Link Details

A couple of things.

1. Todd says Flex is not open source. That is not true. I'm no flex fan but let's at least be honest.
2. Greg mentions a Stock Ticker application as being in the tutorial however, I don't see it on https://pivot.dev.java.net/nonav/tutorials/index.html at all. I'd really like to see how the data binding works.

I see now the Stock Ticker app is on the main Pivot page. My bad. :)

Gregg,

This is Todd Volkert. My bad on Flex -- when we started development on Pivot, Adobe had not yet announced their plans to open source Flex, and we unfortunately didn't keep up to date on its open source status.

As for the stock tracker, you can also browse the source for it here: https://pivot.dev.java.net/source/browse/pivot/trunk/tutorials/src/pivot/tutorials/stocktracker/

Got it. Thanks Todd. I'm 80% through a rather large commercial Swing application and its been a nightmare. I've been looking at other tools for Version 2.0 when the time comes so I'll dig deeper into Pivot in the coming months as I try and determine our approach.

Todd,

Pivot looks really interesting in comparison to Swing applets. It seems as though JavaFX will be built on top of Swing and though it will make things nicer to use through JavaFX Script, that will only be so because of additional bloat. So a person who wants to do a RIA applet with JavaFX and the new Java 6u10 Java kernel would end up with users who have to download all of the 6u10 kernel, then all of Swing (big!), then all of JavaFX, then all of your applet classes/jars. In contrast, with Pivot, an end user would only download the 6u10 kernel, Pivot (which is very small), and my applet classes/jars.

I'm sure I'm preaching to the choir, but it seems like ditching Swing for a lighter weight framework such as Pivot is extremely wise if a technology wants to compete with Flash/Flex. So hats off to you guys.

My single complain with Pivot in its current form is a poor/boring look & feel. The default look & feel for Flex apps is great. I have decent Java2D skills, and I had kicked around the idea of creating a Flex-like L&F for Swing, but I'm wondering if Pivot would be the better code base to contribute that kind of work. Are you guys open to (heavy) L&F contributions?

-Bryan

Bryan,

Thanks for the kind words. We are certainly open to new L&F contributions! The best place to discuss this in detail is the Pivot forums, so I've created a discussion thread to continue this discussion: https://pivot.dev.java.net/servlets/ForumMessageList?forumID=3178.

I'd loved if we had a toolkit called Swing 2. Where all the Swing concepts are
reconsidered and rewritten from scratch. Pivot is something in that direction
and in general the idea is definitely good. On the other hand there is a set of
problems should be solved by such kind of toolkits:
- Default look and feel should be smarter. I even consider that mostly people
won't need a L&F switching impossibly is the default one is good)
- Visual GUI builder is needed
- Business-oriented components are necessary (lookup fields, advanced tables, tree tables etc)
- Docking framework is needed
- Something like Application framework is required
- Code refactoring should be supported between XML and Java

The mentioned issues are somehow solved in Swing already or
at least some ready libraries exist. So, there is quite a long way to
make such thing like Pivot mature. At the moment I'm not sure if
such task is solvable by volunteers. It would be nice if Sun or some
other big company invested in such thing. But Sun is now investing in FX.
May be the Pivot is a technology of future but may be a kind of experimental
technology for fun :) In any case I wish you all the best guys. I believe
that the community will anyway benefit from such projects like yours.

Sorry, a small correction the sentence about L&F should be:

- Default look and feel should be smarter. I even consider that mostly people
won't need a L&F switching possibility if the default one is good.

Hi Maxim-

An updated L&F is definitely on the priority list for a future release. To your other points:

- Visual GUI builder is needed
[GB] It's pretty easy to build a Pivot app just using WTKX. A visual tool would be nice, but would require a significant time investment that we aren't currently able to justify.

- Business-oriented components are necessary (lookup fields, advanced tables, tree tables etc) -
[GB]
- Lookup fields are pretty easy to build - the tutorial includes an example of one.
- Pivot's tables already support some fairly advanced behavior - what specifically do you think is missing?
- It is possible to build a TreeTable in Pivot - we plan to write up an example at some point. However, like Swing, we probably won't include it as a standard widget.

- Docking framework is needed
[GB] Dockability is a nice feature for certain types of applications, but lots of other applications wouldn't get much benefit from it. As such, providing support for docking in Pivot has not been a priority.

- Something like Application framework is required
[GB] Specifically, what do you have in mind? Pivot is itself an application framework - what do you think is missing?

- Code refactoring should be supported between XML and Java
[GB] Not sure what you mean by this.

To your point, though - Pivot *is* less mature than Swing or SWT/JFace; we haven't even officially released version 1.0. :-) However, we think we have created a solid foundation upon which features like this can be built. We are looking to the developer community to help us build them because a) we don't have time to do it all and b) we can't possibly envision all of the different ways in which a developer might want to use or extend the platform. We'd love to see some of the larger organizations help support and contribute to it.

-Greg