When the Internet and the HTTP protocol were first conceived, information transfer was envisioned to be one way, web server to web browser, and the information on the server rarely changed. As the Internet matured, dynamic web page technology was developed, delivering updated content as soon as the user requested it. This was nice, but what if the website wanted to deliver new data after the page loaded, and have it trickle onto the web page itself without reloading?

Ingenious web developers figured they could keep the HTTP connection alive after the initial connection attempt. New information could flow to the user over an HTTP session that launched seconds or even minutes earlier and never closed. However, many firewalls ...

Get HTML5: Your visual blueprint™ for designing rich web pages and applications now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.