Your average active computer user has more and more data. The first computer I effectively administered had 170 megabytes of hard disk space. Difficult choices had to be made about the relative merits of Doom versus Simcity. Now, just my primary email account has 1500 megabytes of data in it. I have 15 gigabytes worth of photos I have taken (all since 2005) and 20 gigabytes of music.
All this has been made possible by dramatically falling storage prices, combined with the spread of broadband internet. Soon, I expect that this combination will reach its logical conclusion. Right now, people are constrained by the size of their smallest hard drive, as well as by the difficulty of accessing larger remote drives. Eventually, I expect that most people will have a multi-terabyte disk connected to the internet at high speed and securely accessible from virtually any device in the world over the internet. The biggest question is whether this will be an ‘answering machine’ or a ‘voicemail’ solution.
The answering machine option is a big disk purchased by an individual consumer (perhaps a rack of disks, so that cheaper bigger ones can be added to the array as they become available). A company that made three things easy would have a license to print money. The first is integrated ease of use. iTunes music on the big disk should be immediately accessible from a person’s laptop or iPhone, provided they have internet access. The same should be true for saved television shows, photos, etc. The second is effortless backup. It is perfectly feasible to have a disk that is big enough to ensure that the failure of any one component does not lead to any loss of data. The third is security. The big disk should be secure enough against outside attack for use in storing commercially sensitive materials; likewise, the connection between outside devices and the disks should be secure. Probably, this means different levels of access for different sorts of devices, managed through a good user interface.
The voicemail option is to leave all the kit to someone else and just buy a service. Lots of companies are moving towards this model. In many ways, it’s a lot more efficient. Maintaining adequate but not excessive space for a million users is easier than doing the same thing for one; there are also economies of scale, since you can have specialists do all the technical work. The downsides of this model are mostly security related. You need to trust the service provider to keep your data safe. You also need to trust them not to apply arbitrary constraints on how you can use it, as Apple has sometimes done.
I predict that most people will use the second model exclusively, and will pay little or nothing to do so. More technically savvy people will run their own drives, but will probably use external services for (free) unencrypted or (subscription based) encrypted backup. Personally, I can’t wait. External hard drives have the feel of a 1980s solution, rather than one that is aware of the potential of the internet.