Why Local First Matters to Businesses

Overview

Adam Fish has been building sync engines for 13 years — from a failed CRM startup, through Realm (acquired by MongoDB), to founding Ditto in 2018. Ditto's differentiator: true peer-to-peer sync over Bluetooth and WiFi, no internet required. But this talk isn't about the tech. It's about a harder problem: making businesses care.

mindmap
    root((Local-First
        for Business))
        The Timing
            Hardware evolution
            Edge devices are supercomputers
            Connectivity still unreliable
            Edge computing is the open platform
        What Businesses Care About
            Things don't break
            Fast with fallback
            Revenue and customer experience
            NOT the term local-first
        Common Traps
            Overemphasis on collaborative text editing
            Benchmarks over real-world usage
            Diminishing returns on DX
            Ignoring economics and durability
        Real-World Impact
            Restaurants and POS
            Airlines and crew coordination
            Defense and national security
            Humanitarian and healthcare
        Peer-to-Peer Networking
            Bluetooth and WiFi mesh
            No server single point of failure
            EU forcing WiFi Aware standard
            Untapped opportunity
        The Bigger Purpose
            Economic value enables durability
            Durability enables social impact
            Firefighter safety
            Remote healthcare

Key Arguments

The Status Quo Is the Real Competition

After 13 years, Fish's most controversial takeaway: traditional request-response architectures are fine for most businesses. That's the competition — not another sync engine, but the default of "keep doing what we've always done." Great technology alone won't overcome that inertia.

Edge Computing Is the Open Platform

The timing argument is compelling. The iPhone 10 in 2018 was a supercomputer compared to the original iPhone — serious CPU, GPU, battery. Silicon carbon batteries launching now add 30-50% more energy density. All this compute power sits largely untapped. Cloud computing defined the last decades; edge computing defines the next ones. Local-first lives at the edge.

Businesses Don't Care About "Local-First"

They care about their business: making money, delivering great customer experiences, avoiding downtime. Fish's challenge to the community: stop speaking in technology terms and start speaking in business outcomes. Decentralized architectures mean systems don't break. Local data means faster fallbacks. High-quality software means more revenue.

The Traps That Kill Local-First Products

Four traps Fish has watched people fall into:

Collaborative text editing — the ultimate CRDT showcase, but most businesses aren't collaboratively editing documents
Benchmark obsession — how people use software in the real world diverges wildly from benchmarks (Realm learned this the hard way)
Developer experience past the point of diminishing returns — matters, but doesn't translate linearly into business value
Ignoring economics — if the software isn't sustainable, it won't last. Ditto's customers are deploying into systems expected to run for decades

The Networking Gap Nobody Talks About

While mobile hardware leapt forward over 10 years, wireless connectivity barely improved. 5G helps, but wireless is fundamentally unreliable. The "last mile" and even the "last few feet" of connectivity remain painful for businesses. Ditto's mesh networking — devices forming ad-hoc networks over Bluetooth LE and peer-to-peer WiFi — turns "offline" into just another network topology.

The EU is about to make this easier by forcing Apple to adopt WiFi Aware, the open standard for peer-to-peer WiFi. Android and iOS devices will finally talk directly over WiFi without Apple's proprietary AirDrop stack.

The talk's emotional core: economic sustainability enables durability, and durability enables social impact. Fish shares how Ditto powers wildfire firefighting — a Cal Fire presentation revealed a firefighter died because a tanker crew didn't know someone was on the ground. A simple lat/long coordinate could have saved his life. Remote healthcare, humanitarian nutrition screenings, defense operations — these use cases only work if the company behind the technology survives long enough to serve them.

Notable Quotes

"After those 13 years, one thing I have learned is that unfortunately traditional architectures — they're fine."

"Businesses don't really care about local-first. What they care about is they're running their business."

"Ignoring the economics of what we do ultimately means we might not be building durable stuff."

Practical Takeaways

Frame local-first benefits in business language: reliability, speed, cost reduction — not CRDTs, peer-to-peer, eventual consistency
The networking layer (mesh, peer-to-peer) is an underexplored differentiator beyond pure data sync
Build sustainable businesses around the technology or the impact disappears when the company does
Edge computing is the next platform shift — the hardware is ready, the software isn't

Connections

local-first-software-pragmatism-vs-idealism — Same conference, complementary angle. Wiggins frames the idealist/pragmatist partnership; Fish argues the pragmatist side harder — without business adoption, the ideals stay academic
the-past-present-and-future-of-local-first — Kleppmann envisions commoditized sync infrastructure; Fish shows what happens when you try to sell that to restaurants and airlines instead of developers
unexpected-benefits-of-going-local-first — Linear's story validates Fish's thesis from the other direction. Artman discovered business value (developer productivity, infrastructure savings) that Fish argues should be the pitch, not the afterthought
why-local-first-apps-havent-become-popular — Bambini diagnoses the technical adoption gap; Fish diagnoses the business adoption gap. Both point to the same conclusion: the technology works, the packaging doesn't