|
|
Optimize with a SATA RAID Storage Solution
Range of capacities as low as $1250 per TB. Ideal if you currently rely on servers/disks/JBODs
Page 2 of 5
These IDEs are even turning up in places where you might not think of them as IDEs. WordPress, for instance, has an editor module that lets you play around with the PHP and HTML of your website. This is a dangerous tool that gave me all of the power to crash a blog, which I did once or thrice. But the power is seductive and quite useful. You might find yourself rearranging the HTML tags daily because the editor is only a click away.
| JSFiddle | Icenium | Cloud9 | Codenvy | |
|---|---|---|---|---|
| Target user | Developers of Web pages who need small blocks of JavaScript to add features and fix problems | Developers of cross-platform mobile apps | Developers of server applications | Developers of server applications |
| Product | Client-side JavaScript | HTML5 for iOS and Android | Server-side apps with Node.js, Python, Ruby, or Java | Server-side apps with Java, Spring, Python, Ruby, or PHP |
| Preferred JavaScript flavor | jQuery, Dojo, YUI, Moo, Processing, ExtJS | jQuery, Kendo | Server-side Node.js | Blank slate |
| Command line access | No | No | Yes | No |
| Platform integration | JSFiddle site hosts shared "fiddles" | Apple App Store, Google Play, other Android markets | Cloud Foundry, Heroku, Windows Azure | Amazon Elastic Beanstalk, AppFog, CloudBees, Cloud Foundry, Google App Engine, Heroku, OpenShift |
| Pricing | Free | Free trial; $16 to $19 per month | Free basic version; $12 per month | Free basic version; $9 to $99 per month |
Cloud power, cloud latency
The not-so-good news about the browser-based IDEs is that the fun often ended with the press of the Build button. The systems
still largely do the job, but the lag can turn into a real drag. Your experience will vary depending upon the speed of your
Internet and the code you write. Mine fluctuated immensely with the day and the rain. (One of my Internet links is wireless.)
The performance was often acceptable, but it was never as fast as running a local IDE on a machine with enough power and RAM. While much of the lag was probably caused by the network latency, I wouldn't be surprised if some was due to the CPU sharing common in clouds. Nothing beats having your own eight-core machine sitting at your desk.
Well, that's not exactly true. For most tasks, building the code wasn't very complicated and the network latency was very, very noticeable. Big projects, though, might gain significantly from the cloud. CloudBees, for instance, has a raft of machines just waiting to jump on the builds. If your build is made up of N independent tasks that can run simultaneously and your value of N is large enough, there's a good chance the N tasks will run simultaneously on N machines in the cloud. Even the fat eight-core machine on your desk can't do that.
For now, I think the attractiveness of cloud-based IDEs will depend heavily on the type of code and the size of the build. People building Web apps that run in the browser will be the most likely adopters, in part because the cloud is often storing the files. The latency doesn't get in the way very much.
The next may be geographically diverse build teams that are working together on the same files. Atlassian, the dev tool company known for issue management software like Jira and Git hosting services like Stash, is at work bolting a pairs-programming editor onto Firebase. This is bound to lead to more collaborative editing done over distances greater than two feet. The IDEs here also unlock geographically diverse pairs programming. Heck, you might even call it triple or quadruple programming.