The Healer's Turn
A deep dive into Shroud of the Avatar's Life magic, what the genre learned about healing from Ultima Online onward, and twenty-four concrete proposals.
Life is the school this game trusts you with when someone else is dying, and it has quietly become the school nobody talks about. In eighteen months of community threads there is not a single request to improve it — because there is no healer role to have opinions about, only a splash skill everyone carries. The engine under Life is genuinely strong: the deck system it never exploits, ground fields it never places, wards it never casts, and an entire authored spell sitting dormant in the data. This document measures the school against the genre that invented healing, and proposes twenty-four changes built on one deal: mend the healer's hands first, then teach the school some new verbs, and decide in the open what a Life mage is.
Ultima Online shipped the ancestor in 1997: healing anyone, anywhere, with a reputation price, and resurrection as a service worth walking for. EverQuest made the cleric socially indispensable — its 96% resurrection handed people back hours of their lives — and accidentally proved that a heal which ignores the deficit deletes the triage game. Dark Age of Camelot priced heals in cast time and taught the spreadheal. WoW industrialized triage with layered HoTs, typed dispels, and mana pressure. Guild Wars 1 showed that preventing the hit outranks curing it, and that a small skill bar forces heals to differ. FF14 made routine healing charge a payoff you choose the moment for. City of Heroes proved the most beloved supports barely heal at all — they move every resource except hit points. The genre's verdict is consistent: a healing school is interesting exactly in proportion to the verbs it has beyond restore.
Measured against that bar, SotA's Life magic has a strong skeleton and one verb. The community's silence says it plainly: in roughly two hundred wishlist and balance threads over eighteen months, not one asks for anything from Life magic — the school predicted to make "everyone a healer" a decade ago instead became a splash everyone carries and nobody plays. The gaps cluster in three places:
The deal that runs through all twenty-four proposals: Tier 1 mends the healer's hands, in the open, with the broken code named. Tier 2 teaches the school the verbs the genre proved out — spending currencies the deck already charges: stacks, combos, draws, and focus. Tier 3 lights what was already built and asks the community the questions we cannot answer alone. No new verbs on broken hands, and no tenth heal. Nothing here makes healing bigger before it makes healing legible, and nothing spends new engine work where authored data will do.
| # | Proposal | Tier | Size | Server | One-line pitch |
|---|---|---|---|---|---|
| S1 | The area heal joins the attunement economy | 1 | S | none | the one heal that ignores your specialization stops ignoring it |
| S2 | Purify's list, cleaned and published | 1 | S | none | the cleanse table has a duplicate and a stray entry; fix it, make it data, print it |
| S3 | The healer's hands | 1 | S | none | ally cycling works, self-cast rebinds, the in-range color comes back to life |
| S4 | Numbers you can read | 1 | S | none | big heals read big, and overheal stops vanishing silently |
| S5 | Resurrection states its odds | 1 | S | none | the cast bar shows rez chance the way it already shows taming chance |
| S6 | Every spell looks like itself | 1 | M | none | orphaned effects rewired, shared casts differentiated, the missing fizzle added |
| S7 | Overheal becomes a ward | 2 | M | none | a slice of surplus healing lingers as a small, capped absorb |
| S8 | Sanctuary: the prevention line | 2 | M | enum | two glyphs that stop damage before it lands, on plumbing that already exists |
| S9 | Three of a kind becomes a new spell | 2 | L | none | stacking three copies transforms a heal's kind, not just its size |
| S10 | Grace in bloom: the first Life and Life combo | 2 | M | enum | Grace primes, Blast detonates; Life finally combos with itself |
| S11 | Life joins the combo table | 2 | M | enum | three cross-school combos where Life is the anchor, not the guest |
| S12 | Bloom: heals bank a payoff | 2 | M | enum | every heal banks a charge; a spender glyph turns three into the big moment |
| S13 | Purify pays out | 2 | M | none | cleansing heals per effect removed and converts what it strips into a blessing |
| S14 | Mend the line: the triage heal | 2 | M | enum | one glyph splits a pool across the party, weighted to the most injured |
| S15 | The Ankh returns | 2 | L | none | the authored, never-shipped Summon Ankh becomes the placeable healing totem |
| S16 | Come back whole: resurrection tiers | 2 | M | decay | higher skill restores a share of death decay; the great healer is worth finding |
| S17 | Radiant reprisal: the smite hybrid | 2 | M-L | enum | a small holy line whose hits splash minor healing to the party |
| S18 | Grace of the giver | 2 | S-M | enum | a new innate: every Life glyph cast on an ally also heals a little |
| S19 | A healer band for the balance targets | 3 | S | none | the balance targets get their first support band; everything above tunes against it |
| S20 | Healing beyond the party | 3 | S | open question | one flag gates all cross-party healing; both sides argued, your vote counts |
| S21 | Mending the townsfolk | 3 | M | none | friendly NPCs become healable; the guard beside you stops being untouchable |
| S22 | The healer's HUD | 3 | M-L | none | attunement on screen, HoT marks on party rows, healing threat made visible |
| S23 | The pacifism tax | 3 | S | open question | the school's signature bonus rewards not acting; keep it, floor it, or invert it |
| S24 | HolyStrike and HolyJudgement | 3 | S | open question | two authored stubs ask: pure support forever, or a spine? |
The pieces in place are real: nine castable spells and five passives in a clean prerequisite tree, a heal-over-time with a built-in bleed cleanse, an area cleanse that also grants death resistance, an anti-undead cone, a resurrection whose reagents go free at grandmaster, and a deck system that lets a healer stack identical glyphs for a bigger cast — heals stack at +25% per copy, better than any damage school gets. Underneath it, the combat engine ships chained effects, ground fields, wards, and child spells that other schools use every day.
