September 18, 2009, 7:19 am
Last night, I had the distinct pleasure of speaking at the monthly Fort Wayne, IN LUG meeting. Apparently, they thought I knew something about Linux. It did not take me long to prove them wrong.
Just kidding. In all honesty, it did go well, and was more of a long discussion than a full-blown speech, since I am big on interaction. The topic of the discussion is "Linux Everywhere You Want It," with the idea that with Linux being pretty much everywhere in platform-space today, the idea of the traditional desktop paradigm is evolving from the sit-on-the-PC-with-files-and-folders model to anything-that-can-provide-a-good-window model.
The big drivers for this shift, in my opinion, are two-fold: first, the flexibility and stability of Linux allows it to be ported to all of these different platforms, which forms a great basis for the growth of this new model. The second, which is no big surprise, is the cloud. After all, the platform-as-a-window model only works if there's something good to look at. If my bay window looks out over an alley rather than a river, you can see how it wouldn't have much appeal.
This is, for me, pretty standard stuff, and for these LUG members as well. We talked about the benefits of the cloud: for me, the former configuration manager, getting more apps from the cloud means less support and maintenance issues on the client machine; for one LUG member, the ease of seamless upgrades to reduce costs.
Of course, there are cons to the cloud right now, things that will need to be worked out, the most basic need being maintaining availability. When Gmail went down last week, it caught many people (myself included) unprepared on what to do when we couldn't access our messages or file attachments. But another con was raised that I had not considered before, and I think it deserves a little more examination.
A FWLUG member raised this objection to the cloud: as a developer, it concerns him that the cloud will eventually leave the traditional desktop behind in terms of innovation. Developers, he maintained, have a much easier time mashing up local applications and code than apps out on a cloud somewhere. Will innovation disappear if more apps are out on the cloud?
I think there a couple of reasons why this won't happen. For one, I don't believe the desktop model we have now is going to vanish completely. There will always be a market for those users who want local apps (be it for control, development, security, what have you). The cloud is poised to capture those users who really don't know or care what an operating system is.
The irony is, this ignorance of operating system is one of the things that Microsoft counts on when the average consumer goes to a big-box retailer and wants a new computer. They buy the computer that will run apps they have already or the apps they think they need. They aren't thinking "I need to stick with Microsoft as a loyal customer," they just want what works.
But if the cloud provides a platform of apps that will be compatible with their old apps and files (and that time is coming soon), then the argument "don't buy that, it won't run your apps" will fail, and suddenly price and value becomes a much bigger consideration--something in which Linux-loaded platforms can easily succeed.
For those users who do know more about what's going on with an operating system and their computers, there will be a desire for the tools and flexibility of a local OS, and Linux can fill that niche handily.
That was the answer I gave last night, but after a night's sleep, it occurred to me the audience member who originally raised the point could have his concerns realized: if the cloud gets really popular and we find ourselves in a world where client machines turn into dumber and dumber boxes, then innovation and development might get pinched, because there would be less of a place to "play" for current desktop and application developers.
The good news is, even if the market for the traditional desktop grows so small that platforms won't continue to provide it, I believe open source will ultimately save the day. Open source gives cloud app providers the opportunity to give developers a chance to innovate out on the cloud. In fact, they will want more developers to participate, since it will raise the chances for innovation.
This will be the real tipping point for the cloud-as-platform: when apps are not only hosted, but created en masse on the cloud. It's coming: Alfresco just announced yesterday a dev program on Amazon EC2 services, and SpringSource acquired Cloud Foundry back in August to enhance their cloud development options for customers.
The cloud shouldn't kill development and innovation, but innovation may have to move to a new address as applications increasingly move from clients to the cloud.