incubator-cvs mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Apache Wiki <wikidi...@apache.org>
Subject [Incubator Wiki] Update of "PivotProposal" by GregBrown
Date Fri, 19 Dec 2008 18:18:20 GMT
Dear Wiki user,

You have subscribed to a wiki page or wiki category on "Incubator Wiki" for change notification.

The following page has been changed by GregBrown:
http://wiki.apache.org/incubator/PivotProposal

New page:
= Abstract =
Pivot (http://pivot-toolkit.org) is an open-source platform for building rich internet applications
in Java.

= Proposal =
Pivot combines the enhanced productivity and usability features of a modern RIA toolkit with
the robustness of the Java platform. Pivot applications are written using a combination of
Java and XML and can be run either as an applet or as a standalone, optionally offline, desktop
application.

Like other modern development platforms, Pivot provides a comprehensive set of foundation
classes that together comprise a "framework". These classes form the building blocks upon
which more complex and sophisticated applications can be built.

Pivot was designed to be familiar to web developers who have experience building AJAX applications
using HTML, CSS, and !JavaScript. However, it provides a much richer set of standard widgets
than HTML, and allows developers to create sophisticated user experiences much more quickly
and easily. Pivot will also seem familiar to Swing developers, as both Swing and Pivot are
based on Java2D and employ a model-view-controller (MVC) architecture to separate component
data from presentation. However, Pivot includes additional features that make building modern
GUI applications much easier, including declarative UI, data binding, effects and transitions,
and web services integration.

= Background =
The web has become the defacto standard method for application delivery. However, functional
requirements for web applications have begun to scale beyond the capabilities of the browser.
Even with the addition of scripting support, dynamic element manipulation, and asynchronous
server communication, it is difficult to create a user experience in HTML that is truly on
par with that of a desktop application.

Rich internet application (RIA) development platforms are a means of bridging the gap between
the web and desktop experiences. Using browser plugins, these platforms allow developers to
build applications that look and feel more like native desktop applications but are deployable
via the web, like traditional, HTML-based web applications. RIAs also often incorporate visual
effects intended to enhance the overall user experience, such as animations and other dynamic
elements.

Adobe Flex (http://www.adobe.com/products/flex) and Microsoft Silverlight (http://www.microsoft.com/silverlight)
are arguably the most high-profile of these platforms; others include !OpenLaszlo (http://www.openlaszlo.org)
and Curl (http://www.curl.com). Pivot itself falls in this category.

= Rationale =
Pivot was created for two primary reasons:

 1.#1 To provide a viable option for developers who want to build rich client applications
in Java. While it is technically feasible to build a Java-based RIA using the Swing toolkit
(http://java.sun.com/docs/books/tutorial/uiswing), Pivot offers a number of advantages that
make it a more compelling, modern alternative:

  * Provides an XML markup language called WTKX for simplifying user interface construction.
Flex, Silverlight, and !OpenLaszlo all offer a similar feature; web developers are comfortable
with the markup metaphor, and it can considerably reduce overall development time.

  * Components are not limited to an "atomic" preferred size; they are allowed to report a
preferred size as constrained by either width or height - this facilitates such features as
label wrapping, which Swing does not support.

  * Defines a consistent data model that is used throughout the entire framework; for example,
JSON data returned from a REST service is serialized into the same data structures used by
a table view component to present data. No additional translation is necessary, which can
significantly improve performance. A common data model also reduces the learning curve for
new developers.

  * Includes built-in support for REST-based data services, which Pivot calls "web queries".
This provides parity with Flex, which comes with out-of-the-box support for RPC via the AMF
protocol, and Silverlight, which supports both SOAP and REST-style services. Swing does not
include any built-in facilities for server communication, making it less convenient to work
with.

  * Includes built-in data binding support, which allows data returned from web queries (as
well as other types of data services) to easily be mapped to form contents.

  * Includes platform-level support for visual effects and transitions (i.e. animations).

  * Defines a single Application inteface that is used for deploying both desktop and web-based
applications - multiple codebases are not required.

  * Takes advantage of newer Java language features such as generics, enums, for..each, and
annotations.

 Pivot differentiates itself from the recently-released JavaFX primarily by allowing developers
to build applications in Java, rather than the new JavaFX scripting language. Additionally,
JavaFX's widget support is based on Swing, which suffers from the limitations outlined above.
In a sense, Pivot represents what we think Sun should have done instead of JavaFX.

 1.#2 Provide a freely-available, open source alternative for RIA developers. Flex, Silverlight,
and Curl are all proprietary platforms. We believe that a large part of HTML's success was
its due to its openness. While we certainly hope that developers will use Pivot to build revenue-generating
products and applications, we believe that the platform itself should be free and driven by
its technological merits, not by corporate objectives.