Every Life cast runs the same road, and two of its stages are where a healer's intent dies today:
flowchart LR A["Glyph drawn\nrandom from deck\nor locked slot"] --> B["Stack / combine\n+25% per copy\nno Life-owned combo"] B --> C["Target resolution\nself wins unless an\nally is hard-locked"] C --> D["Cast\ninterruptible\nno success readout"] D --> E["Formula\nattunement x effectiveness\nx time since last damage"] E --> F["The heal lands\nfixed-size float text"] F --> G["Overheal\ndropped silently\nno feedback, no capture"] style C stroke:#a33d2e,stroke-width:2.5px style G stroke:#a33d2e,stroke-width:2.5px style F stroke:#c9a55a,stroke-width:2px
Stage three is the quiet killer: a beneficial cast defaults to yourself unless an ally is explicitly locked, and the lock tools are the broken keybind and a click. Stage seven throws away every point past full without telling anyone — in a school whose stacking bonus practically guarantees overshooting. The Tier 1 cards operate on exactly these stages, in the open.
Healing is the genre's oldest support job, and its design history is unusually well documented — every era found a different answer to the same question: what does the healer do between emergencies? The short version: the games that answered "nothing" lost their healers, and the games players still praise gave healing more verbs than restore. A companion review of the Chaos school established the method this page follows — measure the tree against the engine, then against the genre, then propose mostly data. Chaos was over-tuned and under-built. Life is the opposite: under-expressed and un-tuned. It has never appeared in the balance change log at all.
Ultima Online: cross-healing as social fabric, healing as a moral act with a reputation price, resurrection as a world service, and a 25-year retrofit arc that still ended at HoTs, absorbs, and auras. EverQuest: the resurrection that handed back hours of lost experience made the cleric socially indispensable; Complete Heal proved a deficit-blind heal deletes the triage game. DAoC: cast-time versus instant is a deeper axis than big versus small, and the spreadheal showed a spell can carry targeting intelligence. WoW: layered HoTs made healing proactive gardening, typed dispels kept it a knowledge test, and its overheal meters built a decade of healer skill culture. Guild Wars 1: prevention outranks cure; Divine Favor unified a whole class by making every ally-cast also heal — the direct model for S18. FF14: lilies made routine healing charge a payoff — the direct model for S12 — and its healer-damage debate is the cautionary tale S23 exists to avoid. City of Heroes: the strongest support sets never healed at all; routing support through the enemy — the model for S17 — put the support player in the same game as everyone else.
| Capability | UO | EQ | DAoC | WoW | GW1 | FF14 | CoH | SotA today | SotA proposed |
|---|---|---|---|---|---|---|---|---|---|
| Prevention (barriers, caps, next-hit guards) | ◐ late | ◐ | ◐ | ● | ● | ● | ● | ○ | ● S7/S8 |
| Conversion (damage to healing and back) | ○ | ○ | ○ | ● | ● | ● | ● | ○ | ● S7/S17 |
| Banked resource (heals charge a payoff) | ○ | ○ | ○ | ◐ | ○ | ● | ○ | ○ | ● S12 |
| Triage tooling (deficit-aware distribution) | ○ | ○ | ● | ● | ◐ | ◐ | ○ | ○ | ● S14 |
| HoT depth (stacking, priming, detonation) | ◐ late | ◐ | ◐ | ● | ◐ | ● | ○ | ◐ one HoT | ● S9/S10 |
| Cleanse gameplay (typed, with payoffs) | ◐ cure | ● | ◐ | ● | ● | ◐ flat | ◐ | ◐ flat | ● S2/S13 |
| Resurrection depth (tiers, restoration) | ◐ service | ● 96% | ◐ | ◐ | ◐ | ◐ | ◐ | ◐ hidden odds | ● S5/S16 |
| Utility beyond hit points | ◐ | ● | ● | ◐ | ● | ◐ | ● | ○ | ◐ S11/S16 |
| Ground and zone healing | ○ | ○ | ○ | ● | ◐ | ● | ◐ | ○ | ● S15 |
| A solo loop inside the school | ● | ○ | ◐ smite | ◐ | ◐ | ◐ | ● | ○ | ● S17 |
| Feedback legibility (waste and weight visible) | ◐ | ◐ | ◐ | ● | ◐ | ● | ◐ | ○ | ● S4/S5/S22 |
| Cross-target reach (beyond the party) | ● | ● | ● | ● | ○ instanced | ◐ | ● | ○ one flag | ◐ S20/S21 open |
● native · ◐ partial or era-limited · ○ absent. "SotA today" = the current development branch, 2026-07. S20 is deliberately marked open: cross-party healing is a stated question for the community, not a promise.
