After investigating some of the nitty-gritty details of data storage last week, now it's time to look at another place where people store their data: the cloud! When does it make sense to use cloud services? What are the advantages compared to the alternatives? For that matter, what are the alternatives to the cloud? What does it do to your efforts at organization if you choose one or the other?
Deciding whether to use or not use cloud services depends on technical, economical, and even philosophical factors. Below, I'll explain as many relevant advantages and disadvantages as I can while attempting to remain as unbiased as possible about the non-technical reasons. My hope is for you to understand the landscape and be able to make your own well-informed decision, regardless of what it is.
But first, we need to answer a key question...
What is the Cloud?
If you have a smartphone, you're almost certainly using at least some cloud services right now. The most effective cloud services are the ones you don't have to think about, and might not even realize exist. Apple's iCloud is a prime example of this. You have to turn it on intentionally during your first phone or tablet setup, but after that, it's basically invisible. (Well, at least until you run out of storage space and they ask for a subscription upgrade.)
A few years ago, someone quipped that "the cloud is just someone else's computer." In a literal sense, this is true. At a hardware level, everything about bits and bytes and hard drives and all that still applies no matter where you store your data. If it's not on your computer, it's on someone else's, even if the actual device might not look like what you typically think of as a computer.
However, this snarky trivialization isn't quite fair, even if technically accurate.
Anything colloquially referred to as “the cloud” comes along with some additional value that makes it an enticing alternative. Otherwise, we would always choose to keep our data on our own devices or storage media.
Below, I've listed a few examples of popular cloud services and some of the features that make them appealing:
- Google Photos provides automatic photo and video synchronization, album arrangement, photo tagging, facial and object recognition, comprehensive search abilities, and customizable sharing.
- Gmail provides label-based organization, automatic filtering, mail categorization, spam detection, mail search, multiple account imports, and a universal interface across many platforms.
- DropBox, OneDrive, iDrive, pCloud, and others seamlessly synchronize and back up your important data on a robust, fast remote storage platform.
- iCloud integrates tightly across the entire Apple ecosystem (Mac computers and laptops, iPads, iPhones, and watches) and provides a one-stop shop for relatively secure backup and synchronization of many different kinds of data.
A better definition of "the cloud" is therefore not just someone else's computer, but a computer or server that you access over the internet and the services and databases that run on those systems. By contrast, we'll refer to anything that you have to administer yourself on a computer or server you manage as "local." (I like "the ground" as an opposite term, but I've never seen anyone else use it, and I don't want to confuse anyone.)
If you'd like a more visual explanation, check out the "What Is the Cloud?" video from the Goodwill Community Foundation on YouTube.
Why Use the Cloud?
Is it possible to replicate the features that these services provide without handing over your data to anyone else? In most cases, yes. But is that the best choice? Let's examine the main benefits of cloud services compared to self-managed local solutions.
- Scalability. Cloud services can easily provide as much or as little as you need in terms of storage space, computing power, data transfer, and so on. All you have to do is ask (and pay, of course). In contrast, increasing the capabilities of a local system often requires expensive hardware purchases, and if you decide later that you don't need all that power or capacity after all, recouping those costs takes still more time and effort.
- Accessibility. You can access cloud services from anywhere with an internet connection, which for most of us these days means everywhere. Responding to an email or organizing and sharing a few photos is just a few taps away; you don't have to wait until you get home. Of course, it's possible to configure a self-hosted system for remote access over the internet, but this takes knowledge and time that many people don't have, and it also raises security concerns.
- Cost savings. Factoring in the amount of time and money required to locally replicate almost any common cloud service with the same degree of functionality and performance, it's often difficult or impossible to win financially with a local solution. Inexpensive or free self-hosted systems often fall short in areas that make them objectively or subjectively worse than their cloud-based counterparts. If you can pay $10 per month for a service that works perfectly all the time with zero effort, or you can pay nothing but spend six hours setting it up and another hour or two every month maintaining it, which would you choose? How much is your time worth?
- Low maintenance. Cloud providers generally handle all relevant software updates and maintenance for you behind the scenes, so you don't need to do anything to stay up-to-date with the latest features, bug fixes, and security patches. Doing this yourself requires knowledge and time, both of which might be in short supply. And, if you happen to fall behind, you might inadvertently expose yourself to unnecessary security risks.
- Backup and recovery. Most successful cloud services are built on an extremely large scale with redundancy in all key components. This means that there are hundreds, thousands, or sometimes even more functionally identical parts which differ only slightly in configuration based on which subscribers may access them, and everything is duplicated (sometimes many times over). This way, if anything fails for any reason, system administrators can swap in a replacement without the customer even realizing anything went wrong behind the scenes.
- Security. Cloud service providers often have a team of experts specifically focused on security and compliance to prevent loss or theft. Achieving the same level of security on a self-hosted local service can be challenging, especially if you open up that system to the internet for your own personal remote access. Granted, your single server makes a much less appealing target for nefarious hackers than a giant cloud provider, but it is still a risk. On the other hand, if you hand over the security of your data to a third party, you should trust the company managing it.
