Open Source AJAX toolkits

twofish writes "InfoWorld columnist Peter Wayner recently reviewed six of the most popular "open source" Ajax toolkits. The article sets out to see if they are enterprise ready in comparison to commercial products such Backbase, JackBe, and Tibco's General Interface. The six open source projects covered were selected because each has a high-profile in the developer community and support of one or more stable organizations. "

The toolkits covered are:

1. Dojo
2. Google Web Toolkit
3. Microsoft Atlas
4. Open Rico and Prototype
5. Yahoo AJAX Library
6. Zimbra Kabuki AJAX Toolkit

Whilst the definition of open source is broad, the round-up is quite helpful.

Java != Javascript

[andrewman327]

From TFA: "[...] JavaScript is pretty close to a superset of Java[...]. It's not complicated to strip away some typing information from the Java code and end up with something that resembles JavaScript."

This is in response to Google's toolkit, which allows users to code in Java instead of Javascript. I think this feature is a real winner to Java coders. Who wants to code Javascript when you can use Swing? Regardless of what TFA says, there is a difference between the two programming experiences.

In summary, if you are already proficient in Java, Google is the way to go.

Re:"Open source?"

[achacha]

The only reason large corporations push some toolkit as "open source" is because:

1. It's a crappy product that their marketing people cannot justify as promotion cost
2. There are better free products
3. They are trying to get their foot into the niche so they can then charge for the "Professional" version
4. They don't understand the space yet

This is common for Microsoft and now becoming common for Google.

Sadly AJAX is still the "silver bullet" of web based companies and the buzzword of the moment. So many companies are using AJAX for the sake of using it despite the fact it is not applicable to the ir use case; sometimes it is easier to wedge something in and use a buzzword to sound cool and relevant.

Re:Java != Javascript

[StarvingSE]

The author of TFA is just dumb and doesn't know what he is talking about. First he says that Microsoft Atlas is open source. Then, it sounds like he truly believes that Java and Javascript are related in some way. Besides some similar syntax, they are both mutually exclusive.

when are people going to realize that Javascript and Java share only a name???

Re:Java != Javascript

[andrewman327]

The confusion of Java and Javascript is one of my biggest pet peeves in computer science. I am fairly proficient in Java, but I still have to look up which command to use the once a year I actually write in Javascript. Google's engineers worked hard to design a system to convert Java into another format only to have this journalist completely disregard it.

It's times like these that I am glad I get to tag articles.

Re:Erm...

[LiquidCoooled]

<i>and for all other effects write your own functions just like always (copy/paste from your personal library and adapt)</i>

Or you just do exactly what digg does and take your own javascript library and include everything you possibly can do "just in case".
I'm actually surprised kitchenSink.js isn't included.

This is just an example from the standard front screen of digg without any cookies or logins to concern itself with.

  <script src="/js/spellChecker.js" type="text/javascript"></script>
  <script src="/js/utils.js" type="text/javascript"></script>
  <script src="/js/xmlhttp.js" type="text/javascript"></script>
  <script src="/js/comments.js" type="text/javascript"></script>
  <script src="/js/wz_dragdrop.js" type="text/javascript"></script>

  <script src="/js/hover.js" type="text/javascript"></script>
  <script src="/js/label.js" type="text/javascript"></script>
  <script src="/js/dom-drag.js" type="text/javascript"></script>
  <script src="/js/switcher.js" type="text/javascript"></script>
  <script src="/js/prototype.js" type="text/javascript"></script>
  <script src="/js/scriptaculous.js" type="text/javascript"></script>

  <script src="/js/lightbox.js" type="text/javascript"></script>
  <script src="/js/aboutdigg.js" type="text/javascript"></script>

Re:"Open source?"

[Anonymous Conrad]

Last I checked, neither Atlas nor GWT were open source in any sense of the word,

But you can download the Atlas source code [codeplex.com] and at first glance the licence [codeplex.com] meets the Open Source [opensource.org] definition: it's a simple no endorsement, no liability, no patent disputes licence. So what's the problem?

Re:A better review (w/ actual code samples)

[wranlon]

I've always liked my own AJAX framework, Engine for Web Applications [imnmotion.com], but it never seems to make it farther than the appendices (if even) - here are some good toolkits, see appendix A for some other stuff that showed up in Google.

If you're interested in JS toolkits... (Dojo, etc)

[ChrisZermatt]

...make sure you check out qooxdoo.

Its not the best known, but its one of the most promising toolkits in [very] active development. I've been involved (sort of -- following the mailing list) and its open source & very slick.

http://www.qooxdoo.org// [qooxdoo.org]

The 0.6 release is expected in the next day or so, and is a big jump over 0.5. The only area that is still a bit weak is the documentation, but there is a good group of developers working actively on getting that properly sorted for the next release.

Re:Erm...

[tenchiken]

As someone who has tried to do what you suggest, and then worked in pain to deal with all of the cross browser issues, the strange XMLHttpRequest behavior, systems for relability,etc, the bloat is well worth it.

Should have reviewed DWR

[bryanbrunton]

As many have noted the article is really quite clueless. However, any review on Ajax toolkits is not complete with a mention of Direct Web Remoting [getahead.ltd.uk].

Central idea behind DWR is it exposes methods of Java Beans over the web. Create a server side class and then call methods from javascript like this: MyBean.method(). It couldn't be simpler.

I have used DWR in my just released online version of Risk, called Grand Strategy [denizengames.com].

Re:Java != Javascript

[Selanit]

An early development version of JavaScript was code-named "mocha." All the way through the old 4.x series of Netscape Navigator, you could access the JavaScript console by typing "mocha:" in the address bar. How I wish they had just adopted that name for the language as a whole! It would have prevented so much confusion.

Re:Java != Javascript

[larry bagina]

Mocha was renamed to LiveScript, which was then renamed to JavaScript (and later ECMAScript). "JavaScript" is actually a Sun registered trademark. When JS first made it's oh-so-buggy appearance, I thought Netscrape was trying to jump on the Java hype, but I think Sun paid them to change the name.

Re:"Open source?"

[hey!]

5. Because they can't allow a hostile competitor to obtain control over a piece of software infrastructure that is critical to them.

i.e., they have learned the lesson of Borland.

e.g., Oracle can't survive in the long term if Microsoft gains control over server platforms

e.g., IBM can't survive inthe long term if they have to use Microft's own tools to complete with it.

So: yes, support of open source is self interested in cases like these. But not necessarily cynical or pernicious.

Re:Java != Javascript

[Anonymous Coward]

And now it's part of Java (1.6 comes with Rhino), and it pretty seamlessly loads and scripts java classes. In fact, interop with Java was part of LiveConnect. So "nothing to do with java" is basically an ignorant tirade tossed out by people who expect it had to actually be java and not just script it, which it did ever since the name was changed.