I'm in favor of nearly anything that allows players more choice. Frankly, there simply isn't a legitimate reason not to permit changing AC's that isn't based on an inappropriate conflation of social considerations with game mechanics.
The most frequently cited and reasonable-sounding objection goes something like this: "If you can switch AC's at level 50, I'll be forced to group with 50's who don't know how to play their classes." This begs the questions, first, who is forcing you to group with any individual in the first place, and second, whether the absence of AC changes protects you from playing with bad pugs (hint: it doesn't). To the first, flatly, you can decide not to group with someone if you don't want to. If someone's going to lie about how new he is to his class, then he would probably lie about other things to maintain a spot in the party under other circumstances. Are we really under the illusion that there's no scrutiny when it comes to making recruiting decisions to guilds, raids, and high-end parties? To the second, on the off chance that in your haste to get a pug rolling you find yourself saddled with a recent AC convert who indeed is no good, is the task of ejecting and replacing that individual so daunting? If so, how is it any different from the task with respect to someone who's just a bad player in general? I'll grant that the frequency of these circumstances may increase, but I would likewise suggest that if that's your primary concern, you have nothing to lose by seeing a mechanism implemented and tested on the appropriate test server in order to provide data with which to actually validate that claim.
Regrettably, most of the other objections are rooted in an idealization of MMO culture that's either outdated ("Classes are rigid, hard-locked playstyles with no overlap whatsoever, and someone who can play Class X is completely incompetent and disqualified with respect to Class Y") or elitist ("Convenience is the devil's work").
I'm frankly not interested in patronizing maxims about decisions having consequences. You're not arguing that someone who impregnated his girlfriend should be a father to his child; you're arguing that someone who clicked a box on a video game--purchased for recreation and being paid for monthly--and then at some indeterminate future time decided it was a mistake, should have to repeat anywhere from five to hundreds of hours of the game in order to correct that error, or else no longer enjoy the game or quit playing outright. I have a job and a family if I want consequences to go with my decisions. When I sit down at my computer in my valuable free time, I'm not looking for consequences: In-game, roleplaying consequences that tell a story? Sure; hell, that's part of what makes SWTOR attractive. Waste-my-personal-recreation-time consequences? No thanks; I'll pass on that.
Moreover, I'm not interested in the equivalent of "f- you" or "go to hell" thinly disguised as argument or solutions. Many of us did not have the advantage of playing many hours during the beta, and many of us don't have the time to devote--or the interest, in many instances--to leveling up multiple characters substantially. We're dependent on the community at large to inform us how a given advanced class performs at a given level.
For example, I understand that a Madness Assassin is a very late-blooming spec. On paper, it looks like something I'd be very interested in. However, I have no way of knowing for myself how the class will perform until I get my Assassin up to level 40 or so. If I find, at that point, that despite their best efforts other players haven't been able to convey to me accurately what the class is like, I may find myself extremely dissatisfied, and at that point regret my decision to play an Assassin at all (since, based on the same information, it looks like the Madness Sorcerer is a lot closer to what I was looking for). By way of example, the ability to change my AC at this point is ideal, merited, and an excellent way for BW to keep me engaged as a player and happy as a customer.
Many of the responses here amount to dressed up equivalents of "Well, sorry you chose wrong; f- you. Reroll." Is that fair? Maybe. Indeed, the weight of MMO experience tends to inform that investing hours of gameplay into a character doesn't guarantee future satisfaction in that character by any means. But it's not right. There is a fundamental inaccuracy in the analogies to WOW, EQ, or any other previous MMO, and that's the product of a failure to consider the class story--which is the fundamental hallmark feature of this game. The story is why I'm playing at all. Asking someone to reroll a different class in WOW is less consequential because, in essence, all classes share the same story in WOW. There's no real unbalance in the consequences. Asking someone to reroll a different AC in TOR, on the other hand, is essentially different from rerolling a different class outright. You're asking the player to repeat identical content that he would not repeat with a genuine alt.
Does allowing someone to change AC's beyond level 10 have risks? Sure. Moreover, it likewise opens up a door for player freedom and customization that is unusual for MMO's, although not without precedent (Rift being the premiere precedent). But the arguments I see here against it typically amount to glorifications of "I don't like this." If your argument to not have AC changes amounts to nothing other than you think someone should be penalized for making a bad choice with a lack of information, there's simply not much there. If you're worried about social consequences--well, you may be on to something. But I would submit that the risk that those consequences might materialize--if they do at all--is vastly and overwhelmingly outweighed by the benefit that this feature would confer on players who could take advantage of it.
The current AC structure in SWTOR is designed to maintain traditional MMO party composition mechanics while attempting to insure that necessary party roles are likely to be available when needed, at times to the displeasure of some parts of the community who, for various reasons, believe that, for example, certain classes shouldn't heal or tank. Allowing AC changes would fulfill the same end of making sure that players could assemble a party with the requisite roles filled without simultaneously depriving a player of his ability to complete other content or, more importantly, experience the story--the lynchpin of the game's experience--in the way he or she wants and without undue duplication, repetition, and grind. The fact that the developers have expressed a desire for this feature to have a cost attached to it, if anything, reflects a middle ground between the stances expressed in this thread. While I disagree that that's necessary, I'm willing to defer.
Ultimately, this is about choice. Prohibiting a player from making a choice with the benefit of the most possibly available information makes the choice less meaningful, and the argument that players should be forever locked into a class at level 10 gains more traction from gaming tradition than it does from common sense. If I change, for example, from a Powertech to a Mercenary at level 45, you are correct that there is a learning curve ahead of me, and that I have a long way to go to mastering my new class. That fact, however, does not alone make it a bad feature. Would it tend to suggest additional features? Sure. I can see that tutorials would be wise, and allowing other players to see how much total experience (within certain parameters, lest it become a server first pissing contest) a given player has logged in his current AC would be a perfectly fair indication of whether that player knows what he or she is doing, and I would likewise advocate that people forming groups should have the benefit of all available information when making choices also. In-story, there's no reason that a Sith who has mastered defense could not upon study and reflection learn when and how to use two lightsabers, for example.
