What does Google Closure Library offer over jQuery?

Considering

• business background
• community support
• available extensions
• default set of features
• simplicity of use
• and reliability

why do you prefer one over the another?

I'll try to add my piece of information.

More than another JS lib

As I understand it, Google Closure is not only another JS library, but it is also a set of tools that will allow you to optimize your JS code. Working with jQuery gives you good tools and a lightweight library, but it does not minify your own code. The Closure compiler will. The closure inspector may also be useful, as sometimes minified code has a different behavior than the original one, and is a pain to debug. It integrates with Firebug and support unit tests, which are both developers' best friends nowadays.

Documentation

I guess that as any new library VS a well established one, it will lack the availability of tons of extensions and tutorial that jQuery has. However, being pushed by Google should ensure that support and reliability will be both pretty good. The current documentation and tutorial both seem really good, too.

Features

The features of Closure look decent, though, and its modular architecture is promising, too. I guess Google has been using it internally for a long time, which means that you could expect all basic features (and more) to be implemented, and probably in a very optimized and scalable way. They are trying to present it as the STL of JavaScript, so they should have polished it.

After looking at the features more closely, it seems that this may be a step forward for web-applications development compared to existing libraries as jQuery. It guess it benefits internal developments at Google, but things like detecting the online state (see goog.events.OnlineHandler), easy integration of AJAX requests and JS actions in the browser history (see goog.History), or the legions of great widgets they provide (see goog.ui package) may help all of us building even more awesome webapps ;) !

It comes with templates features that integrates with Java (who said GWT ?), so this may also be another plus for Closure.

Ease of use

Finally, it looks pretty simple to use. The syntax may be a bit more verbose than the short $ jQuery function, but with IDEs and auto-completion, it's not a real problem. Moreover, I'd say we can expect a good integration in IDEs like Eclipse, coming from Google.

EDIT: as requested, let me say a few words about the GWT reference. Google Web Toolkit is a Java library that allows to create AJAX-enabled web interfaces and that generates (and optimizes) the required JavaScript code. As Google Closure allows to create Templates that can be used both client- and server-side (using JavaScript and Java), my guess is that it will soon be possible to use them jointly (if it's not already the case).

In my brief look at the API I find the differences between jQuery and Closure to be striking.

jQuery is basically just a simplified way to do many frequent operations in a cross-browser way.

Closure is a framework that is very new, in that they provide a cross-browser way to use the <canvas> tag, for example, and they have added new events.

So, this is adding onto what we typically do with javascript, they are taking many operations that people want to do and putting them into the API.

For example, they have an event to tell if the online state has changed. So you can tell if the system is online.

They have javascript functions that use tools such as Google Gears, which continues with the fact that they have extended what can be done with Javascript.

It will take me a couple of days to digest all the changes, but I can see that this could have a big impact on web applications that can be developed.

The biggest advantage of Closure Library is that is is designed for Closure Compiler. This opens completely new possibilities for JavaScript development...

The compiler has several cool features:

• It compiles readable JavaScript into compressed machine-readable JavaScript - it has the best compression ratio in "ADVANCED" mode.
• Documentation of the code with JSDoc Tags is important: the compiler reads it and you get warnings during compilation for typos in documentation, wrong use of a @constructor, wrong type of a variable, misuse of a field annotated with @private and @protected, etc.
• If you write a reusable JavaScript library, such as OpenLayers or Google Maps, you formally export your public API - and the compiler optimizes your internal code.
• The end applications can be compiled together with the library - and then the unused parts of the library are removed from the produced code. Dependencies are solved automatically by the compiler.
• Compiler accepts constants to remove unwanted functionality - this allows compilation only for particular browser such as Mobile WebKit, for only one of Quirks mode or Strict mode, compilation without support of IE6, etc.
• Debugging with FireBug is possible even for the compiled version of the source code.
• Compiler supports generation of dynamically loadable modules, which can significantly speed up loading of the end application, because the code for advanced functionality can be loaded only when it is required.

For details have a look at: http://blog.klokantech.com/2010/12/closure-compiler-for-openlayers-3x.html

Edit: take a look at this youtube video it may answer some questions about Google Closure better.

Probably the best sources of information on google closure are project discussion group, wiki, doc pages, demos and a yet unfinished book by Michael Bolin that is now available from safari books site.

one thing I can tell right away - there is a steeper learning curve for closure vs jQuery but it may be well worth it due to the library's vastness, clear organization and the benefit of using it together with the compiler and the templating tool.

closure library in that respect is more like dojo than jQuery, and some concepts were borrowed from dojo, according to Michael Bolin.

google closure compiler uses JSDoc documentation system which simultaneously (if created by the programmer correctly) provides documentation and enables catching many errors at compile time.

while function names are more verbose than jQuery's, the compiler shrinks the code (using various optimization tactics) and the type checking will save a considerable time debugging the code, so time typing in the longer names is probably not an issue. At the same time longer names add readability.

library supports browsers running in the quirks mode so that scripts could be embedded by other sites using "quirky" html

library works with (but does not depend upon) a javascript templating system called soy that simplifies filling documents with content.

like jQuery google closure allows traversing dom structure with the string-based queries using a dedicated component of the library.

closure library relies on dot-delimited namespaces more like Java - a very strong organizational feature.

using such namespaces will incur overhead in uncompiled code, but in the compiled code those things are replace with short variable names.

I just posted a pretty exhaustive article about Google Closure which answer this question on insideRIA.

...Closure rulez! ^_^

Maybe I'm not getting jQuery, but I haven't seen a real UI widgets collection there (there are plugins, yes, but you never know how well-tested they are and often there is no clear winner and/or the plugin lacks documentation).

Closure has, among other things, a widgets collection (see the demos tab), including, say, imageless buttons used in gmail.

More generally, it has more functionality implemented as part of the release. It may not be a big thing, but I get annoyed with the sea of jQuery plugins when I'm looking for something as simple as a ajax history module or autocomplete.

Overall it's a huge library + set of tools and I'll be getting acquainted with it just to know what's available.

I appreciate most of Google's contributions to the open source community and I'm sure it's got some cool stuff but overall I find Closure bulky, overwrought, and inelegant. If you must turn everything into Java I suppose it's downright spiffy.

Edit:

Fair enough. I didn't really compare. Closure is like a giant warehouse with every possible tool you could possibly want located... somewhere. Sort of like .NET or a massive Java library. Once you find what you need, you can find highly specific stuff that does highly specific jobs. And then for production you can remove all the cruft.

JQuery on the other hand, is more of an easily modified sonic screwdriver.

The Google Closure Library allows you to compile and optimize your JavaScript. It's not a library like jQuery is. jQuery is something that provides you with functions that allow you to write your own javascript faster.

Google Closure would help you make your own javascript code minimized to allow for a speedier delivery over the Internet.

Long story short, Google Closure is a tool while jQuery is a library similar to Prototype.