Mohtalim

http://business.timesonline.co.uk/tol/b ... 917414.ece
http://www.live-documents.com/products/basics.html

What these web-based app companies fail to realize is that large corporations are not going to put their data on the web. They need to sell the server-as-appliance or maybe just the server software. Then they could be really popular.

They need two options. The appliance like you mentioned for large companies, but they also need to provide hosting for smaller companies that don't have servers and sys admins. This would require them to setup some kind of unique hosting experience for each company that promotes the idea that your documents are siloed off from everyone else.

What's sad is that I can already think of what the second option's architecture needs to be. A layer of functional load balanced web application servers in front of a layer of encrypted databases.

I think there's a more fundamental issue in being able to control how these documents are going to look when they're printed. I haven't used office live, but the google one is really only useful for group draft editing and then needs to be cleaned up in a real document editor for printing and/or converting to pdf. Even the office live page mentions fidelity difference when talking about the different ways to share an office doc which to me is a bad sign.

As far as appliance or hosted, both will work, but I doubt that small/medium businesses will really want to go the hosted route. Of course, I didn't think they'd want to do that for email either, but they have. I'd rather see complete solutions for everything on a single appliance for small businesses and the ability to add load balanced peers for scaling up to medium size businesses.

If the open source community (or Red Hat for that matter) started packaging releases that did this, I'm sure it would drive others to follow. Hey Red Hat! Get to work on this.