Over the past year or so, there has been way too much ink, electrons and air spent on cloud computing. The only constant through all of this hype is that no one really knows exactly what the hell it all means other than it has something to do with the web. I’ve heard stories about how The Cloud will render all PCs to little more than thin client shells running everything out of a futuristic browser or less insane stories about how just our data and bandwidth-light applications will live in the cloud. I personally don’t buy the hype and think that the future of teh cloudz will be far different that what the hype machine is making it out to be.
For people who have paused long enough to catch their breath, cloud computing is nothing more than server-side computing available over the open internet (rather than a LAN). The thing that makes this round of server dominance is the rise of wireless internet, particularly cellular data networks, which allows you to get the internet anywhere. The idea is that given the availability of data everywhere, the ever-increasing speed and robustness of wireless networks and the power of distributed computing, you’ll be able to do all of your computing through the browser.
The thing is, most of the people who are so into cloud computing already live in their browsers. When all you do is troll around on the web for news, live in Gmail and do some light text entry, then yes, doing everything in the browser seems quite doable. If you combine your Google Docs with Pandora and use an Android phone (or a Pre in a week or so), upload all of your pictures to Flickr, then guess what? You’re already living in the cloud! Congratulations! You know what else? You have ceded control over all of your data to someone else, usually a large corporation.
That’s my biggest problem with the cloud. I am fully aware of the irony of complaining about this on a blog hosted by someone else, but bear with me. Sites like Flickr and Facebook are notorious for deleting content and banning users, often by deleting entire accounts, which can have far-reaching consequences. Or, they might decide that they own your data and content in perpetuity. Or, the company that’s holding all of your super-encrypted backups or your RAW files just goes out of business, giving you a day to get your stuff back. You and everyone else who has files on that server, that is. 24 hours doesn’t seem quite like enough time for any given server to upload how many terabytes, especially over American intertubes.
And then there are the ISPs. American internet connections are slow and expensive and cellular data is even worse. The future does not promise to get any better, as the large cable companies are still pushing towards tiered connections with ridiculously low data caps, all at a higher price. How many videos do you plan on editing online when you have to count your bits after a YouTube binge?
There are too many roadblocks and bridge trolls in the way for cloud computing to really ever take the place of local computers. The question that I never hear asked is why? Why do I need online storage when 1TB HDDs cost less than $90? When my two year old iPhone has more storage, RAM and processor than the Mac I had in college? When my old-ass Quicksilver G4 (or a Mac Mini) makes a perfectly adequate server? When Amazon suggests that you buy your own drive, fill it up and mail it to them because networks are that much slower and unreliable than FedEx?
Not good enough? How about because anything you put on a remote server can be looked at with a court order because your data doesn’t deserve Fourth Amendment protection when its on someone else’s box? Your TCP/IP requests aren’t protected and neither is your email. I use Gmail like all the other cool kids, but I’m fully aware that the stuff I send and receive is there for everyone to see. That’s why I use Mail.app because Google doesn’t give you this:
Cloud computing is seriously over-hyped and under-thought. Yes, there are things that live on the web that should live there. Yes, the browser is becoming increasingly powerful. But the fact is that client-server computing is not going to be better than native apps on increasingly powerful and cheaper hardware running operating systems that are super-optimised to take advantage of that hardware. I think that this is all a fad, but one that will be quite persistent.
I think that people will be utilizing web-based apps in greater numbers as web standards become more powerful and are adopted by browsers. I also think that native apps will always have their place and will always be better, first-class citizens. There will always be serious compromises when using remote apps over native, not the least because web apps will always be somewhat lowest common denominator unable to take advantage of local hardware to any significant extent nor interact with the native OS in a first-class way.
Finally, there’s the security issue. If you want to be in control of your own shit, then you leave it on a box that you physically control. End of story. That’s not to say that persistent data access is a bad thing. Next time!