Do you even know what cloud-based computing is, and do you know how much of the things you do on your computer rely on cloud computing? More to the point, do you appreciate the potential downside to using cloud resources?
If you use a computer for anything (and you’d not be reading this if you didn’t) you need to understand the downsides and risks associated with cloud based computing. We’ve written this in reasonably non-technical terms, so please do continue reading.
First, a definition. Cloud based computing is anything that uses resources that are not within your local area network. These resources might be programs, and/or they might be information/data. In other words, if you can’t walk across the office or from one room to another in your house and physically touch the box in which the data is located, then we’re talking about cloud based computing.
Here’s one immediate example : Email. Most of the free email programs (actually, probably all) are primarily cloud based. The email interface itself is driven through a remote (ie cloud based) website, and the emails that you ‘receive’, ‘open’, read, and reply to (and the copies of replies you send) are all located somewhere else in the world rather than on your computer, too.
If you’re not connected to the internet, you can’t read, receive, or send emails if you use this type of service. Okay, so of course you can’t send an email if you’re not connected to the internet – but what about not being able to read your older emails, too? Did you even know that there are lots of ways that you can still read the stored emails you’ve previously received and sent, without needing the internet connection?
Cloud Computing is Growing
Software and service providers are keen to promote cloud based computing, because they can turn around and charge monthly fees for such things. Instead of a one time selling price, they can now charge monthly fees for the use of the software, and while it sometimes seems appealing to us to accept a low ongoing monthly cost along with ‘free upgrades’ instead of buying software outright and then buying new versions in the future too, clearly the software providers are betting that, overall, they will make more money from renting software access on a monthly basis – even after deducting all the huge costs of setting up and maintaining responsive internet access for their software.
The proliferation of extremely fast and inexpensive internet makes cloud computing practical for service providers and for us as users. It is now possible to work with programs and their data almost as quickly if they are cloud based as if they are locally based (and certainly as quickly as if they are LAN based).
Our concern about cloud based computing is not based on financial reasons. Those are what they are, and you’re free to make your own decisions as to buying or renting, any way you wish. It is also not based on the end-user experience. That is perfectly good, most of the time, for most things.
Our concern is based on what would happen if/when your internet access is disrupted, and you can no longer access all the information you have stored online. Indeed, it isn’t just a case of what happens if the ‘last mile’ of internet access to your home/office and computer is disrupted. You are also vulnerable to what happens if the company providing the internet cloud service has disruptions to their access too, or if they have any other type of computer problems, and you’re also vulnerable if the internet as a whole has some sort of disruption (unlikely but far from impossible).
Maybe you are prudently keeping backup copies on your own local computer, but what use are those backed up copies if the program to access the data you’ve stored is an online program?
Even a Level 1 event could see temporary loss of access to your cloud data and computing programs. A Level 2 event could see a much longer term loss of access, and a Level 3 event might see the permanent disappearance of your data.
We’re not just talking abstract theory here. These things have already happened, and not just the occasional email outage that many of us have sometimes experienced. Please read on.
How 100 million Users Unexpectedly Lost Their Data Earlier This Year
Here is an appalling example of the vulnerability of cloud data, and one that occurred in a completely non SHTF type situation.
A cloud storage/sharing service, Megaupload.com, grew to become one of the internet’s largest sites, with 25 petabytes of data stored (in case you don’t know what a petabye is, it is 1,000 terabytes, or 1,000,000 gigabytes, so to write it out in full, 25 petabyes is 25,000,000,000,000,000 bytes of data).
By January 2012, it was the 13th most visited site on the internet, with more than 12 billion files hosted for over 100 million users.
It seems that its users were a mix of bona-fide ordinary people storing their data on Megauploads servers, and also pirates sharing copyrighted data and people storing/sharing X-rated files too.
On 20 January 2012 Megaupload’s founder was arrested in NZ, and their servers (in the US) shut down, due to claims of copyright infringement. Everyone who had data on the Megaupload servers were, suddenly and without warning, unable to access it any more.
