Maybe the Backup Problem Will Resolve Itself In Time

I keep saying this over and over again but it is like speaking to a wall. The most valuable piece of technology related advice I could EVER give to anyone in my lifetime is probably this:

BACK UP YOUR DATA AS OFTEN AS POSSIBLE

Your hard drive will fail – it is just a matter of time. If you don’t back up your data, you will lose it. To bad no one ever listens to me. The people who agree with this notion already backup their data quite diligently or simply choose not to. But that’s their conscious decision and their gamble. The people I try to reach usually shrug and claim they don’t have anything that important on their drive. Only after it fails they come back crying about their vacation pictures, their mp3 collection, their unfinished screenplay and all the other stuff. Btw, when people say “and all the other stuff” 90% of the time they mean their pr0n folder – just so you know.

What’s worse – these people never learn. It’s like tilting at windmills. You keep telling them, and they shrug it off and say they will take their chances. Then every time the hard drive fails, they go through a genuine nervous breakdown because all of the irreplaceable stuff they have lost.

In the past I often wondered how to fix this issue. How to force reluctant users to back their data up in a non-obtrusive and non-annoying way. There are many ways to do it (from incremental backups running in the background to periodical scheduled disk imaging and etc) but all of them require some setup, support and at least some degree of user cooperation. Which as I mentioned is often hard to obtain. For example, I’d love to set up some sort of backup plan for my frequent offenders but they simply thing it is a waste of time and wont let me. And so the cycle repeats itself every few years.

Recently however I saw a light at the end of this tunnel: Cloud Computing.

cloud.jpg

I noticed this trend a while ago, but I didn’t really realize it’s scope until recently. Regular users are slowly but surely moving their data off their hard drives and onto the cloud where it could be shared and/or accessed from more than one location. I’ve been commenting on this for a while now but I have failed to call it by name up until now.

For example, a while ago I remarked how many users at my workplace completely bypass the file system most of the time. They use their email both as remote file storage, backup and a version control system. It made no sense to me at the time since we used POP based email and Outlook, but this is a trained behavior. Out there in the wild, POP is being slowly phased out and replaced with a rich, AJAX driven webmail clients such as Zimbra, Gmail and etc. Email these days is almost always remote, so when people email their files to themselves they are actually pushing their data onto the cloud.

Your average user keeps their important files such as homeworks, term papers and resumes in their Webmail. Their pictures are safely stored in their Facebook and/or Flickr account with various privacy and access options applied to them. Same goes for just about any other type of file they produce. If it is to be shared, it ends up on some sort of social sharing service. If it is private, it goes into the email. Their applications are hosted online. They pirate movies by steaming them from less often patrolled YouTube clone sites. The local drive is just a temporary staging area which carries no importance.

Quite a few people already migrated all their irreplaceable data to the cloud. Among them are some of my students and co-workers. They already live in a world where hard drive failure is just a temporary setback. To these people my insistence on regular backups may seem silly and/or old fashioned. After all, their data is already sitting on some nebulous cluster in the sky and chances of losing it are astronomically low. I’m just a silly little man, tied to a local hard drive and obsessed with data redundancy, when their data is redundant by the virtue of being stored on the cloud.

As this trend continues to grow and expand, backups may indeed become a thing of the past. The art of making backups will become an exclusive domain of geeks with large local collections of software/multimedia and corporate IT professionals. In the future most regular people will happily live off the cloud and never worry about backups.

Naturally storing your data on remote clusters owned by 3rd party companies have their own downsides and trade offs. Privacy is a big concern. So is access speed and offline availability in case of a network outage. However, many people are willing to pay this price just so that they don’t have to listen to me drone about backups. Which is not that bad actually. It is one less thing for us to worry about.

This entry was posted in futuristic musings. Bookmark the permalink.



