gwt-ext vs. mygwt

23 views
Skip to first unread message

Garrett

unread,
Dec 13, 2007, 9:59:13 AM12/13/07
to Google Web Toolkit
Hi guys,

I've spent a fair amount of time with mygwt and it's pretty cool. I'm
planning to do the same with gwt-ext, but I'd love to hear from anyone
who has done this already. gwt-ext looks very impressive from the demo
(thanks to ext itself), but the approach is much different than the
native mygwt, so it definitely seems to have trade-offs.

- Complexity to code
- Ease of integration with backend services (RPC especially)
- Performance
- anything else (the good, bad and ugly)

thanks in advance!

Dimitri Roschkowski

unread,
Dec 13, 2007, 12:03:29 PM12/13/07
to Google Web Toolkit
Now it's about 2 month, I started to work with GWT. First I used mygwt
to get some advanced widgets, but I was missing something. So I
started to use gwt-ext. It has more widgets than mygwt, so I changed
the whole mygwt code to gwt-ext. And that's already the answer on your
first question.

> - Complexity to code
The complexity is quite the same. Often I just had to change the
import file to change mygwt -> get-ext. So the complexity is quite the
same.

> - Ease of integration with backend services (RPC especially)
gwt-ext has its own backend integration for forms. I didn't spent much
time on that kind of staff, I just used the getValue() method to get
the Value und put it through the GWT RPC Service to my servlet server.

> - Performance
Well, performance. gwt-ext is based on Ext, so it need some additional
ressources. My application is currently in development state and not
optimized, it loads about 500kb js and 400kb css files on the first
run. The performance in the application itself is ok.

> - anything else (the good, bad and ugly)

There are some ugly things in gwt-ext (especially with forms), but
nothing that's unsolvable.

Reinier Zwitserloot

unread,
Dec 13, 2007, 12:22:51 PM12/13/07
to Google Web Toolkit
MyGWT has better performance, has much better integration with GWT,
and is in my opinion somewhat better designed.

GWT-EXT, on the other hand, is the fastest, simplest, brute force
route to bridging GWT to Ext. It is therefore a lot closer to the
latest Ext release and includes more widgets.

Personally I'd go with mygwt unless there's a thing mygwt doesn't do
but which is supported in gwt-ext.

shagwt

unread,
Dec 13, 2007, 11:11:01 PM12/13/07
to Google Web Toolkit
extjs 2 and a gwtext wrapper around it should be able to address
design issues as they move to a component model.
my experience with ext

maturity - ext js has a wide user community , support etc. gwt-ext
forum on google groups is an active community.
native integration - ext is not very natively integrated to gwt way
of doing things. rather a quick wrapper.
interoperability with other gwt components = is possible and if there
are issues, you can get around them
rpc integration = ext does provide a rpc store integration but its a
commercial license.
browser compatibility, accessibility -- with proper use of browser
effects you could get some ada compliance going. browser compatibility
should come with ext.
ease of use - yeah.. two thumbs up to gwt ext sanjeev jivan . we
have junior developers with basic java know how writing thick client
console applications!


hth,

haggett

unread,
Dec 14, 2007, 9:03:22 AM12/14/07
to Google Web Toolkit
For a large enterprise application we are developing for an aerospace
customer, we chose GWT and MyGWT. We evaluated gwt-ext, but one
important factor ruled out its use -- the need for Javascript/DOM
expertise.

We originally selected GWT because our developers could concentrate on
Java. In prior projects we needed Java developers and Javascript/HTML
developers. With GWT, it's just Java (yeah, and a little CSS for
sure).

We chose MyGWT because it is built on GWT, rather than "going around
GWT" as gwt-ext. We worried about how gwt-ext would integrate with our
testing framework and the GWT event model.

Having worked for a few months now with MyGWT, I can strongly
recommend it. It's solid and looks good out of the box. Darrell
provides excellent response to users and manages to push out updates
very frequently.

Cheers, Allan

bwtaylor

unread,
Dec 14, 2007, 12:36:01 PM12/14/07
to Google Web Toolkit
Ditto, except ours is a large internal enterprise app.

Pavel

unread,
Dec 14, 2007, 1:16:55 PM12/14/07
to Google-We...@googlegroups.com
MyGWT looks extremely nice but the only reason I went with gwt-ext is
the Drag&Drop support in trees... I use it in all of my current
projects. I'd say that if MyGWT adds D&D to trees I'll seriously
consider switching.

Pavel

nico...@gmail.com

unread,
Dec 15, 2007, 9:56:31 AM12/15/07
to Google Web Toolkit
Well I decided DON'T USE THEM becuase I try both libraries demo page
and they didn't work with Konquer :(


On Dec 13, 3:59 pm, Garrett <gsu...@gmail.com> wrote:

Casey

unread,
Dec 15, 2007, 12:17:00 PM12/15/07
to Google Web Toolkit
I don't want to start a flame war but are you really willing to throw
away two great libraries because Konqueror doesn't work properly? Do
any great web applications work with that browser? I am talking about
stuff like Google and Yahoo mail, Google reader, etc... According to
Yahoo it's not even a grade A browser (http://developer.yahoo.com/yui/
articles/gbs/). Also, since you are probably one of the few Konqueror
GWT users that is looking at GWT-EXT and MYGWT would you be willing to
submit bug fixes for either library? Personally I think patching MYGWT
would be a lot easier since its native code and not a hybrid. Again, I
don't want to start I flame war, I just think that you should look at
how many of your target users are not using a grade A browser. Then
again, maybe your target demographic is KDE users, which in that case
you would be correct.
Reply all
Reply to author
Forward
0 new messages