Whenever I go to a major conference or I deal with social collaboration or web 2.0 tools, I see a lot of widget options. Here’s my take on the whole widget vs portlet question:
1. Widgets are great because they are easy to create and use RESTful services.
2. Widgets do have a problem if you are trying to create a more enterprise type portlet. If it will have a lot of transactions and you want to bullet proof it, you will probably want to create a portlet.
3. Portlets work best when you have interaction between portlets and pages. That’s not to say that widgets don’t pass data parameters around but portal and JSr 286 have matured enough to make it much easier to user.
4. Widgets aren’t going away anytime soon. IBM has widget based solution in their mashup server. Oracle has the Ensemble product which should still be in the Fusion stack.
5. Portal vendors are supporting mashups in their products. IBM has support for widgets via a portlet. Liferay has a decent blog post on the topic as well. The GateIn Portal (formerly JBoss Portal) also provides gadget support.
Also, I found a good blog post on a deeper view of widgets vs portlets at the WebSphere Portal Blog.
I’m probably going to change my mind at some point but right now I don’t see widgets or portlets going away soon. Portal vendors are going to support the whole widget approach as it matures. Portal developers are going to continue to develop portlets and use existing portlets already developed. Widgets become just another option to use on your portal or a way to even push portal functionality out of your portal.
本文探讨了Widgets与Portlets在企业应用和社会协作工具中的不同作用。Widgets因其易用性和RESTful服务的支持而受到欢迎,但对于高交易量的应用,Portlets提供了更稳定的选择。此外,文章还讨论了门户产品如何支持Widgets并利用Portlets实现组件间的交互。
57

被折叠的 条评论
为什么被折叠?



