DailyJS

History of JavaScript: Part 3

Alex R. Young

Subscribe

@dailyjs

Facebook

Google+

language history hoj

History of JavaScript: Part 3

Posted by Alex R. Young on .
Featured

language history hoj

History of JavaScript: Part 3

Posted by Alex R. Young on .

This article is part 3 of our History of JavaScript series. Last week we
looked at the development of the Document Object Model, and the browser
wars, which culminated in Microsoft losing the antitrust case yet
reaching total domination by the time Internet Explorer 6 was released.

Rhino

SpiderMonkey was written by Brendan Eich at Netscape, and was
implemented in C. This was the first JavaScript interpreter, and is
still used by Firefox, Adobe products, and many other applications.

Rhino began life as a part of a planned Java version of Netscape
Navigator, and was initially created by Norris Boyd. A Java Netscape
never emerged, but Rhino did. Sun licensed it, and it was eventually
released to Mozilla in 1998. It was released as an open source project
soon after.

Rhino originally compiled JavaScript into bytecode, but compilation took
too long, so an interpretive mode was added.

ECMA-262 Edition 3

At the end of 1999, ECMA-262 edition 3 was published. This edition added
regular expressions, exception handling with try and
catch, and other improvements. Most browsers today still
ship with a superset of edition 3.

After this point, ECMAScript entered a strange time in its development.
A compact version was published to suit embedded devices, which appeared
in the HD DVD specification. E4X was also published, which is ECMAScript
for XML. This version of the language adds native XML support, which was
thought to be a useful alternative to DOM interfaces.

ECMAScript 4

Work on ECMAScript 4th edition started after the release of the 3rd
edition in 1999. An interim report was released in 2003, at which point
work on the standard stopped. ActionScript and JScript.NET implemented
parts of this early standard.

In 2005, work started on ES4 again. The group wanted to draw on
Microsoft and Adobe's experiences, and expected to complete the 4th
edition by the end of 2008.

The feature laundry list was just over two pages long by October 2007:
class-based OO, multi-methods and operator overloading, optional type
annotations, strict mode, generators, and many more features were
expected to be supported.

At this point, Douglas Crockford spoke out about the complexity of the
proposal at the Ajax Experience conference. Brendan Eich wasn't
impressed -- he'd been working on edition 4 for 2 years already. He
admitted that Microsoft's support had been lacklustre. A longer response
was published on Brendan's Roadmap
Updates
.

RIP ECMAScript 4

This represented a turning point in the political discourse of
ECMAScript 4. By the end of 2007 it seemed obvious that this dispute was
serious and work on ES4 was untenable. By August 2008, Brendan Eich
announced the Harmony project, which aimed to reunite parties and
continue development on a new ECMAScript standard. ECMAScript 4 and
JavaScript 2 had died.

Even though ECMAScript 4 seems naive in retrospect, many developers had
already built open source libraries that could work with it. Even savvy
developers like John Resig had blogged and talked about the derailed
standard.

This difficult period demonstrated that JavaScript's strengths lie in
its simplicity, which is an important lesson for the community at large.

Next Week

Next week I'll review the frameworks that made the Web 2.0 revolution
take off.

References