We’re back in action this week! Our fully remote team spent last week in Bali coming together as a group and 

starting the execution work for the quarter — and boy do we have A LOT of stuff to ship.

We took a light touch this week as we’re getting back up to speed but we managed to ship a bunch of sorely needed improvements to our documentation. We also want to hear from you about pricing and packaging.

Docs now have full-text search, a new UI, better snippet highlighting, and plenty more

This quarter we have a number of improvements to documentation lined up. First up — doc search. We received a lot of feedback previously that searching through docs was … not fun. 

The first thing you’ll notice is that we rewrote the search UI. Results are now organized by hierarchy root with each section in results mapping to a section in the sidebar. 

We’ve also added full-text search. Previously doc search relied on tag-based search so this is a big improvement. If you’re curious, we’re using 

. We’re using the hosted version for the moment but in the future we might migrate the service over to Railway and write a blogpost for how to get up and running.

Finally, we added some hotkey support for arrows and selection to make it easy to keyboard your way through the search results.

In addition to doc search, we’ve updated docs for accuracy, clarity, and comprehension.

 The LAST thing we want is for you to use the new doc search and end up on a page with bad information. We made a number of passes and fixed typos, updated images, documented a couple undocumented CLI routes, and more. 

More changes will be coming soon (especially to Nixpacks docs!) but in the meantime we’ve also merged a number of community contributions into the docs. 

Have you had a less-than-stellar experience using the docs? I

f you’d like to call out any items for improvement, feel free to 

If you’ve been a reader of our blog, you know that we’ve been dealing with a 

 in user growth. At the same time, we are pushing the product forward as fast as we can to support more complex production use cases and teams.

The complicating factor for us during this stage of growth is that we deal with a frankly outrageous number of bad actors trying to spam free compute from our service. We 

 two weeks ago and have been doing our best to make it more difficult to abuse the service. 

The best tool we have at our disposal for incentivizing good behavior is pricing and packaging. We’re working to update and clarify pricing to provide better incentive alignment with you the legitimate user as you onboard to Railway. 

As an example of things we’re looking at, we think one incentive misalignment is that Railway does not currently get cheaper as you increase use — but it should! Our updated pricing will do a better job incentivizing deep use on the platform with price breaks for committed spend.

 about your experience with Railway pricing and packaging. 

 and let us know what you think. We want to know what you find confusing, what you find doesn’t align with your use case, and what kind of transparency you want to see from us in pricing.

We’re looking forward to hearing from you. 

We laid the groundwork to add better error handling to the CLI with custom errors

We now stream both build and deploy logs in the CLI when running 

We added deployment ID to the response when calling 

We added Windows-specific shell detection and batch script support for 

We made it easier to copy deployment IDs on the deployment page

We added light mode to the verification state

https://res.cloudinary.com/railway/image/upload/v1682717482/og-images/changelog-docs-search_n8y1n0.png

Docs Search, Docs Copy Updates, Pricing and Packaging RFC

Changelog #0131