Where did the online backup service model go wrong? After showing great promise a few years ago, the whole idea of backing up critical files – or all files – to a secure location somewhere else in the world has never quite caught on for mainstream users.

Some blame the abundance of cheap external hard disks, others an unwillingness to pay to store something not valued until it has gone up in smoke at which point it becomes suddenly and incredibly precious.

There is another explanation that fits more accommodatingly with the history of technologies that people have to use in a personal way – consumers and small business users just haven't 'got it' yet. When they do, they'll be rioting in the streets for it and the issue will have been settled in favour of offsite backup once and for all.

Spare Backup is a recent and brave attempt to buck the apathy, and it comes with an innovation we liked despite having some reservations at first. Where most backup utilities/services ask the user to decide what to back up and how often, Spare Backup chooses the files itself based on what it finds on a hard disk using something called “pre-sets”. These are categories of data, such as documents, email, music, digital images, etc, which it has detected by analysing the applications on a PC's hard disk.

The selection process runs once when the software in installed, the length of which depends on how many files it finds. Each data category can be either backed up online, backed up locally to a CD/DVD, backed up both online and locally, or simply ignored. Files can also be individually selected or excluded, including from networked drives. Powerful though this schema is, the interface can be clunky at times, and the addition of the odd 'back' button to spell it out for us dimwits would be helpful.

Once the data sets to be backed up are as desired, and the time of backup specified, files are compressed and encrypted (using 256-bit AES) before being sent off to two separate and secure servers. Accessing these files later on means taking a mental note of the backup key so that the files can be decrypted once they have been copied back to the hard disk. The advantage of all this encryption is that it means nobody – even those working in the data centres themselves – can see the data in its unencrypted form. The downside is that it adds overhead, in both directions.

The initial backup of 1GB of assorted data on a standard 256 kbps second broadband uplink, took so long we gave up waiting. Surprisingly, no indication was given as to how long any of this might take, nor could you reconfigure the backup process once it had started. A subsequent backup of a test 10MB took over 15 minutes. Broadband or not, online backup times are still sclerotic compared to local backup.

That said, the program can be tasked to run during downtime on any day of the week, and set to turn off the PC when finished. When saving new versions of files, Spare is able to work out which files have been changed (including Outlook email), and doesn't therefore save each file repeatedly. The biggest and longest upload will therefore be the first one, with subsequent backups taking less time.

The least satisfying part of the program is the restore process, which can be done in a single lump on a date basis, or by searching for a specific file. But what happens if you want to restore a group of files, the names of which you can't remember? The only option is to identify them from the backup report built into the software and then hunt them down one by one. To its credit, the system saves up to ten revisions of every file, clearly numbered in the restore window. Getting back a much older version of a backup would be pretty easy using Spare. The system also supports Unicode filenames, so special characters will reproduce correctly.

Online backup will be second nature one day. For now, it's important to think of it a different way to conventional local backup. It's not as quick because it can't be for bandwidth and security reasons, and so it best suits scheduled backup regimes for modest volumes of data.

Spare Backup's ability to choose files for you sounds great but the sheer volume of files it finds means that is has to be used with care. The temptation is to cut the file list down manually, but that undermines the point of such a design to some extent. The restore, meanwhile, is a bit laborious. On the plus side, you can use the system to reach your files in a highly secure form from anywhere with an Internet connection. From a security point of view, Spare Backup is top-notch.

Revise the interface slightly, improve the restore, and perhaps wait for a time when broadband offers better upload times, and this service could be excellent. Online backup is a powerful concept, but it’s not quite there yet unless you happen to be a small business with a dedicated uplink with decent bandwidth.

The basic package offers an annual subscription for $60 (plus tax for US residents) for 50 gigabytes, equivalent to $4.99 per month; the month-by-month rate is $6.99 for the same volume. Only unique files count towards the volume calculation, with backup above the 50 gigabyte limit counting as commercial usage.

Spare Backup’s website can be found here .


Online backup is neither particularly cheap or quick. It works best for restricted data sets, or for small businesses with access to a dedicated uplink.