In parlour LARPs (the types that Sydney and Melbourne conventions called freeforms, much to the confusion of the rest of the world when we finally became connected by the web), magic was often ceremonial, dramatic and integral to the storyline. Fast casting may have happened through the revelation of an effect card to nearby players, but since games like this tended to be socially driven affairs where combat was frown upon in the narrative set up and every freeform tended to be written as a one-off event, there was nothing to compare it mechanically with anything else. Games like this didn't care about balance, in fact they'd actively flaunt it in many cases.
In boffer LARPs, things were different. People could play warriors, and occasionally people could play magic users. I remember reading through the pages and pages of arcane documentation that constituted boffer style LARPs back in the early to mid 90s... most seemed to consider Rolemaster too streamlined as a system, and ramped up the complexity from there. Magic was some of the most overly convoluted parts of these systems, and often didn't seem to do a whole lot anyway (outside of elaborate rituals that the players might be trying to stop...or similarly elaborate rituals by the good guys in an attempt to shift the balance of the meta-game). Other effects of magic might be long and time consuming, such as crafting mystic items.
It seemed to be the case in games like this that magic was actively discouraged unless it was specifically used for the purposes of GM-controlled storyline.
More recently (the early 2000s onwards), I started hearing things about "spell packets" and "spell balls" as away to take magic away from the elaborate ritualism of earlier eras, and more toward a mechanism that could compete with combat on the battlefield. The thing is that a lot of these systems were very GM intensive... certainly not something simple enough for players to self-regulate.
One of the things I like about "Australian Freeform" games is the fact that the storyline is front-loaded as a situation the players walk into with their character's motivations, certain immediate agendas to pursue, ad other agendas that kick in when specific events unfold... when one agenda is completed, that often works as a trigger effect setting off the agendas of other players. The situation changes, it tilts in a Fiasco sense, and alliances/enmities/events that came before may no longer be relevant (or they may be the only thing that;s important anymore). During the course of play GMs don't prompt storyline in this style of play, they simply act as facilitators answering questions, confirming how things might unfold and co-ordinating the negotiation process between players when an outcome isn't obvious.
One of the things I liked at first with the "Clans of Elgardt" game was the fact that the combat was played out in real time through Boffer tactics, the storylines were partially pre-loaded by GMs, and there was an intended overall narrative, but all the details were filled in by the players with their own agendas. During the course of play GMs didn't need to micromanage actions, they didn't need to describe the events of the world or the NPCs encountered...in fact there were enough people and enough things going on that there felt like a world of events happening. No-one knew the whole story, and as soon as you answered one question you'd discover that there were two more along the way hat now needed answering. The last faction to shift the direction of the game didn't seem to understand this concept, and pulled their branch of the game back to the railroaded, GM-centric, gameplay of the past...like many elements of the OSR it felt regressive. It didn't help that my attempts to revamp the magic system were met with resistance by significant members of that branch.
This has all been a bit messy so far, a meandering through various issues, but they all basically link back into the ideas I'd like to see for magic in a LARP. Combat can be handled by boffer rules, particularly a trait/ability driven boffer system where experienced characters earn keywords that can be called in the heat of battle to invoke effects if the hit connects. Social effects can be handled with a combination of natural discussion, with ideas like in-game status pulled across from Minds Eye Theatre (if you are both a part of a group and that player has more status within the group, you have to listen to them and accept their request, suffer status loss within the group, or completely separate from the faction because you don't accept their position). Magic needs to be something more though. I like my magic to be equal parts mysterious and manageable. I accept that not all players will want to use magic as a part of their character concepts, but I'd like there to be a degree of balance for the players who do use it. They shouldn't feel like they are wasting their time in the pursuit of the arcane while they remain completely ineffective on the field, nor should they be obliterating the warriors on the field with "lightning bolts".
To those ends I'm looking at a series of "edges" that characters earn as they gain experience.
Combat edges give a player more weapons (or armour types) to play with, or provide more hit points.
Influence edges give a player more ways to socially manipulate the events of the storyline or other characters.
Knowledge edges give more background information and allow characters to build things.
Magic edges give players access to a school of magic, or an elemental affinity, which in turn open up specific spells.
In ascending order...
Magic might be slow, in the form of rituals. It might be faster, in the form of spells requiring a laying of hands for 5 seconds. It might be instantaneous, where a glancing blow passes on an effect. It might be ranged, and manifest in the form of a spell ball.
In most cases magic will follow established patterns, but those patterns may replicate other effects, or might be completely different to the way other effects in the game work. The more exotic (or the more obviously powerful), the harder the effect is to learn.
I need to think more about this part of the game system I'm working on, because everything else has basically been refined through 18 months of trial and error. I also need to make sure the magic system under development has some kind of internal narrative consistency.
In Elgardt, you buy spells with experience, then generally get a number of spell uses equal to your "Soul" attribute (which caps out at 5), so you never get the issue in the video where a spellcaster seems to have an unlimited number of lightning bolts. To make things a bit different, at this stage I'm working with the notion of magic crystals, which may be carried in a bag and used for spontaneous casting, implanted into items with pre-defined effects, ground up and infused as potions or delivered intravenous to create magical mutations, and other effects that have been mastered by different magic schools. A character may have any number of crystals, but they are unstable and have a half-life (literally at the end of every session, you halve the number of usable crystals in your possession, and you're constantly on the lookout for more, or needing to purchase more if you want to sustain your magic levels). Crystals are exhausted to invoke magical effects while away from a place of power (a dry erase mark is wiped off the laminated crystal card to show that it has been used). While at a place of power, or in the presence of a spirit (ie. a GM/game admin), a character may refresh a crystal, or channel energy through a crystal and into their essence to gain a spell ball effect. Needs more thought.
More information to come.