Give players the ability to make choices and gain the benefit of those choices and the power to correct errors. In fact, I would even advocate that players should be able to change AC's at low cost and, perhaps not at will, but as frequently as their circumstances should require. It appears, from the bulk of the arguments here, that the only thing we really have to lose is the satisfaction of those members of the community who have a sense of entitlement with respect to how others experience the game. That's a price I'm willing to pay.
The Problem I have a deeply impressed knee-jerk reaction to Recount, Gearscore, and their related ilk because I feel they represent the edge of the slippery slope toward the mathematization of relationships. A system that provides detailed relative data among participants in content is less useful for evaluating one's own performance, and more useful for judging the performance of others. In a sense, Recount is the standard bearer for heavy-handed guild dictatorships and exclusivism. It is the refuge of the min-maxer and a menace to the targets of his ire. That said, I will gladly concede that there may exist in the community a subset of players who value data for data's sake--those players who take no offense to the heavy beast breathing over their shoulders, and embrace the risk of abuse the same way one embraces Vegas slots, with the outstretched hope that perhaps this time things will be different. The Solution For those players I would like to propose the following: Do not allow recount. Allow, instead, absolute, complete data transparency. Upon initiation by any group or raid member (with a cooldown for spam control), the party members may vote anonymously to enable data parsing. If parsing is enabled, all players have access to a complete, unabridged record of the group's/raid's activities. To be more blunt, this is a keylogger, and it reports every action taken by a character in-game, with the exception of recording chat as "[timestamp] Entered chat until [timestamp]." It reports actions, cooldowns, player movement (coordinate by coordinate, including facing and rate of turning), button mashing, interface actions, and millisecond-to-millisecond reports of states, statistics, incoming and outgoing damage, healing, and effects. Moreover, it should report the technical calculations involved in getting from base numbers to final numbers. Thus, you can see how much someone would do in base damage and how much the damage is modified by gear. Additionally, you can see the odds of a given occurrence whenever there's interaction between an effect and an enemy, so you can gauge whether a player is performing poorly by his own fault or just unlucky. You can also see if a player is extremely efficient and just undergeared, or performing poorly but being carried by his or her gear. You want data? You got it. All of it. Everything the server knows about what a player is doing during the raid, the system makes available. The Reasoning Why all this information? Why not? You want to data mine; have at it. The data is the key to a successful end-game guild. I advocate for this because Recount just doesn't go far enough. It goes far, by MMO standards, but it's ultimately just organizing data the server already reports. It's a tool for judging player performance. If you're getting into the business of judging others, though, why not be more thorough? Would you be shocked to find out that your raid leader never touches his mouse in a five-minute encounter? What if your tank is just coasting along in gear three tiers higher than the content and is only doing the bare minimum to keep aggro (and you were ************ at the sorc for not ramping his dps!)? Maybe you'll discover that someone missed an interrupt because it was on cooldown from the last interrupt--just like he's been telling you for the last six times you ran that fight. Maybe you'll learn that that whore the GM's been cybering is spending more time talking to people during the raid than she is contributing? Who knows what remarkable secrets you might uncover! It is only then, once you have had the opportunity to examine your colleagues with the most powerful microscope possible and comb through their errors with only the finest instrument, that you can accurately assess whose fault it is that you're not succeeding where you know you should. Only then can you assist them in improving themselves. Know that the other members of your guild may fear such a thing, and I trust many of you receive my suggestion with some trepidation at first blush. Once you embrace the power that comes with this knowledge, however, you will find that your guild will succeed and thrive as an organic machine, stripped of the trappings of trust and relationships. This is a computer game, and so it is driven by data, by numbers. The numbers are what matter, and no one has anything to fear who is giving proper dedication to his role in the guild. Often people laugh when someone says poopsock. Those people aren't cut out for end-game content. Your guild members may be angry with you if you suggest that such a thorough, robust system be implemented. Some of them will no doubt become furious at the notion that you or anyone else can prowl through the minutiae of their online behavior during raid time. Remind them that that time is sacrosanct, and it is a sacrifice they must be willing to make to the guild. If they cannot make it, then cast them into darkness. Once they have been cast out and been deprived of life-sustaining end-game epics, they will return, humbled, and submit. Your guild will hate RecountRedux. They will despise it as its claws hook themselves into their petty lives. But it will unite them, unite you as a guild, and make you stronger. You must channel the very reasonable repugnance at the idea that game performance Big Brother is the gateway to success, harness it, and make that repugnance serve you. When your guild begins to squabble, and members fight among themselves over the truth revealed by this system, you have nearly arrived. Direct their fury away from their destroyed relationships and to the unliving, inhuman game content, where it should be. In the end you may find yourself and your guild miserable shells of who you once set out to become, but you will be shells wearing gear with spectacular stats. And that's what matters, isn't it? TLDR: Recount is the path to the Dark Side. And not the cool storytelling dark side. The guild-imploding, enemy-making one that makes the game antithetical to fun. If you think someone is under-performing, ask. Maybe talk with your raid before and after encounters and find out why things are or aren't working. No one needs a blue ribbon for top DPS or a brown one for worst healer. Build a guild with a raid force composed of people you know and can trust. It might mean that you have to repeat encounters a few more times before you get them down. But you won't end up hating your guild afterwards. That's all.