last updated: 10 January 2019 (approximate reading time: 8 minutes; 1618 words)
I make my living by writing down words.
From a practical perspective, every day I fire up my word processor and type away. But from a technological perspective, my livelihood relies on digital files which are processed on high tech devices that are a combination of mechanics and electronics and which then rely on complex software to make the machines actually run. In other words, I rely on bunch of electronic that can go wrong.
With every computer, the question is not if, but when a bad thing will happen. There is no doubt that a bad thing will happen, the only uncertainties are:
- when the bad thing will happen, and
- which bad thing will happen first.
Knowing that a bad thing will happen at some point, these are some of the processes that I have put in place to look after my electronic files.
In looking to preserve my files—the data on which my livelihood relies—I have several principles:
- The first consideration is that when the bad thing occurs, I need to be able to keep working without a hitch and without any delay. If my machine explodes, I want to be able to access my work from another machine immediately—not after a day, not after an hour, but immediately. There may be inconvenience, but I want to keep working without pause.
- I don’t have one single backup or one single process. I have several backups (so that if one backup fails I have other backups I can call).
- I have several processes to account for different potential problems and different eventualities. For me, this is all about layers of protection so there is (as far as practical) no possible single point of failure.
- And most significantly, this protection needs to happen automatically without my intervention because if I need take action, I am liable to forget/be busy/lack the energy.
What Are the Risks?
There are several ways that bad things can occur, but it’s worth taking a moment or two to look at the problems that may occur and the risks I’m trying to mitigate.
Potentially the biggest problem for me is human error—in other words, me doing something stupid. This may be me deleting a file or any other of a range of issues. In terms of where the risk lies, this is the most obvious and potentially the most consequential.
If the power fails, then any unsaved data is lost. This is probably a single file (or a highly limited number of files) issue, but it may still be significant.
If a computer fails, rather than a disk, the data may still be accessible. It’s mightily inconvenient and time consuming to fix this problem—it means removing the drive and connecting it another computer—but it is achievable.
Files can become inaccessible for any number of reasons. At the most basic, an individual file or small number of files may be corrupted. At the more complex end, an entire drive may fail making every file on that drive inaccessible.
In these circumstances some data may be recoverable, but data recovery from bad drives costs money, takes time, and any recovery cannot be guaranteed, so I look for a better option.
An entire drive can also become inaccessible for other reasons.
In the case of fire or theft, the files will become inaccessible, almost certainly on a permanent basis.
Ransomware and malware are, in essence, computer viruses which mess with individual files, groups of files, or entire disks of files making the files inaccessible. The divide between the two is that—theoretically—with ransomware, the files are locked and on payment of a ransom, the files will be unlocked.
Yeah…I don’t believe that either…
Responses to the Risks
Having identified the risks, I then apply some processes to address these potential challenges.
In applying these processes, my aim is not to have one solution that solves every problem. But rather, I’m looking to have layers of protection so that my files are stored in several places allowing me to have the best of both worlds when contemplating ease of recovery and certainty of recovery.
Before we talk about defense, there are some basic hygiene procedures that I follow:
- I keep all of my software up to date—mostly this happens automatically and behind the scenes.
- I run an anti-virus and firewall program—both of which are kept up to date.
- I’m cautious about what software I install.
These steps reduce the risks from more common problems. They do not prevent or fix any issues, but they reduce the risk of problems and hence reduce the likelihood that I’ll have to rely on my backup strategy.
First Line of Defense
My first line of defense is the obvious one: I have autosave switched on. When I am writing, my working document is set to save once every minute. This addresses the basic risk of power failure.
Saving a file also means that when it comes to backups (which I’ll come to in a moment), the files which are backed up are the most recent version of my files. If, for example, I only saved once an hour, then potentially my backup file might be an hour out of date.
Second Line of Defense
Putting files online makes them readily accessible. So, for instance, in the event of a computer or a drive failure, the files will be immediately available. In other words, this is the part of the strategy that means I should be able to keep working without a pause if I have a problem.
As I said, this isn’t a backup—all that has happened is the files are being synchronized to the cloud. Since files are synchronized, they can be easily deleted or written over (especially from a mobile device). Remember what I said about human error…
This is the stage in my workflow where the files I create when I’m mobile (so files I create/edit on my phone or a laptop) are synchronized back to my desktop machine. Once files created from mobile are on my desktop machine they are automatically included in my backup process.
While not designed to provide a backup, both OneDrive and Google Drive do offer some level of history, storing files for a period. This means that recently deleted files can be recovered through these services reasonably easily.
Third Line of Defense
My third line of defense is the first genuine backup.
I have File History engaged.
File History is a feature of Windows which takes a backup of my files every hour (and then stores these backups on a separate drive). This gives me:
- A backup of my files—so if my main drive dies, I’ve got a very recent version of the files immediately available on a separate drive.
- I’ve got chronological versions of my files (with snapshots taken every hour), so if I want/need to get back to an earlier version of the file (that human error thing again…), it’s a mouse click away.
As an aside, there is a Mac equivalent of File History—it’s called Time Machine.
Fourth Line of Defense
My fourth line of defense is to manually copy the contents of my hard drive to a separate external hard drive. This protects against nearly every risk (fire, theft, drive failure, and computer death).
This is the only manual element of the whole process. The manual element of this step is, of course, the weakness.
Having external drives also means that I can store the files in a separate location adding another layer of security (it’s unlikely that the location of my external drive would suffer fire/theft at the same time as my main computer).
Fifth Line of Defense
The fifth line of defense is an online backup service. Like File History, this makes regular timed backups of all my files, but instead of saving them on a local drive, the files are stored in the cloud. I use Crashplan, but there are many other similar services such as Backblaze.
The benefits of this approach are:
- Most obviously, my files are kept safe. This protects against human error, drive failure, fire, and theft.
- Secondly, I get the incremental history meaning I can find an earlier version of a file if I want.
- This incremental history also helps if a file becomes corrupted. A corrupted file will be backed up as a corrupted file, but I can still access an uncorrupted earlier version of the file.
- In the case of ransomware locking my entire drive, the files are still accessible.
The downside to an online backup service is that it takes a considerable time to recover an entire drive’s worth of a data. The recovery process is fast when only one or two files are needed, but with very large amounts of data (especially, in my case, photos), the recovery period is not swift.
Because of the time needed to recover a backup, it is my least preferred option. I have many other ways to recover data which are must faster, however, this backup gives me a high level of protection against many threats.
Will This Work For You?
So, should you adopt a process like this?
Probably not, unless you’re someone who makes their living by editing computer files. But if you have files that are important to you—for instance, if you have photos that matter to you—then I suggest you do something more than just hope your hard drive doesn’t develop a fault.