I'm done making desktop applications (2009)

Original Article ↗ Hacker News Discussion ↗

A 2009 essay arguing that web apps monetize and convert better than desktop software prompts a re‑evaluation in light of today’s landscape of Electron apps, mobile app stores, SaaS saturation, and AI tooling. Commenters weigh trade‑offs between native desktop, web, and mobile: ease of distribution and analytics versus performance, UX quality, user control, privacy, and long‑term ownership. Many conclude that the “right” platform depends on goals—commercial vs. open source, hobby vs. business—and that despite the web’s dominance, desktop apps remain vital for serious creative work, self‑hosted tools, and users wary of the browser-centric, subscription-heavy model.

Age and Context of the Post

  • Many point out it’s from 2009; tooling, distribution channels, and user norms are very different now.
  • Some argue its conclusions were reasonable then (thick clients, Swing, Flash/Java era), less so in 2026.
  • Others jokingly propose the inverse title for today: they’re “done with web apps” and rediscovering desktop / on-prem.

Desktop vs Web vs Mobile: Monetization and “Metrics”

  • Original thesis: web apps monetize and convert better than desktop apps.
  • Several commenters dispute how big the desktop “funnel” penalty really is; with good installers and flows it can be close.
  • Others say mobile now has the strongest B2C monetization, largely because app stores have payment info on file.
  • Counterpoint: mobile spend per user is often small; ad-supported models require huge scale.

Open Source, Hobby Projects, and Motivation

  • Many note the article assumes “goal = make money.”
  • For open source or hobby work, concerns like conversion, piracy, Adwords, analytics, and A/B testing often don’t matter.
  • Some push back: even non-commercial projects that want impact still need to care about onboarding, usability, and metrics.

Browsers, Electron, and “Universal App Engine”

  • Strong criticism of Electron as “worst of both worlds”; others note employers/governments still pay for Electron-based tools.
  • Debate over whether the browser effectively is the universal app engine versus a bloated, inefficient, historically-accidental one.
  • Some emphasize webapps’ huge distribution advantage; others stress native apps’ efficiency, offline use, and deeper system access.
  • Several lament GUI toolkit pain (GTK, early SwiftUI) and concede Electron/web often “just work” despite their flaws.

UX, Stability, and User Control

  • Desktop proponents value stable interfaces, predictable upgrades, offline capability, and local data/control.
  • Web critics complain about constantly changing UIs, onboarding popups, and dependence on remote servers.
  • Some highlight that self-hosted webapps on a LAN blur the “desktop vs web” distinction.

Infrastructure and Deployment

  • Web/SaaS is praised for centralized updates and simplified support.
  • Others note the cost: servers, DNS, DDoS risk, cloud complexity, surveillance potential.
  • A recurring theme: for many internal tools, simple desktop apps or on-prem webapps would have sufficed.

Side Tangents: Startups, Piracy, and Ethics

  • Discussion touches on indie SaaS eras, low-hanging-fruit niches being mostly gone, and the danger of building complex products without validating demand.
  • Debate over software piracy, search engines surfacing cracks, and whether “don’t be evil” ever genuinely guided large tech firms.