Tier 1 repairs the act of healing, published with the broken code named. Tier 2 teaches the school prevention, conversion, and utility — almost all of it data authored over effect machinery the engine already ships. Tier 3 is the identity work: the balance band, the HUD, and the questions this page deliberately leaves open for your vote. Every tuning number is an illustrative starting point: final values ship through the combat balance program's test-shard changesets and its weekly measurement windows.
Every single-target heal in the school scales with Life Attunement: invest in Life Magic Specialization and your Healing Touch, Ray, Grace, and Blast all grow. Healing Burst — the school's only group heal — does not. Its formula simply omits the attunement term, so the one spell a group healer leans on is the one spell your specialization never touches. The fix gives Burst the same attunement multiplier its siblings use, with the base trimmed in the same changeset so the median cast lands where it does today and the gain goes to the players who actually invested.
Seen in: every game with a coherent stat system; this is the classic spell-coefficient audit.
| Today | Proposed | |
|---|---|---|
| Healing Ray benefits from attunement | yes | yes |
| Healing Burst benefits from attunement | no | yes |
| Burst output, uninvested caster | 100% | ~85% (base trim) |
| Burst output, invested healer | 100% | ~120–130% |
Pure data: one formula edit in the Burst spell asset, shipped as a balance changeset so the expression regenerates through the real formula tokenizer. Attunement reads both the caster's and the target's stats, so the change gets the standard caster/target/observer live verification like every formula edit.
What Purify actually removes is a hardcoded list in the combat code, and the list has bugs: one entry appears twice, and one entry is a fire resistance stat that has no business being cleansed. Players cannot see the list, and designers cannot edit it without a code change. The fix deduplicates the table, removes the stray entry, moves the whole list into a data asset, and prints the categories in Purify's tooltip — so a healer finally knows what their cleanse answers.
Seen in: WoW's typed dispels made cleansing a knowledge game precisely because the categories were legible.
Small code change (the table moves from a constant array into a data asset) plus tooltip strings. The Fire magic page proposes the same dedupe from its side of the fence, and whether Purify keeps stripping fire effects is a cross-school rule the two pages decide together — one change ships, not two. This card is also the enabler for S13: typed payoffs need a typed, data-driven list.
Three input repairs in one card, because they are one experience: reaching the person you mean to heal. The cycle-friendly-targets key runs a loop that never advances its own index, so it retests the same ally and gives up — cycling through a party mid-fight simply does not work, while the hostile-cycle key beside it works fine. The hold-to-self-cast modifier is hardcoded to one key and cannot be rebound, unlike nearly everything else on the keyboard. And the targeting UI has a color authored for "friendly and in range" that never shows, because the code that computes range hands the display the out-of-range color unconditionally. Fix all three, and the basic loop of healing — pick an ally, know you can reach them, or snap to yourself — works the way the code always intended.
Seen in: every MMO with a healer; DAoC and EverQuest shipped functional ally cycling in 2001.
Three small client fixes, zero data, zero server. The ally-cycle repair can mirror the working hostile-cycle implementation line for line. The rebind goes through the existing keybinding registry. The range color needs the caller to pass real range state; the display model already branches on it.
Land a big hit and the damage number grows with the hit. Land a big heal and the number is always the same size: a 400-point Healing Blast reads exactly like a 12-point tick, and a heal that rounds to zero shows nothing at all. Meanwhile any healing above the target's maximum is silently discarded — the healer has no way to learn that half their cast was wasted. Proposed: heal text uses the same magnitude scaling damage text already has, tiny heals show a minimum readable value, and overheal renders as a dimmed remainder after the effective number, in the style "+142 (38)". You cannot make triage decisions with numbers you cannot see.
Seen in: WoW's overheal meters drove a decade of healer skill culture; FF14 dims capped values.
Display-only. The overheal value is already computed at the exact spot it gets discarded; it needs carrying into the combat-text event instead of dropping. The remainder display gets a combat-text option so group fights can turn it off. Deliberately ships before S7 decides what overheal is worth.
