Our digital lives are as much an entity as our physical lives. We each manage an online 'person', who has interactions with other online 'people' and services, much as we would in the physical world. Our digital lives have become much more than just an interface to our physical selves. We do the majority of our socialising online, often with people we only know online. Most of our possessions are online, in the form of photos and documents and conversation histories. We watch movies online and listen to music online. Many of us conduct our entire workday online: students, office workers, researchers, managers. Our physical lives are simply fit in around the margins. And yet, we don't treat these digital selves of ours very well. In fact, we don't really treat them as anything at all. This is a problem.
It's not just me who believes that:
- There is a growing concern about the dangers of online socialisation. From cyberbullying to catfishing to problematic online communities and more, online socialisation is seen as a new frontier in human interaction with many new dangers, or old ones in new forms. Governments and researchers are scrambling to respond.
- The digital space is almost entirely owned and operated by corporations. Unlike the physical world, where you might buy a house to live in or take advantage of public amenities like beaches and parks, the digital world has very few things you can claim for yourself or use without cost. This means that when you place your digital self in the digital world, by default you do so only by the grace of these companies. As such, we spend time dressing our digital selves in account data that can be deleted, or lost, or sold; we purchase tools for our digital selves with online subscriptions that can change in price or be cancelled; and we agree to the wholesale loss of digital privacy in order to better deliver us, not information, but algorithmically-curated content designed to drive our purchasing behaviour. Nothing about your digital self is owned by you anymore, and the information your digital self consumes is mediated by people who don't have your interests at heart. Again, governments and researchers are scrambling to respond to these issues, but also concerned are many tech insiders which seems, frankly, like a pretty troubling indictment.
- Your digital world is home to most of your stuff. Unless you're a particularly off-the-grid person, your photos, documents, and other files far outnumber the stuff you have in your residence. This stuff can be lost with a simple power outage, or a dropped hard drive. This stuff can be lost if there's an error in the cloud that overwrites your files. This stuff can be lost when a proprietary format is no longer supported and the world moves on. Governments and non-profits are scrambling to answer the question of maintaining data for the long term, but not many people care about your lost photos—that's on you.
- There are a vast number of digital threats that we typically ignore or poorly understood. Viruses, malware, ransomware, spyware, phishing attacks, man-in-the-middle attacks, the list goes on. These things don't just brick your computer, but steal your sensitive documents, financial information, and private photos. We know this, and so we generally install an antivirus and hope for the best. But this seems like an odd response to protecting our online self: the place we put most of our stuff, and the place we live much of our lives. It's not just that installing an antivirus and crossing our fingers isn't enough, because it's better than nothing. Rather, it's that we don't treat our physical selves with this much apathy. We know what kinds of threats face us in the real world—neighbourhoods to avoid, what to do after dark, behaviour that's risky or safe—and we conduct ourselves accordingly. There seems to be no reason not to do this with our digital selves.
So, we're going to give our digital person the attention it deserves, and build it a digital house. A digital architecture if you will. We'll take back control of our digital space from the companies that own us, learn to manage the threats that face us, and most importantly learn to maintain our data in the face of trouble.
A conceptual starting point
As I was starting this project for myself, there was one concept I had in mind that I found helpful, and another concept I learned later that I wish I'd known.
- We're not going to become experts, and often are going to barely understand what's going on, especially at first. That means we're likely to fuck things up. If something goes wrong with anything we're doing, we want to be able to sort out the problem super quickly. The logical solution to that is to make sure that at any given time we have no single point of failure. This is actually true all the time. Whether our computer is stolen, or if it's frozen because we messed it up, we should be able to replace that functionality immediately. If our files get deleted because of some error in the cloud, or because we did something stupid, we need to be able to get those files back from somewhere else. So whenever we're thinking about doing something, we're going to first figure out how to make sure that a screw up isn't going to bother us.
- This ties in very neatly to Unix philosophy, which I discovered later on in this project but was super helpful conceptually. Unix is the operating system that underlies most things that aren't Windows (e.g. Mac, Linux). One of their core tenets is that each program should do one thing, and do it very well. The other tenet (that we care about) is that each program should work with others easily. For our purposes, this means that our digital life should be modular—composed of pieces that can be swapped in or out as our needs change. This is not going to seem important just now, but it will become important as we go. Especially when we're talking about being able to fix our fuckups. We aren't going to have any one big thing that does everything, so any point of failure will be small and easily replaced. This is a luxury we don't have with our physical selves, and it will make this process much, much easier. But also it means that we are going to approach things from the point of our problems rather than working backwards from whatever software we're already using and finding a replacement for that.
We could consider a couple of other conceptual starting points, but these I think should be more obvious:
- We want things cheap; and
- We want things as simple as possible.
Finally, this is no command by command walk through. I've had enough trouble following super specific guides and modifying them to suit my needs. Or coming across something that doesn't work and spending ages figuring out what thing is out of date. So instead, this will set up a problem, describe the solution, describe what needs to be done to achieve the solution, and describe the endstate. Then you can go and search for things like 'install this program', or 'configure this thing' yourself. Or post on a forum 'I'm trying to do this specific thing, but don't know how', and they will answer the question with current advice. Then you won't have a situation where you copied some thing I did and it broke and you don't know why.
