TL;DR
Get monitors, keyboards and dev gear delivered free — and shop member deals
- Fast, free delivery on millions of items
- Access to Prime Big Deal Days deals on October 6–7
- Prime Video, Amazon Music and more included
A developer says he moved a static site generator from Deno to Node.js 26.10.0 after using Node on a SvelteKit project. He reports that the migration required few code changes and that builds ran 15% faster in his test; those results describe his projects, not a general benchmark.
A developer has moved a static site generator from Deno to Node.js 26.10.0, reporting that the migration took little work and that builds were 15% faster in his test. In an October 3 account on dbushell.com, he said recent Node.js work on a SvelteKit client project prompted him to revisit the runtime; the speed result is his own observation and does not establish a general performance advantage.
The author said he had used Node.js heavily during the month on a client project built with SvelteKit. He found that Node supported the ECMAScript features he wanted and that some APIs he remembered as inconvenient had been updated. He chose Fast Node Manager, or FNM, to switch between Node versions, saying he had heard it was better for that purpose than NVM. He also said he selected PNPM as his package manager and configured aliases for npm and npx commands used by some scripts and installation guides.
For the site generator, he replaced Deno’s file system API with node:fs and replaced Deno.serve with Hono’s Node adapter, which wraps Node’s HTTP server. He also swapped Deno’s standard path module for node:path. The author described these as the required changes for a minimum viable migration, while noting that the code still favored Deno-style patterns and that he might explore other Node APIs later.
The move did not eliminate all packaging work. The author said Node’s handling of TypeScript files does not extend to files inside node_modules, citing the runtime error ERR_UNSUPPORTED_NODE_MODULES_TYPE_STRIPPING. He used Tsdown to bundle his package and said he needed extra configuration. He also described configuring PNPM’s trust policy for his own packages after self-hosting Forgejo, saying that limitations in his setup meant the packages had lost provenance.
A Smaller Runtime Migration
The account offers a practical example of how one developer’s runtime choice changed as Node.js and its surrounding tools evolved. For teams maintaining Deno applications, his experience suggests that moving a small service or utility may require fewer code changes than expected when its runtime-specific APIs are limited. His migration relied on a small set of replacements, including node:fs, node:path and Hono’s Node adapter.
The reported 15% faster builds may interest developers weighing migration costs against build times, but the article does not provide benchmark procedures, repeated runs, hardware details or a broader project comparison. The figure should be read as a result from this site generator under the author’s conditions. It cannot show whether Node.js will build other projects faster, or whether further optimization on either runtime would change the result.
The post also points to trade-offs beyond speed. The author described using PNPM settings to delay package updates and restrict trust, while noting extra steps to bundle TypeScript packages. Those choices reflect his security and packaging preferences; they do not amount to a comparative security assessment of Node.js, Deno, NPM or PNPM.
Fast Node Manager (FNM) for Node.js
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Why He Returned to Node
The author said he had relied on Deno for a long period, partly through habit and familiarity, and had not been enthusiastic about server-side JavaScript recently. His work on a SvelteKit client project brought him back to Node.js, where he found the modern runtime more capable than he remembered. The post is a first-person account of a developer’s tool change, not a formal product review or a controlled comparison.
His decision also followed several frustrations with Deno. He cited broken ZSH integration, frequent “429 (Too Many Requests)” responses from JSR, and bugs he said affected concurrent HTTP requests. These are the author’s reported experiences; his post does not include independent verification, issue references or responses from Deno Land Inc. He also criticized the company’s direction and described layoffs and its remaining work in negative terms, which are his assessments rather than independently established findings in this account.
On TypeScript, the author cited Node.js v26.10.0 documentation stating that type stripping is unsupported for files under a node_modules path. He interpreted this restriction as a deliberate policy choice to discourage publishing packages written in TypeScript. The cited restriction is the documented behavior; the author’s explanation of the motivation is his interpretation. He also expressed interest in native type annotations in ECMAScript, while acknowledging that their arrival is uncertain.
““Node got a glow-up, wow!””
— The author of the dbushell.com report
As an affiliate, we earn on qualifying purchases.
Limits of the Build Result
The report does not explain how the 15% build-time difference was measured, which hardware or workload was used, or whether the author repeated the comparison. It also does not include timings for other projects, independent tests, or a response from Deno. The result therefore remains a single developer’s account of one site generator.
It is also unclear whether the author’s reported Deno problems affected other users or were limited to his environment and projects. The post names issues with shell integration, JSR rate limits and concurrent HTTP requests, but offers no supporting reports or technical details. The article does not establish how current Node.js and Deno compare across workloads, nor does it resolve the author’s questions about package provenance and ecosystem security.
As an affiliate, we earn on qualifying purchases.
Further Node.js Tuning
The author said he may explore additional built-in Node.js APIs to improve the migrated generator, whose code still favors idiomatic Deno patterns. That work could change performance, but he did not provide a schedule or promise a follow-up measurement. For now, the published account records a completed migration and the author’s decision to remove Deno from his machine.
Readers considering a similar move would need to assess their own runtime-specific code, package workflow and performance needs. The report identifies a small set of changes in one project, while leaving broader compatibility and speed questions open. No wider migration plan, Node.js release milestone or Deno response was included in the source material.
As an affiliate, we earn on qualifying purchases.
Key Questions
What did the developer move from Deno to Node.js?
He migrated a static site generator to Node.js 26.10.0 after using Node.js on a SvelteKit client project.
How much faster were the builds?
The author reported that builds were 15% faster after migration. His post does not state the measurement method or show that the result applies to other projects.
What code changes did the migration require?
He replaced Deno’s file system API with node:fs, changed Deno.serve to Hono’s Node adapter, and swapped Deno’s path module for node:path.
Why did he leave Deno?
He said Node.js had improved and cited problems with Deno’s ZSH integration, JSR rate limits and concurrent HTTP requests. Those points are his reported experience; the post does not independently verify them.
Does Node.js support TypeScript packages in node_modules?
The author cited Node.js v26.10.0 documentation saying type stripping is unsupported for files under a node_modules path. He used Tsdown to bundle his package instead.
Source: hn
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
