TomSoft

Widgets: the next portability nightmare?

October 16th, 2006

The topic of Widgets standardisation will become a hot topic next year. We already haves so many different Widget framework based on browser (Google, Opera, NetVibes, WebWag), a few desktop based Frameworks (Yahoo!Widget and Dashboard), and now some emerging mobile widget framework (Widsets, and of course Mobidgets! ) that it’s quite sure that we can not continue into the same direction.
In this post, I will try to provide a view on the landscape of widgets in terms of interoperability (between platform) and try to explore the common part between theses framework.

First, I’ve been to “Widgipedia“, and from their FAQ, they already support these 14 platforms: (this was more than I was expecting)

• Yahoo! Widgets (Windows)
• Yahoo! Widgets (Mac)
• Dashboard (Mac)
• Standalone Widgets (Windows);
• Standalone Widgets (Mac)
• Microsoft Vista Sidebar (Windows)
• Opera (Windows)
• Opera (Mac)
• Opera (Linux)
• DesktopX (Windows)
• Kapsules (Windows)
• Samurize (Windows)
• KlipFolio (Windows)
• AveDesk (Windows)

I assume that some of them are interoperable, especially between the mac/windows version, but it seems that it’s not always the case.
Also there are some missing components, like Netvibes, or WebWag which have their own framework, and probably Windows!Live? Flash is also a probable candidate for a Widget environment….
Mobile applications have also their own frameworks, for instance Widsets or Mobidgets

So what levels of interoperability do they have. For now, none of them are interoperables!. When you do a Google widget, than you have to recreate it for Yahoo!Widget for instance.
The positive view is to say it’s not so hard to convert a widget from one environment to the other….
Is there any hope that this will change in the near future. I am not an expert of each of these widgets framework, but we can identify two major things:
the “representation/descriptive part ” generally XML based
the scripting language used is JavaScript/ECMAScript…

That’s a good start, but then you face all the little details: where are stored the user preference, how to access to some utility function, is it CSS/HTML based, etc….

So what’s next? Three options:

• One of these framework become the de-facto standard, and all the others are either killed, or align to it….
• A standardisation body (W3C?) take the lead on it and try to solve the problem….
• Status quo, only a few widgets engines/framework are still alive and tools are created to “transcode” one widget into another?

None of this solution is perfect, but it will be interesting to see how this will evolve in the coming monthes, especially in regard of mobile who will bring new widgets framework, but who will be hardly compatible directly with other framework for many technical reasons….

I am sure that it will be an interesting topic to discuss in the upcoming “Widgets Live!” conference in SF! (which is the same day than this interesting mobile2.0 event, in SF too!)

And in the second part of this post, I will explore potential solution and direction for a common widget framework…Stay tuned.

Update: one the reader (Lindsey), pointed out this work on W3C about packaging deployment Web Application Packaging Format Requirement. This doc is a starting point, but contains a good table about various packaging format used between Google, Yahoo and others…

Read also this post from C.Enrique some monthes ago on the same topic….

Picture taken from the “hyalo-weather” widget….

Technorati Tags: widgets, mobilewidgets, standardisation

Entry Filed under: Ajax, Google, MobileAjax, MobileWidgets, Yahoo