Resurrection can fail. Its success chance scales directly with your Life Attunement, so a character below full attunement can watch a four-second cast end in nothing — with no warning that failure was even possible. The casting UI already computes and displays a success chance for exactly one thing in the game: taming. Extend that display to Resurrection. A healer deciding whether to spend four rooted seconds mid-fight deserves the number.
Seen in: every modern game with a chance-based action shows the chance.
Generalize the cast bar's execution-chance branch from "is taming" to "the spell declares a chance expression", then declare it on Resurrection. The displayed number and the actual roll read the same inputs on the same machine, so they can never disagree.
Six of the school's spells share one charge-up effect. The biggest heal in the school wears the starting flash of a spell three tiers below it. Casting PurifyBurst on nothing plays no failure effect at all — the cast just ends. Meanwhile finished Life effects sit unused in the project: a HealingRay effect, a HealingTouch effect, a Purify ending, an authored lasting-heal glow that nothing references. This is a wiring pass, not an art commission: connect the orphans, give the spells that share a start effect their own tints and timing, dress the Ankh (S15) in Life instead of the Earth effects it currently borrows, and give PurifyBurst its fizzle. A school reads as finished when its spells are visually distinct.
Seen in: the community's own per-school spell color and effect threads.
Asset reference edits on the spell definitions plus tint and timing variants of existing particle effects — reuse with tint, never new emitters, so the particle budget in group fights holds. The one code check is whether PurifyBurst's fizzle hook fires at all.
Today any healing beyond full health is deleted. Proposed: a slice of the surplus lingers as a brief ward on the target that soaks the next hits — illustratively, 25% of overheal becomes a ward capped at 15% of the target's maximum health, lasting 6 seconds, refreshing but never stacking. The +25%-per-copy stacking bonus stops feeling wasted on a nearly-full ally, the pre-fight top-off becomes a real opening move, and the school gets its first taste of prevention without a single new glyph. The caps are the design: an uncapped absorb deletes the triage game for everyone, which is the oldest lesson prevention healing ever taught.
Seen in: WoW's Divine Aegis; the cap discipline is the post-2010 correction the whole genre adopted.
One conversion branch at the exact spot overheal is currently dropped, feeding the engine's authored-but-unused ward effect type. Verifying that dormant ward machinery actually absorbs is a Phase 0 item. S7 and S8 draw from one shared absorb budget and share a no-stacking rule, stated here so neither can double-dip the other.
The genre's most respected healing archetype plays in the future tense. Two new glyphs open that line for Life. Aegis of Grace (single ally, instant): the next hits on the target are absorbed up to a pool that scales with skill and attunement, for 8 seconds — cast it into the wind-up you can see coming. Sanctuary (party, centered on you): for a few seconds, each nearby ally's damage from any single hit is capped at an illustrative 20% of their maximum health — the giant swing becomes survivable, chip damage is untouched. Prevention is anticipation play: you read the enemy, not the health bars. Both work untrained; mastery widens the pool and the window, never switches them on.
Seen in: GW1's Protective Spirit and Aegis (the per-hit cap and the party block are direct homages), WoW's Power Word: Shield, FF14's shield healers.
If Phase 0 confirms the dormant ward machinery works, both glyphs are data: spell asset, tree row, buff channel, strings, icons. The per-hit cap variant may need a small mode flag on the ward handler — an extension, not a new effect type. The known failure mode is invisibility: prevention that works looks like nothing happened, so both glyphs ship with a distinct absorb-break effect and an absorbed-damage combat-text line.
Stacking identical glyphs is this game's signature combat move, and heals already stack harder than anything else: +25% per copy against damage's +15%. But the payoff is only ever a bigger number. Proposed: at three copies, each Life heal transforms into a different spell. Three Healing Touches become Mending Wave — heal the target, then splash to the two most injured allies beside them. Three Healing Rays become a beam that chains to a second wounded target. Three Graces become a party-wide regeneration. Three Bursts become the triage split (S14). The hold-or-spend decision every healer already makes at two copies becomes the most interesting choice in the deck: spend now for throughput, or hold for a different verb entirely.
Seen in: WoW's Lifebloom three-stack bloom; every card game where quantity changes kind.
The one genuinely new engine mechanic in this catalog: an optional transform reference on a spell, respected when the stack count reaches its threshold. After that hook exists once, every transform is ordinary spell data built from existing pieces (splash, chain, area regeneration), and other schools inherit the capability for free. Transforms are tuned as strict sidegrades against three separate casts, so holding is a choice, not a tax. The hand slot needs a visible tell when a stack reaches transform threshold.