The US DoJ couldn’t care less about this mass destruction it was causing. It said that if anyone had a problem, it was their own stupid fault, because Megaupload had recommended people keep local backups of all data on their servers. Such recommendations were buried in the FAQs and Terms of Service.
At the time of writing (almost exactly four months after the servers were all turned off) legal arguments are continuing as to if people will ever be allowed to access and retrieve the data that is stored on the now inactive Megaupload servers. The US government has acted without any trial to deny these 100 million or more users, located all around the world, access to the data they had stored in the cloud with Megaupload, and seeks to make this denial permanent.
Sure, some of the users are probably accessing copyrighted data for illegal purposes. But probably most users are normal people, and they’ve been penalized exactly the same as the others.
Imagine if the police came and arrested everyone in your neighborhood or apartment building, based on their belief that one person, somewhere, had committed a crime. ‘We don’t know and don’t care who it was, we’re arresting the whole lot of you’.
While loss of data isn’t quite as serious as loss of liberty, if the data that was lost represented a person’s work materials, it could imperil their ability to earn a living, and imagine the tragedy if it was someone’s creative outpouring representing thousands of hours of creativity. What say you’d spent the last five years writing a book, and now all your materials and the book itself have been taken from you, even though you have not done a single thing wrong, yourself?
Cloud vs LAN vs Local
It is convenient to think of computer data in three forms. At one end is cloud based data – information that is stored somewhere and which you access ‘invisibly’ through some sort of internet connection. We say ‘invisibly’ because, as long as everything is working well, you as the end-user probably can’t tell the difference between cloud based programs and data and locally based programs and data.
The two key things about cloud based data are that it is physically remote from your location, and that it is information that you don’t personally control the access to. Sure, it might be ‘your’ information that you own. But – as the Megaupload case surely shows – you don’t control your access to your data.
Being able to access cloud based information relies on many things, none of which should be taken for granted : Your access to the internet, the internet itself, the information host’s access to the internet, and the proper operation of the internet host computers.
On the upside though, if your computer has a problem, you can probably access the cloud based information from somewhere/anywhere else.
Local data is information that is physically located on your computer. You can unplug your computer from everything except the power cable, you can move it anywhere you like, and when you turn it on, you can access all your local information and programs.
Being able to access local information relies on your own computer working properly. If the computer has a problem, you can’t access your information until the problem is solved (unless, of course, you have backed up the information, either onto an external hard drive or into the cloud).
Locally stored information also makes bugging out in a hurry much easier and foolproof. All you need to do is grab your computer. You don’t need to remember to also bring a NAS device, a router, or any other equipment. Just your computer.
LAN based data is somewhere in the middle between cloud and local information. It is information that is stored somewhere close to you, within your business or home network. (Note, if in a business network, although it appears as a LAN resource it could be located anywhere.)
LAN information may or may not be externally accessible by computers outside your LAN. But, and certainly this is the case in a home LAN, it is stored somewhere that is within your physical control. You are not reliant on the internet or anything outside your home to access the LAN based data. You are reliant on your home LAN, of course, and on the proper functioning of whatever device the information is stored on, but that is all.
What You Should Do
The first thing you need to do is understand where your programs are located, and where the data they work with is stored.
There’s not much you can do about remote websites, of course, although storing offline copies of helpful/relevant webpages and the data on them is an excellent idea. You can never predict when websites might close down.
Your own data should be kept on your own computer. By all means, have backup copies in the cloud – indeed, we’d encourage it, so that if something happens to your own computer and the location it was stored at, you might be able to access the data from other computers and other locations.
But you want to ensure that if/when there are interruptions in service, your computer can still work as a freestanding computer with all the information you consider to be ‘your’ data, so you need to keep both the software and data on your computer, not somewhere in the cloud.
For email purposes, you can still keep your Yahoo, Hotmail, Gmail or whatever accounts. You simply need a front end program that will take the email from those accounts and download it all to your computer, probably using the POP3 and SMTP protocols (IMAP4 is good too although a bit more complicated to set up and understand).
We use Microsoft Outlook, which does this very well, but there are plenty of free programs you can use too (for example, Mozilla Thunderbird).