= Current Status =
Pivot began as an R&D effort at VMware. It  was announced as an open-source project in
June of 2008 under the Apache 2.0 license. Version 1.0 was released in October, 2008, and
version 1.1 is targeted for release in early 2009. Pivot is currently hosted at http://pivot-toolkit.org
with development at http://code.google.com/p/pivot.

== Core Developers ==
Thus far, Pivot has been developed primarily Greg Brown and Todd Volkert of VMware.

== Community/Meritocracy ==
The Pivot team is actively striving to build a development community around the Pivot platform,
consisting of both users and contributing developers. Although we are pleased with the progress
we have made to date, Pivot is a large undertaking that needs the support of a larger developer
base if it is to succeed and thrive.

We have written numerous articles and blog entries about Pivot to help generate interest in
the developer community, and will continue to do so as new features are added and new versions
released. We have established a Google Group for discussion of Pivot-related issues (http://groups.google.com/group/pivot-toolkit)
where feedback is solicited and highly encouraged.

As developers (hopefully) begin to express an interest in using and helping to expand Pivot,
their contributions will be evaluated and incorporated as appropriate. Several contributions
from several external developers were actually accepted before we released version 1.0 - notably,
both the Pivot Explorer tool and the pivot.core.util.Resources class used for localization
were provided by outside developers who volunteered to join the project after reading our
initial announcement.

== Alignment ==
We use Ant to build and deploy Pivot, and we host our web-based components in Tomcat.

The Apache developer community is representative of our target audience - developers who seek
out and advocate high-quality, open-source software projects. We consider ourselves to be
members of that group, and we are hoping to share our work with other like-minded developers.

= Known Risks =
== Orphaned Products ==
This project will not be orphaned. The developers have already invested a significant amount
of time in the project, much of it in off-hours. Greg has been working on Pivot since August
2007 and Todd since January 2008. We pay to host the project site, pivot-toolkit.org, out
of our own pockets. Additionally, at least one production application is currently being developed
with Pivot. We want to see this project become successful, and we will continue to invest
whatever time is necessary to help make it so.

== Inexperience with Open Source ==
Pivot has been conceived as an open-source project since day one. However, though we are longtime
users of open source software and have some understanding of the process, we have no prior
experience managing an open-source effort. We are hoping that the community experience within
ASF can help bring our understanding to the next level.

== Homogenous Developers ==
While Pivot's primary developers work for VMware in Boston, Massachusetts, Pivot is not a
VMware project. Most Pivot development takes place outside of work hours.

Pivot is a highly-modularized framework that lends itself well to heterogenous development.
We have already received code submissions from developers elsewhere in the U.S., Germany,
Italy, Canada, and the U.K. We are hoping that, as the capabilities of the framework expand,
it will lead to an increase in overall developer interest and an acceleration of the development
process.

== Reliance on Salaried Developers ==
Pivot began as an off-hours side project. It was officially supported by VMware for a short
time, but was dropped in June, 2008 when the company decided to pursue an off-the-shelf UI
solution instead of a homegrown one. It has continued as an off-hours project since then.

The "lack" of an ability to rely on salaried developers for development may actually be more
of a risk to the project. Most work must be done in off-hours, which may occasionally limit
our ability to make progress.

= Documentation =
Additional information is available from the project site: http://pivot-toolkit.org.

= Initial Source =
Source code for Pivot is currently hosted at Google Code: http://code.google.com/p/pivot.

= Source and Intellectual Property Submission Plan =
VMware released Pivot as an open-source project under the Apache 2.0 license in June, 2008
but retains a copyright on the code. VMware has agreed to sign a software grant for the codebase.

= External Dependencies =
Pivot is a Java-based toolkit and relies on a JRE for execution. Pivot's charting library
currently relies on J!FreeChart (http://jfree.org) 1.0.9; however, charting is an optional
package and J!FreeChart is not integral to the implementation (i.e. it could be replaced with
another charting package with no impact to the Pivot charting API).

= Required Resources =
 * Subversion repository
 * Issue tracking
 * Dev., user, commits, and private mailing lists
 * Wiki space
 * Web hosting (for content currently hosted at pivot-toolkit.org)
 * SSL certificate (for code signing)

= Initial Committers =
 * Greg Brown, Todd Volkert - VMware (http://vmware.com)
 * Eugene Ryzhikov - Florida Power and Light (http://www.fpl.com)
 * Christopher Brind - unknown
 * Sandro Martini - Technomind (http://www.technomind.it)

= Affiliations =
No relevant affiliations.

= Sponsors =
 * Champion: Niclas Hedhman
 * Nominated Mentors: Niclas Hedhman
 * Sponsoring Entity: Incubator PMC

---------------------------------------------------------------------
To unsubscribe, e-mail: cvs-unsubscribe@incubator.apache.org
For additional commands, e-mail: cvs-help@incubator.apache.org


Mime
View raw message