There's nothing a developer can do to eliminate the tendency within the community to demand optimal setups.
Optimization stem from restrictions.
It is not viable due to exp penalties to make groups larger than 6, so you try and squeeze the best things into that small group. Remove exp penalties and you'd see groups of 18, and I'm fairly sure people wouldn't be as picky as to demand 3 BRDs and a horde of WARs for their alliance.
Won't happen. It would be too much work for the developers to create dynamically tuned content suited to groups so vastly varied in size.
Guildleves will most likely have limits on the number of players in a specific instance at one time. I don't expect to see people standing around in major cities spamming for open air progression parties if guildleves are where the bulk of the character progression is done. Other developers have pondered the idea of removing penalties and/or restrictions on group sizes for non-instanced content and realized that it would be impossible to tune. If you have content tuned to be suitable for smaller groups...say 2-3 players...and then make it so that you can be in a group of 18 and do the same content without penalty, the standard will be to arrive in a given zone, join one of the in-progress alliances, blast through the content and then turn around and ***** at the developer that there's not enough to do.
If you make it so that everything can be done in a full alliance without penalty, you'll have to put up with the headache of a full alliance for virtually everything. Maybe FFXI changed over the years, but when I played anything involving 18 people was almost guaranteed to add at least an hour of pointless thumb twiddling to the process while so-and-so warps back to get <item they forgot> prompting so-and-so to get bored and go afk for 10 minutes starting the perpetual leapfrogging of afks resulting in a great deal of anxiety and frustration over nothing.
The only way around all that was through organized shells and/or weekly scheduled events, and even then I was part of more than one scheduled event by an organized group that was scheduled for one time and started 60-90 minutes late because there were just too many unreliable people.
Killing faster is better. If this wasn't true, then the need for BRDs and WARs would lower. To change this you could implement time based zones that sort of send waves of enemies at you, and every hour you stay alive, you get a set exp bonus. (This would of course on the other hand put more focus on healers and sleepers). You could also have an activity (besieged) that lasts a set time (1 hour) and always gives a decent amount of exp for participating in it (15k exp). You could have quests that set you out as a choco scout and for the 15 min the choco lasts you get 5k exp if you have at least reached another region.
That's an FFXI mindset, and my hope will be that SE is doing more than just looking at what was going on in FFXI and trying to tweak. My hope is that they'll be looking at the gameplay experience overall and trying to tune it in such a way that players don't feel compelled to run with only optimal setups because the game is too tedious and painful to play any other way. My hope is that they won't be tuning encounters so that any one specific build is required at any time. I'm not opposed to variety that gives various different builds an opportunity to shine here and there, but not in such a way that players adopt the global impression that if you don't have <x> archetype with <y> build, it's not worth attempting.
You could have gimmicky HNM behavior, like a HNM afraid of pets, but have to be nuked down. So you have to have BSTs, PUPs, SMNs near the BLMs and assault it when it runs for BLMs, but retreat it so that the HNM doesn't run out of nuke range.
And then what happens to the BST, PUP, and SMNs when the encounter is over? When you tune encounters such that they are unmanageable without specific builds, you create barriers to the content. It's not ideal. It fosters resentment in the community. Part of that resentment is directed at other players, and part of it is directed at the developer. What happens if your particular group doesn't have a pet class? Do they recruit to fill a spot for that particular encounter and find themselves with a roster bloated to account for gimmicks, or do they focus on building a group of people that can work well together, get along, and share common goals and then head out to play and enjoy the game?
Anything goes, as long as you emphasize the strengths of different jobs so they actually are needed. Another example could be to have undead, that are immune to damage unless a WHM has cast raise 3 on it the last 3 min. Nobody but a WHM can un-weaken it so a WHM becomes vital for it.
Using your example, again, in an attempt to ensure that every class is viable you encourage gimmicks that force
players to have that particular build in their group. And if they can't find someone of that particular build, that encounter is inaccessible. Why not just tune each particular build so that if they progress their character intelligently, the game mechanics will allow them to contribute to any
encounter in a way that people are happy to have them along?
It helps if you can make up your mind and decide: do you want to have encounters that require specific "classes" or do you not? You say you don't want to see things like BRD and WAR dominance in groups, yet in the same post you just shift that mandatory component to specific encounters with other classes. That's just bad design.
Make all archetypes viable with the only limitations being the way in which a player chooses to develop them. If you play with consideration to what you bring to groups and direct your progression efforts based on that, you'll be viable. That's it. That's all you need. You don't need developers reaching down with the hamfisted hand of dog and saying, "Ya, we made it so that the build that resembles WHM actually sucks compared to other healing/support classes, so rather than retune that build we're just going to throw in encounters that force you to bring that build that sucks and that way everyone is happy. Except the ones who play the class that sucks, because all they ever get invited to do are the encounters that we tuned so that they were required. At least, they used to get invites until people realized they could just progress their own characters that aren't classes that suck to get access to the key abilities of the class that sucks and add it into their rotation for that particular fight and all of a sudden the class that sucks but was needed is back to the class that sucks and isn't needed when all we really needed to do was not be moronic developers and fix problems with balance instead of trying to ram the imbalance down the players' throats."
SE has no excuses this time around. They can't play the novice developer card and expect any sympathy. They need to get it right, and that will start by making sure the game mechanics are inclusive and all archetypes are viable for all encounters. Edited, Nov 2nd 2009 8:56am by AureliusSir