| >>I'd argue the fact TM is in a good spot.|
Yeah, but it's trainable, which is more than can be said about aug, ward, and util right now. I'm not saying we're done with TM, just that adding new stuff to RESEARCH and the Symbiotic Magic system to support TM isn't something we plan to do right now.
>>we should be able to wield them for exp throughout our careers.
But against what difficulty? With weapons, you're constantly facing more challenging opponents. There are no such opponents in the Magic system aside from mana (and now Symbioses), which is why spells are neither creatures nor weapons, but something in between.
>>I see no good design reason for the spell level learning restriction, other than a slight nod to realism.
It's about maintaining the skill training paradigm we have throughout the game. The reason a spell stops teaching is because you can stop adding more mana to it, not because of any other systemic reasons. When you've sufficiently surpassed the greatest challenge that spell offers you, you no longer learn anything from it.
The magic system is built on a foundation where you get the maximum effect from a spell once you've capped the mana in it. Additionally, any capped spell is as good as any other capped spell - An intro spell that you're capping will give you the same strength of buff as an esoteric spell giving that same buff. The Esoteric spell will give OTHER buffs too, but in general once you've got 100% of the spell's mana, you get 100% of the spell's effect.
So, if we extend the training range of a spell to, say, 1-1600 ranks, that means that that spell is only ever getting capped by the most skilled players. Characters can learn from 1-1600 with that spell, sure, but it's also going to be worse for them as a spell when talking about the amount of buff to give.
Changing these various pillars isn't something we can do easily at all, and would require a Magic 4.0 level of release. That's just plain not something we want to do right now, for obvious reasons.
So, given the magic framework we have today, a single spell has to stop training at some point. Symbioses are, in some ways, a loophole that we can use to make your basic spell (ES, for instance) more complex (and thus training to higher ranges) without altering the core principle that ES only trains within a certain range on its own.
>>Lower the exp if you have to, but don't increase the difficulty.
That's where the cyclic stuff seems to be landing, both from a mechanics perspective and also a player preference perspective.
>>give me reasons to be excited about choosing spells based on what they DO, not be sadface about choosing them based on their prereqs and how they train.
Agreed. I hope that between RESEARCH and Symbiotic Magic, we're getting to a place where spell choice CAN be a lot more about what you want rather than what you need to train.
>>Spells with extra management should cost 1 slot less.
This is already somewhat included in our calculations, under the 'harder to use' bucket. It's a matter of degree, of course.
Prereqs serve a variety of purposes. The argument that the prereqs of a spell can force you to take spells that you don't want is a valid one, but the lore is pretty important too. This is one of the cases where the story trumps the ease of use. Prereqs, in concept, say that 'A spellcaster must know how to cast this more-basic version of a similar spell before even attempting this more complex version'. There's a lot of flavor there, and a lot of distinction. One thing we've seen over the last decade or so is a lot of 'vanillification' - blurring of the uniqueness of each guild and the fiction surrounding them. Removing spell prereqs in favor of 'Must have N spells of type X' would be a disservice to a the fiction around the game world.
[This is a discussion in and of itself, if we want to drill deep into it let's take it to a different thread]