Recently, four pillars of the JavaScript community (James Snell, Natalia Venditto, Michael Dawson & Matteo Collina) teamed up to create a resource that lays out nine principles for doing Node.js right in enterprise environments. On this episode, Natalia & Matteo join Jerod to discuss all nine. :link: https://jsparty.fm/347
Ch | Start | Title | Runs |
---|---|---|---|
01 | 00:00 | It's party time, y'all | 00:39 |
02 | 00:39 | Sponsor: Notion | 01:28 |
03 | 02:07 | Hello party people | 01:36 |
04 | 03:43 | How it came together | 03:09 |
05 | 06:52 | The nine pillars | 01:31 |
06 | 08:24 | 5: Avoid dependency creep | 06:15 |
07 | 14:38 | 6: De-risk your dependencies | 04:06 |
08 | 18:44 | Sponsor: WorkOS | 02:50 |
09 | 21:34 | 7: Avoid globals | 08:34 |
10 | 30:09 | 8: Handle errors, log well | 04:37 |
11 | 34:46 | 9: Use API specs, generate clients | 05:37 |
12 | 40:23 | Sponsor: Jam.dev | 01:31 |
13 | 41:55 | 1: Do not block the event loop | 03:53 |
14 | 45:47 | 2: Monitor Node-specific metrics | 04:09 |
15 | 49:56 | 3: Use Node LTS | 04:36 |
16 | 54:32 | 4: Automate testing, code review, conformance | 02:32 |
17 | 57:04 | Bonus pillars | 03:09 |
18 | 1:00:14 | Closing time | 02:54 |
19 | 1:03:08 | Next up on the pod (merch!) | 01:00 |
One thing I'd add from experience is: don't forget that JavaScript is single threaded
No point scaling up by adding CPU cores if you don't also take extra measures to use those cores, because otherwise they'll sit there doing nothing
Another example of "worse is better" in this industry.
Last updated: Apr 04 2025 at 01:15 UTC