More

EDIT: Here's the benchmark.

Every level, mages should gain 1-3 spells depending on their usefulness (1 great spell, 2 good spells, 3 eh spells). Likewise, equipment and magic items should be slowly ramped up so that characters are always getting a little better. It'd be optimal to keep these things out-of-phase with leveling, so that occasionally, the PCs level and get better, and then at the mid-point to next level, craft a new spell and get some better armor.

...

More on spellcraft.

Dude, I think we should just spell out the difficulties with this skill for DMs. You were planning on writing a 'DM's guide' or something, yes? Let me just write an essay for it on what I've learned from years of GMing.

Basically, no matter what system we offer for Spellcraft, we won't get it just right for every world. The spellbook of a mage is the number one thing in that character's life. Herm/Hedge-primary characters are defined by what's in the book, and the DM has absolute control over that, i.e., control over the character's usefulness. Directly, by providing or not providing spells in the course of a campaign, and indirectly by limiting time and money for spell creation costs.

DMs should feel free to make some spells easier, more difficult, or even impossible to make.

This is because a single spell has the potential to drive the campaign. Not only drive the PCs personal quest to make it, but a single casting of a given spell can be the climax of the entire saga. Sarpagal II, it was Anti-Magic Shell. Subquests in the campaign are the same, usually trying to find someone to cast Resurrection or Wish to bring back a dead PC.

(There are analogues of this campaign driving: Narbohring II it was the quest to find the Blackstaff, used once only to destroy itself and take out all of Barrowsreich. Melee-primary types also have character-driven quests, limited by the DM, to get better magic weapons, and also in that way determine the climax of the campaign.)

Also, while priests just have to pray, and fighters can go find weapons on any dead town guard, if a mage loses the book, it's basically a weeks-long Hold Person spell. Which can be interesting the first time it happens, if the story works, but otherwise sucks.

Anyway, the point is that we should just tell DMs to take a heavy hand in crafting the progression of the Hermetic spellbook (Hedge magic is less important for this, which is why spells can be cheaper- the highest circle Hedge spells aren't really going to alter destiny, but 8th circle Herm spells certainly should be able to do that). In parallel, they should be sensitive to the mage's usefulness, relative to priests and fighters, whose effectiveness should also be subject to top-down control.

(But these changes to the system we'll offer should be spelled out early, so that the players don't get discouraged. It's fine to say in the beginning, these spells are 'known at the Guild', and 'these spells are lost to Legend', but to suddenly say 'oh Fireball actually is twice as expensive' when they try to make it is unfair.)

Something to that effect.

...

Also, monster design bores me. But when I've got a moment, I'll write up a mini-campaign by way of example, based on our table top experiences, and design some magic trinkets too. I want to encourage Hedge and Ritual mages, and encourage the use of CHA skills. Think those two things can go together with some magic trinkets.