Step one: file management
So the first thing we're going to do is handle our files better. For the most part, we really only want two things from our files: easy access everywhere, and easy backups. But I also want them to be in one location. All my photos, all my music, all my documents, all my movies, all in one place, rather than scattered across various clouds and streaming services. As it stands, they're easy enough to access, but because my stuff is everywhere, nothing is being backed up anywhere and I have more than once forgotten where exactly a file was across all these things.
First stop: the cloud
The fastest, most accessible, and easiest solution is to use cloud services like Google Drive, or Onedrive. Between that and a harddrive or two, we can meet most of our most urgent needs. Here's the play.
But, the cloud is not enough
That's an absolutely adequate solution. But it doesn't solve things perfectly. For instance, shortly after setting this up, I moved Universities (and therefore cloud accounts) and I had to move all my files. This is the exact kind of hassle that we never want to deal with—our use of some software or system being dependent on the kindness of others. I was now concerned about a) whether the new University would let me store personal files here, and b) what I would do when I no longer had access. I needed something more permanent that didn't cost a heap of money. I did briefly buy a OneDrive for business account: that's some of the cheapest storage you can get at the time of writing without too much hassle if you have less than a terabyte of data. But we still have this looming cloud over our heads:
Putting your files in most cloud services means they aren't owned by you anymore, and this causes a variety of issues. You become bound by complicated and changeable terms of service, and are at risk of losing your files to business dealings like the sale of a company or insolvency. If the company has an outage, so do you. If the company loses the files, so do you. At some point you might run up against financial obstacles, like me and my storage problem. If you run out of room, you need to pay for more. If the company reduces your allowance, you have to pay for more. And perhaps most offensively, the company is likely in the habit of periodically reading your files. If they're responsible, they're likely only checking for reasonable things like child porn. But in the case of companies like Google, they're also using it to add to their data profile of you. Either way, I'm not generally in the habit of giving out all of my files to people, and I'd prefer not to do exactly that thing by default.
So, what we really want is something like our cloud storage solution that doesn't read our files, isn't possessed by someone else, doesn't run out of room, and provides legitimate backups. And we want all of this for a reasonable price.
Next stop: upgrading from the cloud, a simple solution in an unreasonably complicated crowd.
The obvious choice is to set up another computer in the other room, and somehow get access to that from wherever we happen to be. This is the most basic definition of a server. A computer that serves information, in this case files.
This might seem like a big leap, but it isn't.
We already have a harddrive or two, because we're now backing up our files from the cloud (right?). If we don't, we should, because the equivalent cost of about ten glasses of wine to buy a couple of harddrives seems perfectly reasonable to keep the majority of our possessions safe. Better price than any insurance policy.
Many people also have an old laptop lying around that's too slow to use, which is fine because a basic server doesn't do anything that needs speed. Or the battery is gone, which is also fine because we can just plug it in somewhere out of the way and leave it. If you don't, second hand computers are startlingly cheap because computers depreciate in value so quickly. Or you can spring for a hobbyist computer like a Raspberry Pi that costs less than a harddrive. A server, even on an old computer, can be transformed not only into your own personal Google Drive, but also a personal Spotify or Netflix. It can automate your backups. It can turn your bloody lights on. It's a computer. Again, we must consider the value of what we're trying to do here. I want to OWN my digital life, not hand it off to someone else. If I'm serious about that, then this is a logical next step. But we need to make it easy.
And this should be straightforward. Booting up a new laptop is straightforward, so why wouldn't booting up one that does a different job be? But a foray into the territory of home servers instantly becomes ridiculous. Overwhelming talk of network-attached storage (NAS) and RAID levels among other incomprehensible things. Least helpfully, every option is far beyond anything that we could reasonably call 'low cost'.
We don't need all of that. We need something simple. If you want something complicated, you can go and lose yourself in /r/homelab.
Fortunately, after an annoying amount of time spent googling words I don't care about, I came across a project that meant I didn't have to worry about any of these weirder setups.
With that project, we'll have created simple set up that allows us to access our files and stream media from our PC, our phone, and wherever else we might want them. No one owns them but us. If anything goes wrong, it can be replaced easily. And most importantly, we'll have control over our digital self in a way that it deserves.
But, that only takes us so far. We need to go a little bit further and create a small and persistent network, that will make it easier again to access our files, but more importantly, it will allow us to build out our digital architecture in any direction that pleases us.
And with that, we've laid the foundations for our digital architecture. Whether you decide to build on top or not, you now have a digital architecture that means your digital possessions are safe from the most common sources of loss and destruction, they'll follow you wherever you go, and they'll be completely amenable to our next stop without us needing to worry if we accidentally delete a bunch of stuff by accident—security and encryption.
Next stop: hiding files from prying eyes
I've gotta write that first though.
What will follow
From there, we'll continue even further, because our digital lives aren't just about files. We leave trails of our digital lives all over the internet. Accounts on websites we visited once. Newsletters we stopped caring about years ago still filter into our inbox. Account data on social media accounts that would disappear if it was deleted. Taking control of our digital lives isn't just about taking control of our files. It's about managing our digital world. That's what we're going to achieve.
I just haven't written that part yet.