- Innovation. The market for everything constantly evolves and new competitors arise almost daily. In order to avoid losing customers, existing cloud service providers must integrate new features and improve their services to match or surpass whatever the latest and greatest is. This doesn't always occur, of course, but the economic motivation is there. If you set up your own local service, you have to do this yourself, or just live with an unchanging set of features.
As you consider the list above, think of it in terms of the physical things that you buy for yourself: food, clothes, computers, phones, a car, anything. Could you make these things yourself? In some cases, maybe so, but most of the time, you couldn't even begin to achieve anything like the results you can enjoy for just a dollar or two from the local grocery store, hardware store, or technology store. This is an example of the benefits of the division of labor, which is a very good thing.
Cloud service providers do the same thing in the digital space. Even if you could replicate what they do, it often makes more sense economically to pay them to do it for you. Whether this is the better choice in any specific case depends on what's more important to you.
Why Not Use the Cloud?
Despite all of the advantages we just discussed, some people still prefer local systems instead of cloud-based solutions. (Full disclosure: I am one of these people, though I do still use some cloud services.) Let's examine some of the main reasons why local solutions are more appealing.
- Control. This is one of the main motivators for people who choose this route. By hosting services locally, you have complete control over your data and everything that happens to it. You aren't stuck with whatever features or capabilities the cloud service provider decides to implement. Everything is completely customizable, albeit with more effort.
- Privacy. In addition to control, this is often the other main motivator for going local. Since you never need to hand over your data (encrypted or otherwise) to anyone else, you can be reasonably certain that it remains only in your hands. You don't have to parse a giant organization's arcane privacy policy and hope they actually adhere to it.
- Security. Locally hosted services can provide a higher level of security than cloud services, as long as you know enough about the hardware and software you are using to configure it reasonably well. If you are a security-conscious technical person, you might also respond to new updates and patches faster than a large cloud provider would, since you are free to react immediately without worrying about how installing an update might affect all 10 million of your subscribers. However, this is still a tricky topic, and it's easy to get wrong.
- Performance. Locally hosted services can offer much better performance than their cloud counterparts since data doesn't have to travel over the internet. If you work with large data sets such as a huge photo or video collection (or three decades of personal data, as in my case), fast access is especially important.
- Cost savings. "But Jeff," I hear you cry, "Wasn't this already an advantage of cloud services?" Yes! But in some cases, maintaining and upgrading locally hosted systems might be cheaper than paying for cloud services. This requires a calculation of expected costs over time; cloud services add up if you subscribe for long enough, while locally hosted systems have mostly up-front costs with few additional purchase requirements (other than upgrades, which are usually optional). However, don't forget to factor in your own fairly valued time. If you can find a turn-key solution that won't become a time-sink, and you have the knowledge to deploy it, it might still be worth considering for cost reasons.
- Internet independence. Locally hosted systems continue working no matter what the outside world looks like. On the other hand, cloud services depend on the internet. This can obviously be problematic if you have poor connectivity or suffer from power outages. Some services provide an "offline mode" that allows you to continue working for days or even weeks, but for most systems, you're out of luck until your internet comes back up.
Especially those first two points together provide enough impetus for many of us to shy away from cloud services when possible. Sometimes there's no way around using a cloud service when certain features are absolutely necessary; anything heavily focused on real-time collaboration or media sharing often can't work well (or at all) in a locally hosted setting. But for the vast majority of applications, you're free to make a choice.
Jumping back to my "physical goods" analogy, you can make the same kinds of comparisons. Some people do in fact choose to grow their own food, make their own clothes, craft their own furniture, and in extreme cases even build some of their own electronics. This is much more complicated than choosing to do something yourself in the digital world of computers and software, but the basic analogy holds.
What About Philosophical Issues?
The use of cloud services raises other concerns that are sometimes harder to articulate. Let's take a look at just a few of the common ones.
- Centralization. Cloud services tend to centralize data and computing power in a small number of large corporations, which can be a concern for those who value decentralization and diversification.
- Ethical concerns. Some people are concerned about the impacts that large cloud providers have on government policy (and vice-versa). Even if everything about a particular service seems valuable, mistrust in company management can make someone wary enough to avoid handing personal data over.
- Dependence on a third party. Relying on cloud services can create some level of dependence on the service provider just to maintain the normal function of your business or personal life. Often by design, it can be quite challenging to untangle yourself from a comprehensive cloud-based ecosystem if you decide to later on.
There are endless resources, articles, and vehement opinions about these topics, but they are too far outside of the main focus here at Tidy Bytes to dwell on more here.
Do One Thing
Think about any cloud services you use and consider whether it might make sense to try something locally instead.
It might not! That's a decision that only you can make. But it's always best to make those decisions with a clear understanding of why you're choosing one path or the other. Everyone has different priorities, and that's part of what makes the world such an interesting place.