9 Responses to Maybe the Backup Problem Will Resolve Itself In Time

  1. dawgit GERMANY Internet Explorer Windows says:

    Ok, it’s one way. But what happens when the Cloud Breaks? (I know there will be a down pouring of useless information) No, seriously, That’s a precarious place for any type of information that might be of importantance. There already far too much personal information floating around out there, getting hijacked, and abousd. The ‘Cloud’ does have a valuble and bright future, but not I hope as a means of ‘Back-up’. -d

    Reply  |  Quote
  2. dawgit GERMANY Internet Explorer Windows says:

    – sorry all, my spell-checker has the day off. -d

    Reply  |  Quote
  3. Luke Maciak UNITED STATES Mozilla Firefox Ubuntu Linux Terminalist says:

    Well, the thing about cloud is that it is not supposed to break. For example – how many times last year did Google’s cloud go down? Zero. There were glitches here and there I’m sure but as far as I can tell there was no major downtime or data loss. They have many data centers, and their data is distributed, mirrored and very well maintained.

    Statistically, your data is safer on the loud than on your hard drive. It is much more likely that you will get disconnected form the cloud due to a local network outage than that the cloud will be disconnected from you because it is usually distributed, and redundant.

    The big difference is that you own your hard drive, but not the cloud. The big concern is what happens when the owner of the cloud decides they no longer care for you as their customer and delete your data.

    Another is security and privacy. How do you know that the company on whose cloud you are storing your data is not selling it somewhere or sharing it with some 3rd party?

    But I agree – it is not a replacement for backup. But if it is preserving people’s data in one way or another then I’m all for it.

    Reply  |  Quote
  4. dawgit GERMANY Internet Explorer Windows says:

    I think we’re actually agreed on this, you have just stated it more elequently.

    Reply  |  Quote
  5. IceBrain PORTUGAL Mozilla Firefox Debian GNU/Linux Terminalist says:

    Actually, the “cloud” can be a great way to backup. I have now a 2GB Dropbox account, and thanks to some bash scripting each file is automatically encrypted and copied to dropbox.

    Reply  |  Quote
  6. Luke Maciak UNITED STATES Mozilla Firefox Ubuntu Linux Terminalist says:

    @IceBrain: I actually use Dropbox to sync files between my desktop, laptop and work computer. I don’t really use it as backup – just as an extremely convenient way to move files between my computers. Just copy the file to the Dropbox folder and forget about it. I absolutely love it.

    That said, I have a sell script running on my home desktop that backs up my personal files to an external hard drive once a week.

    Reply  |  Quote
  7. vacri AUSTRALIA Mozilla Firefox Windows says:

    I am extremely lazy when it comes to backups… so I run RAID1 at home. It’ll protect my data from a lost hard drive. Not from my own stupidity, admittedly (yet to have that lose valuable data though). I do keep random greyware installers on a small NAS box, but documents and stuff… I just don’t have many of them, and those I do have (like my resume) could probably benefit from being rewritten from scratch.

    Reply  |  Quote
  8. Sarel UNITED STATES Mozilla Firefox Windows says:

    I’ve noticed over the years that any backup scheme that requires the user to do something on a set schedule such as daily or weekly just doesn’t work. It needs to be totally automatic and in the background.

    Recently I started using rdiff-backup. It’s a great little tool that backs up incrementally using delta compression, just as rsync does. However, it also acts as a version control system. You can tell it to keep the last 60 days worth of backups and it doesn’t require 60 times as much space. It just uses a little more space with each backup.

    It can back up locally or to a remote system via SSH and because of the delta compression you can back up huge amounts of data in a very short time. I have it backing up to my linux server with RAID1. A scheduled task in Windows kicks this off every morning while I’m sleeping (usually).

    To me the only pain with rdiff-backup is that it is difficult to setup. I would really like to write a Java front-end for it so users without the ability to get this going can use it.

    If you do decide to use it make sure to apply the 4GB patch to librsync otherwise you won’t be able to backup a file larger than 4GB.

    Reply  |  Quote
  9. Kenny CANADA Mozilla Firefox Mac OS says:

    I agree with Sarel, if the user has to remember to do something, they won’t do it.

    The *only* success I’ve had with getting any Windows users running backups is with online, completely automatic, idiot-proof backups. There are a couple of them, but I usually recommend Mozy. Unlimited backup for $5/month.

    Disclosure: I also signed up for a Mozy Affiliate ID. So if you use my link to signup they may give me a buck or two.

    Reply  |  Quote

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>