Twenty-nine glyph combos exist in the game. Life anchors two, and both live in other schools' hands: one needs a Death rune, the other a Water rune. Life is the only school that cannot combine with itself. The first Life-and-Life combo fixes that with the genre's best two-card play: Blossom of Grace (Healing Grace + Healing Blast). Cast it on an ally carrying your Grace and the remaining regeneration blooms at once — the outstanding ticks land instantly, plus a burst that splashes reduced healing to allies nearby. Prime, then detonate. The heal-over-time stops being a background trickle and becomes half of a deliberate play, and the two most expensive heals in the school finally have a reason to share a deck.
Seen in: WoW's Lifebloom (letting it expire is sometimes correct), GW1's enchantment-consume skills.
Two honest stages. Version one is pure data: the combo output is a strong instant heal plus splash, using the same authoring pattern as all 29 existing combos. Version two adds the actual consume-remaining-ticks payoff, which needs a small effect handler (read the Grace effect's remaining ticks, cancel it, scale the payout). The break-even gets published: detonating with several ticks remaining wins on burst and loses on total healing.
Beyond the flagship, three cross-school combos put Life glyphs in other decks and other schools' glyphs in the healer's. Radiant Dawn (Healing Ray + a Sun rune): a beam that heals the ally it touches and dazzles enemies adjacent to them. Stoneward Blessing (Purify + an Earth rune): cleanse plus a short physical damage cap — the fighter-support handshake. Quickened Pulse (Healing Touch + an Air rune): an instant micro-heal that also restores a sliver of the target's focus, the school's first gift of a resource that is not health. Each combo is one output spell plus one data row; the whole slate is a week of authoring, and it triples the number of decks that want a Life card in them.
Seen in: this game's own combo economy, where one Chaos glyph anchors eight combos and Life anchors zero.
Pure data plus strings and icons. The focus-restore number on Quickened Pulse stays deliberately small: resource gifting is the genre's most loved and most abused support verb, and the mandatory-buff era it can create is on the rejected list by name.
The dedicated healer's oldest complaint is downtime without agency: between damage spikes, nothing you do matters. Bloom gives routine healing a destination. Each Life heal that lands on a living target banks one Bloom charge, capped at three, visible as pips. A new spender glyph, Vital Bloom, consumes all charges: at one, a modest healing pulse around you; at three, a strong party heal plus a short cast-speed blessing. Every unremarkable top-off now also loads the remarkable one. You are never just maintaining; you are always building toward a moment you choose.
Seen in: FF14's lilies feeding Afflatus Misery, the cleanest anti-boredom design of the modern era.
Mostly data: a small rider on the heal spells plus one spender spell; the charge itself is a stacking self-buff the formula system can already read. The spender consumes all charges regardless of count, so timing is the skill. Charge pips ride the existing buff-icon display first; a proper HUD element arrives with S22. The three-charge payout is budget-pinned to roughly two Healing Bursts, stated as illustrative.
Purify is a chore glyph: cast when a debuff icon appears, feel nothing. Proposed: each negative effect removed also heals the target a small amount, and the strongest effect removed converts into a brief inverse blessing using the typed list from S2. Strip a poison and the target gains a short regeneration; strip a hex and they gain a sliver of attunement; strip a root and they gain a step of speed. PurifyBurst does the same across the party at reduced strength. The cleanse becomes a jackpot in exactly the fights that hand out the most debuffs — and knowing what to purify, and when, becomes a knowledge skill.
Seen in: GW1's Smite Hex (remove and punish), WoW's typed-dispel gameplay.
A small handler extension (count what was removed, emit the payoff) plus a fully data-driven payout table. The degenerate loop is guarded from day one: payoffs only trigger on effects a hostile applied, so nobody farms blessings off self-applied debuffs. The heal-per-removed keeps the immediate cast honest so holding cleanses to fish for bigger payoffs never beats cleansing now.
This game has no ground reticle, no mouseover casting, and party-only friendly targeting. When three people are hurt at once, the UI cannot express "heal the worst two first" — so the spell does it instead. Mend the Line rolls one healing pool and divides it across the party weighted by missing health: the ally at 30% gets the lion's share, the ally at 90% gets a courtesy. It is deliberately less efficient per point than an aimed Healing Ray on the right target — aiming stays the skill, and the split is the honest answer to chaos. It arrives both as a standalone glyph and as the three-stack transform of Healing Burst (S9).
Seen in: DAoC's spreadheal, built for exactly this targeting-constrained case.
One new distribution mode in the heal application path (weight by missing-health share, with a floor per target), then the glyph is data. The whole split computes from the same replicated party health the party window already displays. The efficiency discount is published: illustratively 80% of an aimed heal's total.
