I completely agree with you in the fact that there are already some standards for widgets out there, although i don’t think there are that many in mobile (I’d say Nokia Widsets, Plusmo, MojiPage and Yahoo Mobile Widgets are about it, recently including Nokia S60 widgets)

For me things like widsets and plusmo aren’t really “widgets”, they’re (as you mention) nice displays of rss feeds and little more. Yahoo and Nokia S60, on the other hand, include more complex APIs that allow you to do multi-screen apps, more complex input-handling and requests, and even phone integration.

The counterpart of S60, from my point of view, it’s the fact that it is liked to having a Nokia S60 phone, and that is huge distribution for europe, but not so good worldwide (check latest Symbian market studies here) so the idea is to have a platform that is wide enough not to be limited by devices/manufacturers, and I think that is where Yahoo widgets is aiming at

Of course, current support is far from perfect, but at least that is the original goal. Also, as contrary to what happened with Go 2.0, Yahoo Go 3.0 and the yahoo mobile widgets can be executed both in the Java app AND ALSO IN THE BROWSER. When you subscribe to a new widget in your Java app, you can check the exact same widget gets subscribed to if you access through your browser via http://beta.m.yahoo.com. I think that is another of the big advantages: you code once, get access to both app and browser

Of course being “limited” to the browser makes you not have a lot of cool stuff (like access to low level phone features), but in my mind, this is a similar situation to what we had in web apps some years ago: you could go using microsoft activex and have low level access, or use standard html and have little more than post/get abilities… but it was the latter approach that actually drew in more adepts and that shaped the panorama of web apps that we have nowadays…

I would love to see the same rich apps market happening in mobile apps, but for reaching that we need critical mass… and for critical mass you need maximum reach… and for maximum reach you need to restrict your low level access…

Well, there’s my 2 cents… what do you think?

This was something I had some interest in doing last year, even to the point of grabbing a few domains. I’m interested in widgets, but mainly along the lines of small applications that ran on the desktop or browser.

I kept a list of the different widget/gadget standards, and quickly got lost in the volume. There’s just far too many widget standards out there at the moment.

Until those standards converge, or one comes out on top, there’s going to be the divide where widgets are limited to where they can run.

With mobile widgets, I’m not entirely convinced by the Yahoo Go idea – its just a Java application that runs on a small number of handsets. Looking briefly at Ricardo’s slides, there’s yet another different widget standard.

What concerns me most about Yahoo Go is that the handsets it runs on are probably powerful enough to run Opera, so why settle for Go, when Opera runs natively on the phone, allows browsing on the web, and already has its own widget standard?

I haven’t really seen any good widgets either. There’s nothing novel about a widget that requests data from a server and displays it – its just a “hello world” program with an HTTP GET request.

I look at my mobile phone – one that isn’t supported by Yahoo Go – and wonder why not? There’s documentation out there on how to develop midlets that run on the phone. This limited range of handsets makes the Go platform vulnerable to more innovative platforms like the iPhone.

I’m not sure what makes a mobile phone application that’s useful. Sure, being a plain old data aggregator has some benefits. Being able to peek into an inbox is useful. But what is compelling? What empowers the device to be something other than a glorified colour screen on a phone?

I picked up on your thoughts about:
“a knowledge-sharing platform that speeds up the mobile development process, probably a portal aggregating all resources as a directory in one place where developers share insights in the most efficient way they can, so that we can progress to a standardised, incremental approach more quickly than we are today”

This is really what we are hoping that Betavine could achieve, and we are looking for the help of the developer community to define it further and help us make it truly valuable. I’d love it if you could take a look (www.betavine.net) and drop me a note.

cheers, Margaret

Will definitely check out Steve’s presentation. Thanks!

=wil

“… compounded further by the challenge of interoperability – how can you create a functional tool that is workable on any mobile device?”

Without a doubt, the lack of interoperable standards is one of the major contributing factors of the high costs (human resource as well as financial costs) in mobile development.

Speaking of widget platform design (using the term “widget” loosely here — desktop widgets, facebook apps, mobile widgets): Facebook, Yahoo and MojiPage are essentially overcoming the device compatibility issue by having widget code run on the server. Yahoo, of course, has a client that runs on a wide variety of mobile devices, and they have the resources to support it.

I still believe that the mobile web will get better with each new device, and eventually the web will dominate on mobile just like it did on the desktop.