On 17 nov, 03:25, zixzigma <zixzi...@gmail.com> wrote: > I agree with all the issues that were pointed out about Sencha/Ext. > > However, comparing GXT to GWT is comparing apples to oranges. > > GXT is only a widget library, you can certainly build applications out > of just putting together some widgets. > > but to develop scalable apps with great user experience, you need more > than just bunch of pretty looking widgets. > you need architecture. thats where GWT with its MVP, Editor, Server > Communication mechanisms come to play. > > i didnt suggest a cookie cutter widget solution, that you just drag > some widgets into your IDE, and write few lines of event code, and bam > youre done. > ofcrouse not.
I understood you were asking for widgets, not RAD; so saying now that we couldn't compare GXT with GWT (which I'd then understand as "GWT's widget set") puzzles me... > a powerful widget library can "Complement" GWT. > and you would have the same flexibility you have now as how you want > to design your app. [...] > asking deverlopers to build their own widget libraries, > is just plain wrong. goes against re-usability, maintanance, > interoperability, etc. > you well know these widgets must be tested on a number of browsers, > each with different versions, > your core application is one thing, but you spend your efffort > elsewhere. > talk about re-inventing the wheel. Can you list a few widgets you're missing? (and which are not in GWT- Incubator) > do you write your own String library from scratch on every project or > in every company you go to? > do you implement your own LinkedList or HashMap ? > do you write your own I/O library ? your own web framework ? > ofcourse you can, you can create your own programming language or > operating system if you want. > but in a real world, with tight deadlines, and whole set of other > architectural decisions to deal with, > building drag and drop for tree is the last thing on your priority > list. > if this is your line of work or something you enjoy, then that's > great. Drag'n'drop is hard to develop in an "accessible" way; you'd probably have to complement the functionality with some kind copy/cut/paste or move up/down/level-up/level-down. And if you don't need it in your projects, people would blame Google for not providing accessible alternatives. I've always considered D'n'D as a "bonus", an alternate way of doing things, so saying "it's not possible" or "it'll cost twice as much" to our clients has never been an issue for me. I've seen (desktop) apps where D'n'D were the only option, that wasn't intuitive at all (imagine, a "double list" similar to this one <http:// www.cs.helsinki.fi/u/salaakso/patterns/Double-List.html> without the "move" buttons in between) (in addition to not being "accessible"). Of course, YMMV. FWIW, YUI's TreeView doesn't do D'n'D either: http://developer.yahoo.com/yui/treeview/ and YUI's DataTable while supporting resizable and reorderable columns (via D'n'D) requires a bunch of work to make reorderable rows: http://developer.yahoo.com/yui/examples/datatable/dt_ddrows.html (and for GWT, you can try GWT-DND, by Fred Sauer, Google Developer Advocate on AppEngine and... GWT!) > Software Development with GWT is more than just creating widgets, > the challenge is is how you put together everything in a sound end to > end architecture. > Still when it comes to User Interface design, if you have a powerful > widget library, you can implement more complex user interactions. > the core widgets provided by GWT are good, but not good enough. > i can see 10-20 variations/features on each of the widgets. and many > more. Wouldn't there then be a risk of a bloated API, and a bloated widget set that doesn't optimize nicely in all cases? > i agree on this that we as a community can participate and > collectively contribute to this. > this requires a bit of change in the way the things are handled. > > i am not sure if you are familiar with Vaadin.http://vaadin.com/directory > one thing i like about them is the way they get the community involved > and how they make it easier for the developers to submit their add-on/ > plugins. > they have an official section that users can submit the plugins they > have implemented. > other developers who use the plugins can provide feedback. > so instead of "scattered" google code projects that end up with no > activity after a while, > there would be a "central place" where plugins/widgets are submitted, > reviewed. > incubator seems to have this purpose, but i am not sure how successful > it has been so far. You mean something like the GWT Gallery? http://code.google.com/webtoolkit/community.html -- You received this message because you are subscribed to the Google Groups "Google Web Toolkit" group. To post to this group, send email to google-web-tool...@googlegroups.com. To unsubscribe from this group, send email to google-web-toolkit+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/google-web-toolkit?hl=en.
