement these features into their user interface. Just more food...

The game pad buttons are generally colour-coded which helps to make them more intuitive -- say, a red button is "agressive" and fits naturally with functions like attack or fireball. In similar manner "A" comes before "B" and so "A" tends to be associated with primary function, etc.

On the subject of this pet toolbar thing, I wonder if part of the point wasn't... if the whole purpose of quest is to forge some gizmo with special breath, and the ability to control the pet is on short timer and one pretty much only uses one dragon function to that... why have toolbar with multiple buttons to begin with? The whole thing could be just as well entirely removed and turned into cut-scene that happens once player succesfully takes over dragon mind. Which could even make it more "immersive" experience without the awkward break that has player struggle to understand odd piece of UI, taking their attention from actual quest they are trying to complete.

I have no sympathy for people who complete a three-step character delete operation "accidentally". There are some behaviors we shouldn't be required to design around.

I'm not so sure that he was necessarily saying that it was a bad mechanism, either, but rather that it was one that makes the user stop and take notice as it isn't the usual, "Are you sure?"/Click OK method.

Someone just played PSO Blue Burst!

Yeah, but if you are paying CS personnel to answer the phone and talk to those losers who you have no sympathy for, you'll have even less sympathy for them, but they will still cost you money.  Giving an undelete period is a lot more idiotproof method.  And there are a lot of REALLY GOOD idiots out there.