Untill the MMO industry revamps the way we get quests, gives us a reason to learn the lore, and redesigns quests that actually require brainpower to accomplish, the mechanics of npc/player dialog interactions is really a worthless discussion.

I we look at the quests, 99% of them do not require any skill, knowledge or thinking, and have no consequences whatsoever.
As long as it is so, the dialog is basicly worthless.
I guess the dialog and options should be proportionally as deep as the quest is.

[...] mulling over the ideas in this post over at nerfbat, I thought I’d tackle it as it concerns my game [...]

This is another case where I think that you can’t really go with the original EQL style of dialog interaction anymore without being niche.

And I’d like to throw in, I have no problem at all with niche. In fact, I’d prefer it if more games would shoot for niche, building with smaller teams and shooting for well done smaller games… I enjoy playing WoW, but only because of the people I play with, overall the game is not really what I would like to be playing, but its good enough for me to not hate playing it.

It’s less convenient, for one, to figure out what to type and progress the dialog or get the quest. For most people (those who don’t have much RP in them), it just becomes a barrier to fun. Secondly, it would just encourage people further to go to Thottbot-type sites.

There are also certain design beliefs these days that basically state that you shouldn’t put anything in games that make people feel stupid unless they can easily try an “alternate path” after they fail the first time. This doesn’t apply to everyone, but to keep things accessible to a wide audience, it’s not a bad idea to adhere to.

Maybe I should write a follow-up post to my post, “Casual Games Kill Hardcore Gamers,” with one called, “Casual Games Kill Hardcore Games.” This is another case where I think that you can’t really go with the original EQL style of dialog interaction anymore without being niche.

Well… one thing that just feels “wrong” to me is when you have a quest that sits in your journal/whatever. So, give quests a timer of sorts, but don’t make it something extreme.

For example… this quest gives a timer of 1 game week (which would, by ‘conincidence’ be precisely 1 real day).

To use a sort of RL example…

I visit my uncle and “hail” him (yeah, I know… it’s an easy way of saying it, and it sticks hehe). He’s a lazy guy, and tells me he’ll give me $20 if I mow the lawn… (the quest). I could…

* Decline, without any real loss/gain (ie make up an excuse if needed)
* Consider it, and tell him I might if I have time (no time factor associated, no real loss if you don’t do it)
*Agree to do it. He’d expect me to do it within that day, and I’d get paid upon completion (and he’d probably look at me a bit more highly for doing it). But if I say I’m going to do it, and then I don’t do it, you’d imagine that he’d be upset. Same if I did it a week later instead of the day I told them.

So perhaps give your quest developers several options…

* Pure Accept/Reject (much like it is now), either with or without a time frame, and with/without a “failure penalty”
* Add in the “consider” options which removes the penalty (or mitigates it) but perhaps lowers the reward
* Add a time-range such that, instead of “failing” if you don’t finish in time, you get a bonus if you do it within the time range. Perhaps even have multiple ranges… using this you could even make a sort of game in a game…

IE, the quest is “Go and kill 10 rats.” If you complete it in 5 minutes, you get 10 gold. If you complete it in 10 minutes, you get 5 gold. If it takes you longer than 10 minutes, you only get 1 gold. (Of course, the possibilities for stratification and such would be limitless).

OK, all that aside…

As for hailing and either having to select text or give keywords…

I really liked the EQL (EQ1, whatever) style.

I’ll give an example from EQ2 of one reason the old style is better…

In order to gain access to this building, you have to pass this little “test”. The punishment for failure is simply to wait and have to take it again.

So, with the EQ2 style, theres say 5 questions, with 3 or 4 (lets say 4) possible answers for each. So, you can think of the conversation as a sort of probability thing…

Theres 4 “wrong” answers at each level, so 4×5. And there’s only 1 correct path, so thats 21 total possible conversations you can have with that NPC (20 of which are “fails” and only 1 conversation path is the correct one).

Thats a lot of writing a Dev has to do… 20 “fails” for just 1 actual progress. Even copying & pasting thats still wasted time.

Not to mention you’re not truly taking the test… as with enough time you can simply map out the graph (I’m a math guy… but I won’t really go into graph theory hehe) and find the solution… the actual text then becoming meaningless. (I’ve seen hints that do that essentially… Ie, pick 1,3,4,2,5 to win.)

But… if you go with the user-input-response and keyword-check, then you only have to program the correct answers… and perhaps remind the person to only say the answer.

For example, if the answer was “Brother Zerb” and I say “I believe it was Brother Zerb or General Klao” it would reject that as the correct answer because it contains more than just the answer. But, you can whitelist some words (ok, a lot of words) so that it only check for the keywords and ignores white-words when rejecting. Then “I think it was Brother Zerb” would pass, as the answer is there and “I think it was” are all considered just normal dialogue words. But “It was Zerb and Brother Klao” would be wrong, because theres that extra non-word “Klao”. (I made those names by hitting my keyboard, btw).

Besides that, there is also the ability to let the player talk the way we want to.

If I don’t like this NPC, but I have to ask them about “the money,” I shouldn’t be restricted to “Where is the money?” or something like that. I should be able to say “Fork over the money, you son of an owlbear” or whatever.

As for the Fable system… thats one reason I really liked that game… everything had a meaning.

Some solutions being: Icons, generic words that can be selected from a drop-down when designing a quest, or generic words that the designer writes (e.g. “Ask %NPCNAME% for more information”).

The way I see it, there are almost zero positives to writing out player responses in a massively multiplayer game. Instead, it’s a list of negatives.

On the Developer side:
- Designers have to actually write those responses.
- They have to be translated into other languages.
- The custom responses have to be stored somewhere.
- There’s greater opportunity for error.

On the Player side:
- It can be a barrier to the “fun” part (driving the conversation or getting to the quest).
- Players have to read enough of the response to understand where it will take the conversation, which can be confusing.
- It’s putting words into an avatar’s mouth, which is a bad idea in most MMOs.

The benefits being… Uhh… None? None that I can think of, anyway. Writing out responses for a player to select from is a great idea in a single player game in which the player is controlling a character with a backstory, attitude, etc. that you define. In an MMO where players are supposed to be writing their own story with their own character, that’s a bad thing.