Sitting in the game's data is an entire unshipped Life spell: Summon Ankh — three authored spell assets, reserved skill ids, its own power stat already wired, a tree row flagged not-implemented. Ship it as the school's centerpiece: plant an ankh, and for its duration the ground around it is consecrated. Allies standing in the light receive a steady pulse of healing and a small Life Attunement blessing — the school's authored-but-unreachable ground aura, finally reachable, as the totem's gift. The ankh is a physical object: it can be destroyed, it does not move, and choosing where to plant it is the healer's positioning game. Every other school has a summon. Life's was always meant to be a place, not a pet — and Ultima has exactly one symbol for that place.
Seen in: ESO's ground-heal culture, WoW's totem play, UO's ankh as the resurrection landmark.
The rare feature where the data groundwork already exists: the skill ids are reserved, the power stat is wired, and the engine's ground-field volumes are exercised daily by other schools. The build cost is assembly and verification, not invention — a destructible totem object, its pulse, the aura field, and the Life dressing from S6 replacing the Earth effects the assets currently borrow. One ankh per caster, enforced by a buff channel from day one. Every "already wired" claim gets verified in Phase 0 before this card is promised externally.
Death in this game costs something real: decay. Resurrection currently ignores it, so a grandmaster's rez and a novice's rez deliver the identical product, and nobody has a reason to seek out the better healer. Proposed: Resurrection restores a share of the decay from the death it reverses, scaling with skill — illustratively 10% restored at skill 40, 25% at 100, 50% at 200, on top of the existing perk that its reagents go free at grandmaster. The greatest support spell in MMO history was great because it handed people back their time. Suddenly "is there a real healer in the group" has an answer worth gold.
Seen in: EverQuest's 96% rez, the strongest social currency ever attached to a support spell.
The one Tier 2 card that touches an economy system — decay is a sink — so the restore percentages are explicitly the balance program's property, capped at half, and gated on a real cast by a real invested character, never an item. If the revive path proves harder to reach than expected, the fallback version ships first: higher skill shortens rez sickness, decay restoration follows.
BanishUndead is the school's only attack, and it deals literal zero against anything that is not undead — a dead card in most decks. Proposed: a small offensive line grows from it, and its signature is that Life's damage gives back. Radiant Reprisal: a modest holy strike, full damage against undead, reduced against the living, and a share of the damage dealt splashes as healing to nearby party members — illustratively 30% of damage, split across up to four allies. The solo Life mage finally has a loop; the group smiter heals by fighting, the highest-agency healer archetype the genre ever produced, capped hard so it never outraces dedicated healing. The undead resistance wall that taxes this whole line is already a committed balance item; these cards land beside that fix, not instead of it.
Seen in: GW1's Smiting Prayers, WoW's Atonement, City of Heroes' enemy-routed support.
Mostly data — a spell grown from BanishUndead's pattern with a healing splash rider — plus a small formula addition if the splash cannot yet express "a fraction of damage dealt". Conversion caps are the whole ballgame here; the genre spent fifteen years learning that, and the cap is published from day one. The line stays at two glyphs and undead-favored so it never blurs into Sun's territory. Whether it adopts the authored HolyStrike and HolyJudgement names is S24's call.
The school's passives are stat sticks: power, range, mastery, ward. Add one with a behavior. With Grace of the Giver trained, every Life glyph you resolve on another character also heals them a small amount, scaled by your Life Attunement. Purify heals as it cleanses. Resurrection's target stands up with a cushion. Aegis lands with a touch of mending. The whole kit becomes healing-flavored without a single spell changing, utility casts are never wasted on the health economy, and attunement investment pays out on every card in the deck. The cheapest cohesion win in the catalog.
Seen in: GW1's Divine Favor, the class mechanic that unified an entire profession.
Data-heavy, code-free: a new passive row plus a small rider term in each ally-targeted spell expression, following the exact heal-others pattern every heal formula already carries. Kept deliberately small — illustratively a few percent of a Healing Touch per cast — so it flavors rather than feeds, and it reads attunement, not cast count, so spam earns nothing.
Our published balance targets draw bands for damage schools, for the bard's party uplift, and for pets. They say nothing about support: no statement of how much party healing a dedicated healer should sustain, what fraction a hybrid should reach, or how prevention counts against the budget. Without the band, every card above tunes against vibes. Proposed rows, all draft-for-measurement: a dedicated healer sustains party healing near one enemy's expected incoming damage at the content's tier; a hybrid reaches about 60% of dedicated; wards count at 80% of face value against the band, because prevention can be wasted; and the healer's solo clear stays inside the existing everyone-rule of 1.25x the median school. Life has never once appeared in the balance change log. This row is how it starts appearing responsibly.
Seen in: this project's own taming page, which published the pet bands its proposals tuned against.
Today you cannot heal a stranger. Not because of an engine limit: a single per-spell flag allows friendly casts outside the group, and exactly one ability in the game sets it. Flipping it on the heals is an afternoon. Whether we should is a real question, so it goes to the page as one — vote on this card.
