Loosely Couple wrote a while back a couple posts, Avalon: Microsoft's Microchannel and Rich clients, network wealth on why he believes that "applications will run in rich browser clients, and Windows will have settled into its legacy platform niche ... and Microsoft will be working hard to re-establish the market position of."
I was a little skeptical of this claim, but he did present a lot of evidence. There are a number of initiatives and products by various parties including IBM, Macromedia, Dreamweaver, and BEA to make browser applications as rich as desktop apps.
Just recently, he just included additional examples in his Why Yahoo! bought Oddpost. Google is in the midst of releasing GMail, which provides a rich-client email experience within a web browser. GMail raises the bar by extensively using JavaScript to offer rich capabilities and a very responsive desktop-like feel. Yahoo, in a competitive reaction, purchased Oddpost, which produces a clone of Microsoft Outlook in JavaScript and DHTML running in IE.
So, I'll buy into his claim that web applications will become richer and encroach into the desktop turf, but I don't feel that Microsoft needs to worry much yet, since the rich desktop will likewise get richer. In addition, Longhorn applications will adopt many ideas of the web as I mentioned in my earlier post.
They say that they will "build on current browser technology", but that they need to extend it... So these great new rich-web-client things won't work with current browsers. In other words, they require you to download a new client. If you're going to download a new client, why not make it the .NET Framework, and then you can get REAL richness?
What I really see here is that they want to keep competing with Microsoft. However, knowing that whatever they come up with will never match .NET or Longhorn, they instead decide to call it "browser technology" and say that's why it's better than Windows, because it's "browser-based".
Posted by: Michael Giagnocavo | July 12, 2004 at 10:26 AM
Yes, but...Longhorn browser apps will run only in Longhorn, so you not only miss the current xp/2000/98 base installations, but also miss all the OsX, *nixes, palm based browsers, etc.
if you will be developing for Longhorn, why not make a Longhorn native app?
Macromedia have a huge win here, because it can run in all platform with flash plug-in installed.
Posted by: Eduardo | July 12, 2004 at 01:01 PM
Personal opinion, what web browsers badly need is a local persistent format based on some additional namespace schema, only to avoid server-side round-trips after each and every click. This persistent stream would be then synched up with the server-side any time required. I don't know how this actually matches WebDav but this sounds pretty close.
Posted by: Stephane Rodriguez | July 13, 2004 at 01:04 AM
For a captured audience WinForms is alreayd losing out to Web Applications. Developing internally for an enterprise nearly all applications are web based clients.
We are adding lots of rich client interface functionality by developing our own asp.net controls, though you do have to jump through a few hoops to get the browser to play ball.
As we have a guarenteed web client (IE6) we make good use of javascript and web service behaviours to update the interface, data etc without round trips to the server. Ok its still a browser with some of the limitations of a browser but most of our apps are feeling much richer than that when you use them.
That being said there are lots of advantages to WinForms and as deployment becoming less of an issue now they start to look more attractive again.
My opinion is there is a place for both clienst and will be for the forseeable future...
Posted by: Rob Williams | July 13, 2004 at 06:52 AM