Hacker News Reader

Comment by pts_ | Hacker News Reader

It says so right in the readme. It's a formatter, linter, bundler, compiler(transpiler?) all in one.

I’m looking for something more specific than that. To go to such great lengths, the value propositions have to be more than just bundling a few things together. (And I did read the project philosophy section.)

If you’ve used Metro or Webpack before there is very little explaining to do. The existing tooling is bloated, fragmented (think 40k deps) and slow, and the experience of sewing it all together and keeping it in one piece extremely frustrating.

I have, but there’s still a lot of explaining to do. E.g., what’s the extensibility story? To do anything remotely useful with webpack you need to install a million loaders and plugins, but the plugin architecture also lets you do a lot that can’t be done with opinionated bundlers. So, what’s Rome’s approach to this? Enumerating existing tooling’s problems does not automatically guarantee something better.

The author has commented that this isn't yet ready for public consumption, but is instead just a call for other possible contributors to the project. So:

1. Everything you seem to be asking for seems like it is not relevant at this point in the project's lifecycle. 2. IMO people who are capable of providing good contributions don't actually need what you are asking for. They can review the README as is, review the source code, etc.

In short, patience.

I don't think "less bloated/fragmented/slow Webpack and co" is what sebmck intends its USP to be. It sounds more like he expects there to be some advantages to arise specifically from a single tool doing everything related to processing source code, e.g. the linter being able to catch bugs by also being aware of how code will end up after bundling.

It would be interesting to see examples of specific such advantages that could be gained, because I'm having a hard time thinking of any.

I think Deno has the same concept.

Reply on news.ycombinator.com