The case for: healing strangers is the oldest social texture in the lineage this game claims. Ultima Online's roadside healers built reputations on it; sieges and public events fragment parties constantly, and the healer standing next to you should matter. The whole bet of this page is that the healer is a social role, and the party wall is the hardest wall around that role.
The case against: in open-PvP scenes an unattached healer becomes a griefing tool, keeping a flagged fighter alive from outside the fight with no way to answer back — UO needed an entire criminal-flag system to police exactly this. Add kill-credit leeching, and strangers topping each other off could flatten intended difficulty in public scenes. The flag is trivial; the consequence layer is not.
A possible middle: enable it outside PvP-flagged scenes only, or enable it and revive the dormant healing-enters-combat rule in the same change. Stated, not chosen.
Seen in: UO's cross-healing culture and its criminal-flag correction; the sterile party walls most of the genre chose instead.
Cast a heal at a wounded town guard, an escort charge, or a quest companion mid-fight, and the spell refuses: the targeting code only recognizes players as valid friendly targets. For a virtue-flavored school in an Ultima game, being unable to help the villagers is a strange wall. Proposed: friendly NPCs accept beneficial Life casts. This is also quietly a content lever — encounter designers can finally build "keep the wounded knight alive" moments, the oldest escort drama in the book.
Seen in: every game where protect-the-NPC content exists.
A targeting fix that asks "is this character friendly to me" instead of "is this a player in my party", plus health display for friendly NPCs (the nameplate path already renders NPC health for hostiles). Needs a guard for disguise-flavored content where an enemy reads as friendly, and an audit of scripted encounters that do not expect their NPCs to be healed.
The healer's most important numbers are invisible. Life Attunement scales every heal and appears nowhere on screen. Your own Grace ticking on an ally shows nowhere you can read at a glance. Healing generates a quarter-weight of threat, and the game never says so — the first monster that turns on the healer feels like a bug. One coherent surface fixes all three: an attunement readout on your status bar (with the Bloom pips from S12 when trained), a small heal-over-time and ward mark on each party row so coverage reads at a glance, and a subtle tell when your healing has drawn attention. One surface, not three widgets, and no new windows anywhere.
Seen in: every modern party frame's aura tracking, reduced to this game's no-addon, glanceable philosophy.
Pure client UI over data the interface already receives; party rows already render buff icons, so the marks are a filtered emphasis, not a new list. Every element defaults subtle and carries an options toggle, and long-string clipping gets checked on every new label. The threat tell is the only piece needing a new event from the combat layer.
Buried in every single-target heal is a multiplier that grows the longer you have not dealt damage. It is the school's only conditional mechanic, and it is a tax on participation: the moment a healer contributes damage, their heals shrink. Every peer school's signature mechanic rewards doing something — Death rewards risk, Sun rewards timing. Life alone rewards abstaining. And the community data says nobody lives on the favored side of it: there are no dedicated healers to reward, only hybrids to punish. Three options, argued openly. Keep it: it is the one mechanical statement that pure support is a build, and removing it quietly buffs every self-healing hybrid in the game. Floor it: hybrids keep a guaranteed share, illustratively 80%, so the bonus nudges without punishing. Invert it: replace it with momentum that builds while you heal, rewarding the act instead of the abstinence — which also makes S17's smite loop coherent. The decision gates S17 and S18 tuning, so this card carries a deadline, not a someday.
Seen in: FF14's decade-long healer-damage argument, which is what this question grows into if left unanswered.
Two skill stubs named HolyStrike and HolyJudgement exist in the game's data, assigned to the Life school, never shipped. They are the fossil of a decision nobody wrote down: does Life get real offense? This page proposes a modest answer through S17 — a two-glyph, undead-favored line that feeds the party — but the larger identity question deserves its own card and vote. The case for a spine: every other school has one, solo play is most of this game, and the healer-knight is a beloved archetype the genre has proven can coexist with support primacy. The case against: Life's purity is its identity — the only school that cannot hurt you is a meaningful thing for a world to have — and Sun already owns holy-flavored offense. This page's stance, stated honestly: S17 as designed is compatible with either answer, and the stubs stay dormant unless your votes and the balance program both say otherwise.
Seen in: DAoC's smite cleric — played solo, loved solo, argued about forever — and GW1's contained version of the same idea.
