The New York Times has an interesting OpEd article, by Professor Jonathan Zittrain of Harvard Law School, about some of the implications of Internet-based or “cloud” computing. The recently announced Chrome OS from Google is just one indication that computing in the cloud is becoming a more significant part of computing’s evolution. As Zittrain says, there are some obvious benefits to moving toward cloud computing:
Many people consider this development to be as sensible and inevitable as the move from answering machines to voicemail. With your stuff in the cloud, it’s not a catastrophe to lose your laptop, any more than losing your glasses would permanently destroy your vision.
Nonetheless, there are some real dangers associated with computing in the cloud, some of which I’ve talked about before. One fairly obvious one is having all or part of one’s data at someone else’s mercy: There are several aspects of this:
- With some cloud services (music, for example), you may need the continued existence of the service in order to use the data you “bought”. A recent Washington Post “Help File” column gives an example, of a music service discontinued by Yahoo!, which has left some purchasers without any means of accessing the music they thought they had purchased.
- The degree to which your data is protected while stored in the cloud is not up to you, and is often not easy to determine. And the degree to which you have anylegal protection may also be murky; as Zittrain says, “Worse, data stored online has less privacy protection both in practice and under the law.”
- Governments and other snoopers have a much easier time getting access to data in the cloud than they do to data on your PC. The Patriot Act in the US gave the federal government the right to demand information about some of your online activities from the providers; they don;t have to tell you about it, either. And more authoritarian governments (think Iran, or China) do a lot of snooping.
- As I noted in an earlier post, there is a danger of getting “locked in” to a particular cloud computing provider, if the data is kept in a proprietary format that makes it difficult or expensive to change vendors.
Zittrain argues that there is further risk involved, which is more subtle, but no less pernicious. Because the vendor, in cloud computing, has much more control over the environment than a software vendor (even Microsoft!) has over your PC, there is a risk that that control might be used to stifle innovation:
The crucial legacy of the personal computer is that anyone can write code for it and give or sell that code to you — and the vendors of the PC and its operating system have no more to say about it than your phone company does about which answering machine you decide to buy. Microsoft might want you to run Word and Internet Explorer, but those had better be good products or you’ll switch with a few mouse clicks to OpenOffice or Firefox.
As he points out, it is not just competition between commercial application vendors that might be threatened by control of the cloud; it might also make it more difficult to introduce new or niche applications:
Instant messaging, peer-to-peer file sharing and the Web itself all exist thanks to people out in left field, often writing for fun rather than money, who are able to tempt the rest of us to try out what they’ve done.
In this respect, it is good to see developments like Amazon’s generic cloud computing service, and the recently-announced Google Chrome OS. Having open, standards-based competition is one of the best ways to keep all the proprietary vendors honest,