-
Notifications
You must be signed in to change notification settings - Fork 6.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Meta: Replace the current Infrastructure with Next.js #4958
Comments
Folks at Vercel let us know about a few features that would be super useful:
|
Did a quick perf & mem usage test (using Firefox devtools) on nodejs.org/en/download & nodejs.dev/en/download (because those are popular pages & support more or less the same functionality)
Disclaimer: as the maintainer of metalsmith, I'm sad to see its most flagship website move away from it (especially since it's been improving fast since 2022 and I've been in contact with current maintainers regarding certain plugins). The nodejs.dev site does add some overhead in memory usage (x5) and increases the amount of data requested (x25), and note for a documentation page this also holds (11MB nodejs.org/docs -> 60MB nodejs.dev/docs). I personally don't like the opinionated blend-all-concerns, one-tool-fits-all approaches of modern webtools like Next.js (or Gatsby, or Nuxt.js). Metalsmith might require you to put in some more effort for certain features, but you will never be "stuck" because of it. Any Node.js developer with basic JS & markdown knowledge can understand the build in the glimpse of an eye without having to write everything in JS and learn JSX, or browse through tons of fast-evolving docs. Then again, I might be the odd one out because I also couldn't care less about a "light/dark" theme switch. That said, IMO the new features and updated design of nodejs.dev may justify the (slight, depending on internet connection & device RAM) perf/ network latency penalty. |
@webketje just a quick reminder that:
|
I'd also like to comment that this approach of benchmarking isn't really appropriate as there are way too many things at play here (such as extra network requests, service workers, et cetera), nevertheless Gatsby adds a lot of overhead on top that definitely are non-optimal, and possibly some of the causes of the extra overhead. |
Finally, I'd like to appreciate your work with metalsmith and for supporting us during all this time. Your work is not going to waste 🫶 I do understand and respect your own opinions and preferences and I also believe that for the feature set we need we must depart from the current infra. Regardless, I don't want to sound like I'm just being picky on your benchmark, data is data, and indeed things are relatively hungrier now 👀 gonna def prioritise slimming down those fat docs ♻️ |
We're on the same page, all points taken. The quick perf/mem test was just to satisfy superficial curiosity, and the second part is just an opinion/ preference. I got practically attached to this repo as I use it to sanity-check new releases work locally. I give you a 👍 here because not sure how to do it on a comment in Github's mobile PWA |
PS: Minor, but I don't think that it's on a free tier for the .dev, but some credits from @MylesBorins from the regular Google cloud offering |
Copied from #4991 (comment)
|
That makes sense. It would be good even for posterity reasons :) (cc @Trott wdyt?)
It would make sense if both websites planned to be different, but the idea is that nodejs.org will have the exact same layout, pages (content), and feature set from nodejs.dev, and eventually, nodejs.dev will sunset. (The only difference is that all the blog posts from nodejs.org will be kept).
We don't have plans to bring the newly designed API docs from nodejs.dev to this repository yet, the reason being there are too many open topics, such as:
|
BTW @nschonni thanks for putting these together :) |
I not fully fleshed out idea for the |
Closing as we adopted Next.js and this is done. |
This is a Meta issue regarding the current Website Redesign Initiative that has taken place on the https://github.com/nodejs/nodejs.dev repository since 2019.
We believe that in terms of features, design and content the https://nodejs.dev website reached its goals and pretty much has the format we all desired. (Or to a certain extent).
The overall idea is to incrementally move the feature-set from the new website to the current website. The overall idea is explained here and it would, in summary, mean that the underlying infrastructure of the website would be dropped to support Next.js. But what does that mean exactly?
After that, the next steps would be (not necessarily in this order):
This would be a long-term effort work, that I'm pretending to commit myself with and would love the support of the existing @nodejs/website and @nodejs/nodejs-dev teams to make this a reality.
There are plenty of benefits with this approach, and I'm definitely open for discussion and more. There's an open issue here for scheduling a bi-weekly meeting for the Website WG, and anybody interested (being part of the WG-or-not) is welcome here!
The text was updated successfully, but these errors were encountered: