🏝️
islands of coherence
"when a complex system is far from equilibrium, small islands of
coherence in a sea of chaos have the capacity to shift the entire system."
— attributed to ilya prigogine
somewhere, a community already solved it. for every problem that is
local but exists everywhere — homelessness, addiction, trash — some town figured
out an answer that actually worked, and almost nobody copied it. these are
those answers, documented as protocols: instruction sets concrete enough
that a group of neighbors could run them.
01 · rehabilitation & reentry · san francisco · since 1971
🏗️ delancey street foundation
delancey street takes the population every other program screens out — addiction,
multiple violent felonies, generations of family incarceration — and it has
no professional staff. every member, all the way up to the leadership, is a
former offender. residents commit to two-plus years, live together, and run the
businesses that fund the place: a moving company, a restaurant, retail. the
operating principle is each one teach one — you arrive, someone slightly
ahead of you teaches you; within months you're teaching someone newer.
being needed is the treatment.
~95% lasting rehabilitation reported · 50+ years running · 0 treatment staff
the protocol
- recruit from the hardest population — don't screen for the easiest
- no staff/client divide: the people being rehabilitated run the organization, leadership included
- multi-year residential commitment — long enough to rebuild identity, not just behavior
- each one teach one: every resident teaches the person behind them
- fund through resident-run businesses, so the model earns its independence
- graduate into leadership: the c-suite came through the program
the 95% figure is as reported — formal studies are limited; verify
before citing. via the case study library ↗
02 · homelessness · brasília, brazilstory unverified
🧹 paying the homeless to clean the city
as the story goes: around 2007, a civil-engineer mayor stopped treating trash and
homelessness as separate line items and collapsed them into one program —
the city paid homeless residents to collect trash and paint over graffiti.
the results stacked: clean streets, income for the people who needed it most,
rising property values — and it cost less than the separate programs it
replaced. almost no other city has copied it.
two problems, one program · city budget went down · visible homelessness ~0 (reported)
the protocol
- find two chronic problems where one population's need (income) is the other problem's missing workforce
- pay directly and simply: cash for visible public work — skip the program-industrial complex
- price it against the status quo: cheaper than the separate spend it replaces
- measure everything — cleanliness, homelessness, property values, total budget — so the win is undeniable
- publish the numbers so other cities can copy it
told from memory and still being sourced — it closely resembles
curitiba under jaime lerner, so city, date, and details may shift.
via the case study library ↗