Thanks for your thoughts. I respect your opinion and can only draw on my experience. I spent six years making Windows virtualization software for Linux which I thought was the silver bullet for Linux Desktop adoption. My though was it allowed the migration of Windows applications to Linux without a complete rewrite of the appplications.

It seems to me that over the next 10 years the sensible solution will be to migrate to the web for new applications, look at Intuit and their release of QuickBooks and Quicken for the web. I think you are right about security concerns today but overtime I think this problem will be solved through market demand.

For now I guess it’s just a case of predictions and we’ll have to see how everything plays out.

Thanks for your thoughts.

The first issue I see here is that I don’t think that Google Office will have much traction in real business world. People are usually at least a bit more paranoid than your company and will probably not have their employees create and have company documents on a web host, no matter what assurance of security and privacy they have from google. Ditto with all web-based, hosted-by-outside-entity type of business software. Aside from hype there is little gain in web2.0 business software. Outside security issues, it’s all years behind even the slowest developing opensource desktop apps, and most of us would like to have our business-critical stuff chugging along even when we have internet outages, bandwidth spikes etc. Not everyone lives in an American or Western European metropolis, I’d dare say that more than 90% of the planet doesn’t.

The second thing is — it’s not office apps keeping people off Linux desktop. For one, OpenOffice, Abiword/Gnumeric and even KOffice are more than decent replacements for MS Office and if you have trouble moving your users to those than moving them to Google Office will be impossible.

What’s with my engineering department and it’s dependance not on “a CAD software” but AutoCAD-or-bust (make something 100% compatible and keep it compatible for versions to come, charge half the dough and then we’re talking). What about the marketing department that’s used to Illustrator/Photoshop combo — don’t say Inkscape/Gimp ’cause it just won’t cut it (ever tried preparing pro CMYK prepress with those two).

Or what about people who do my finances and accounting, my sales and purchases. They use an accounting software, they cannot afford SAP and Oracle and never will — last I checked 99,9% business IS packages, from simple invoice printshops through acounting apps to full-on ERPs are Windows only.

I can see these things getting on Linux desktops much, much sooner than they get to web2.0 world. Actually I expect the web2.0 bubble to burst way before that time comes, but that’s another topic.

What Linux world currently needs badly is more, better development tools that make it easy for semi-literate quasi coders to write “serious” database backed apps for businesess worldwide, preferably baking easy to make desktop/db-connectible stuff from some codegenerator like Rails or a connect-the-dots kind of CASE-like thing like Clarion. VB did as much for Windows as a platform as Office did. That’s how you get ISVs.

The other thing it needs to break trough in enterprise is less choice: tightly setup server distros that pick the best tools from the flock and deploy/configure those from a wizards will kill it. So far not even RedHat got it quite right, let alone Ubuntu/Debian or Novell. SME’s usually cannot afford to wander in the wilderness. In this regard CentOS did more for Linux in SME world than all other freebie distros together.