Every approved card composes into surfaces the game already has. No new windows: the party rows learn to carry the triage, the chip line carries the healer's economy, the hand answers different questions, and the cast bar tells the truth. The same regions as the Section 1 screen, in the same order — that is the point.
| Phase | Ships | Rule |
|---|---|---|
| P0 · De-risk | Nothing ships. Verify the dormant plumbing this page leans on (the ward machinery, the Ankh assets, the revive path), draft the S19 band, re-run the corrected skill audit for trustworthy Life baselines. | Every "already wired" claim gets proven before it is promised. |
| P1 · Repair, in the open | S1–S6, plus S18 as the one cheap Tier 2 rider that needs nothing else. | All of Tier 1 before any Tier 2 verb; each fix published with what was broken. |
| P2 · The verbs, data wave | S7, S8, S10, S11, S12, S13, S14. | Data over existing machinery; each ships to the test shard under the S19 band with the two-client verification matrix. |
| P3 · The big builds | S9 (the transform hook, then the transforms), S15 the Ankh, S16 rez tiers, S17 the smite line, S22 the HUD. | S17 waits on the S23 and S24 decisions and the undead resistance work. |
| P4 · The decisions | S20, S21, S23, S24 resolve from page votes and thread input, collected during P1–P3. | Open questions close before P3 tuning locks; none of them resolves silently. |
flowchart LR P0["P0 verify plumbing"] --> S7["S7 overheal ward"] P0 --> S8["S8 Sanctuary line"] P0 --> S15["S15 the Ankh"] S19["S19 healer band"] --> S8 S19 --> S12["S12 Bloom"] S19 --> S14["S14 Mend the Line"] S19 --> S17["S17 smite hybrid"] S1["S1 Burst fix"] --> S9["S9 stack transform"] S2["S2 Purify list"] --> S13["S13 Purify pays out"] S4["S4 readable numbers"] --> S7 S7 --> S8 S6["S6 spell identity"] --> S15 S9 --> S14 S10["S10 Grace in bloom"] --> S11["S11 combo slate"] S23["S23 pacifism decision"] --> S17 S24["S24 holy stubs decision"] --> S17 S12 --> S22["S22 healer's HUD"] S10 --> S22 style S9 stroke:#c9a55a,stroke-width:2px style S15 stroke:#c9a55a,stroke-width:2px style S19 stroke:#a33d2e,stroke-width:2px
This page builds on what the school already does well rather than re-asking for it: heals stack at +25% per copy, better than any damage gets; Healing Grace carries a built-in bleed cleanse on every tick; Purify grants death resistance as it cleanses; Resurrection's reagents go free at grandmaster; and the heal-over-time architecture re-derives its numbers from current formulas on load, which makes every future rebalance retroactive for free. The method of this page — measure the tree against the engine, then the genre, then propose mostly data — was established by the Chaos magic design review, and the balance measurement rails it plugs into were built by the combat balance program.
| Claim | Source surface |
|---|---|
| Six of nine actives are the same restore expression | The nine Life spell definitions and their amount formulas |
| Healing Burst omits the attunement term | The Burst spell asset's formula, against its four siblings |
| The cleanse list's duplicate and stray entry | The hardcoded purify table in the combat effect code |
| The ally-cycle loop cannot advance | The friendly-cycle routine, against its working hostile twin |
| Heal text is fixed-size while damage scales | The two floating-text emission paths, side by side |
| Overheal is computed and discarded | The heal effect's full-health guard |
| Zero Life-owned combos; 29 combos in the game | The combo definitions across all spell assets |
| The dormant Ankh: assets, ids, stat, flag | The three Summon Ankh assets, the reserved id block, the skill sheet row |
| Heals stack at +25% per copy vs damage's +15% | The card-count scaling constants |
| Healing threat at quarter weight | The heal-aggro constant in the combat tuning values |
| The community's silence on Life | The 2025–2026 wishlist and balance thread sweep, ~205 threads |
| No Life entry in the balance ledger, ever | The balance change log |
Each shipping card adds its pin in the project's usual style: S1's formula pin at zero and grandmaster attunement, S2's dedupe and data-load tests, S3's cycle-order test against a fake party, S9's stack-resolution tests (two copies boost, three transform), S12's charge state machine, S14's distribution weights and floor, S16's tier math, and a school-wide audit that every formula still evaluates at mastery zero and grandmaster — the "works untrained" invariant, numerically. The one thing no automated test can cover is multi-client effect sync; every phase carries the caster, target, and observer checklist with two live clients.
Shipping game data (the skill sheet, the nine Life spell assets and five passives, the combo definitions, the effect channel data, the dormant Summon Ankh assets); the combat engine's effect, formula, and targeting code; our published balance targets and balance change log; the 2025–2026 community wishlist and balance thread sweep; and the genre record — Ultima Online's healing and reputation history, EverQuest cleric culture, DAoC's healing lines and spreadheal, WoW's HoT, dispel, and overheal-meter era, Guild Wars 1's Protection Prayers and Divine Favor, FF14's lily economy and healer debates, and City of Heroes' support set design.
No account needed. Every proposal above also has its own Discussion box for talking about that one alone; this thread is for the plan as a whole. Comments are plain text, held to basic decency, and may be trimmed.