Cloud9 IDE

» see c9.io as it lived, on the wayback machine

~ the obituary ~

cloud9 was a category i kept seeing on hacker news between roughly 2011 and 2016. my ide habit was already settled by then. i had moved from notepad++ on a turkish windows xp install during university, through gedit on a kubuntu dual-boot, then sublime text after about 2012 on a thinkpad. honestly, the workflow was always a real editor on real local disk. the cloud-ide pitch reached me as a tab i would open out of curiosity and then close back to sublime.

the company that became cloud9 started inside ajax.org, the open-source project ruben daniels and rik arends ran in amsterdam, the same one that gave the world the ace code editor (a fork of mozilla skywriter, originally bespin). ace was the editor widget that ended up underneath a lot of in-browser code experiences for the rest of the decade. cloud9 packaged it inside a complete ide shell with file tree, terminal, debugger and git integration. you opened a tab in chrome and you got a vps with a real linux on the other side. the funding round was five million dollars from accel and atlassian in 2013.

amazon bought the company on july 14 2016. the brand and team folded into aws and the product was relaunched as aws cloud9 about a year later, this time tied to ec2 instances inside the customer's own aws account. honestly, for some reason. the consumer-side c9.io address went read-only in september 2019 and then dark. the aws-side service kept running, but quietly. on july 25 2024 amazon announced no new customers for aws cloud9. existing customers got a sunset path that runs until 2028.

~ the case file ~

~ how it landed ~

in those days the pitch for cloud9 was real. a junior dev on a netbook could open a chrome tab and have a working node.js or ruby or python environment in about thirty seconds, without installing anything locally. the beaglebone black single-board computer shipped with cloud9 as the recommended ide for beginners. honestly, that was a category of user the local-editor tradition did not serve well. the people who needed cloud9 most were the ones who could not install xcode or visual studio on the laptop they had.

~ the slide ~

this part you had to be inside aws to feel, and i was not. once microsoft open-sourced visual studio code in 2015 and shipped extensions that ran remote workspaces over ssh, the cloud-ide value proposition got squeezed from above. by 2019 github codespaces was in preview, running vscode in a browser tab against a github repo, owned by microsoft after the github acquisition. honestly, aws was still positioning cloud9 as the in-browser companion for amazon's own services, but the wider developer market had moved.

amazon itself shipped two replacement directions. the aws toolkit for vscode bundled the cloud9 features into the editor people already used. q developer became the ai-coding strategy that displaced the entire cloud9 motion. either way, by 2024 there was no clear product wedge left for a browser-only aws ide.

~ what replaced it ~

the ace editor library still ships under the ajax.org name and powers code widgets across a long list of consoles, including parts of aws itself (cant remember the full list). that part of the cloud9 legacy outlived the product brand by a margin. the cloud-ide pitch itself did not die, it migrated. github codespaces, gitpod, replit, stackblitz, devpod are all running the same pitch a decade later, mostly on top of vscode rather than ace.

in those days. the original cloud9 was just early. they shipped the right product in 2011, against an audience that mostly still wanted a real editor on a real laptop. by the time the audience came around, the tools had been rebuilt on top of vscode by people who came after.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »