A shift to control your own data

Control the access to your own data...

Control the access to your own data...

Some incidents that happened lately led us to think quite much about the control of our data:

  • In the past couple of months, Facebook had decided to disable the accounts of a couple of friends of ours.
  • In 1999, some very important files of mine were hosted on a “Free 50MB storage” service. A year later I found out that my files were taken hostage, and in order to bail them out of that company I had to pay $10.
  • Censorship is another concern. I’ve given YouTube and Flickr the liberty to censor my movies and photos. Sometime in the future, they might decide to shutdown their service, clean up their servers, go bankrupt, become a racist totalitarian, or who knows what.
  • Some very important accounts of mine are authenticated against my GMail. If someone somehow breaks into it – I can lose control over some Very important accounts of mine.

These are a couple scenarios that happen when we give the control over our own data, to some 3rd parties who might someday betray our trust.

Very annoying.

Very disturbing, when I came to think of it.

Giving the control over my data to some company, suddenly sounds insane. Why should I trust Facebook, Yahoo!, Amazon, Microsoft and Google – with access to my personal data? Why are we forced to trust these companies, when another, very elegant solution is now possible?

Tal Muskal has urged me to think about cloud computing in the past couple of months now. Amazon, Microsoft and Google – all started to offer some cloud storage and computing solutions. These resources are quite cheap, and always available. Together with some open-source development, these solutions can be turned into something very neat.

Imagine that you pay for some storage and CPU time in the cloud. You have full control over the data, and the APIs needed to access this data. All the code is open-source, and the data is encrypted so that you are the only one who controls who gets to see what.

With this scheme, I can control exactly who is permitted to access my pictures. I can allow my Facebook-friends access to my future events, without allowing Facebook-the company access to my list of favorite songs. I can store some audio-clips, and pay $0.17/GB/Month, without the concern that a service will sometime decide to take it down.

This approach is much healthier than the current state we’re in. It removes the need to trust 3rd parties with some very important data. I believe that the ability to have full control over his own data is something that every heavy Internet user should strive for…

6 Responses to “A shift to control your own data”

  • Ori Shalev:

    Nice article…
    Why then you trust Amazon’s S3 more than GMail?
    To access S3 you need secret access keys which you may keep on your computer (unlike your GMail password), and once someone breaks into it you can be similarly damaged?

    Also, to retrieve someone’s Amazon S3’s access identifiers you may only need to break into his GMail account and tell Amazon you’ve forgot your password.

  • Eitan Burcat:

    I don’t have to trust Amazon’s keys either… I can encrypt the data that I store on their S3/EC2 using my own encryption scheme, unrelated to Amazon’s keys management.
    You’ve got a point regarding the taking-over of my whole account and deleting it completely. For this I can rely on an off-site backup.
    Yes, to an extent I’ll have to trust my cloud provider on not taking my CPU and data down.

  • Itay:

    Your alternative is not detailed enough.
    Are you talking about an opensource, cloud computing, implemented as a downloadable client, or interfaced with some main API on some company’s server?

    The international banker conspiracy to take over the world through instant access to All data and enslavement through the monetry and mass media system is only frustrated by the very existence of the open internet.

    Yet. the profit based economy will find the relevant hack undermine these concepts, and Semantinet will need to store it’s data somewhere for faster cashing, interfacing, indexing or something…

  • Eitan Burcat:

    Not detailed enough for what? It’s still just a general idea… I’m sure there are TONS of issues (if and) when I’ll delve into the details…

    The direction I’ve thought about so far (Tal gave me some hints on this :) ), is an open source software running on EC2. Let’s call it an “Angel”. (Because angels seem more appropriate for living in the clouds than the “Agent”s we’re all used to ;) ).
    This Angel can be controlled by the user from any browser, or in a scheme similar to OpenID. No need for a client nor any company’s server (except for the EC2, or Google’s AppEngine or whatever Microsoft will call their cloud…).

    I’m not a fan of conspiracy theories, I have no idea what you’re talking about :)

    This post has nothing to do with SemantiNet.

  • TalM:

    Great article (i like the picture)

    Some advantages i see in personal cloud computing:
    1) Data portability, export/import all my data and metadata or share with other services. – if you decide to migrate to a similar service.
    2) Flickr/Lastfm/Upcoming/Facebook/GMail like services can be either a open source plugin you install on top of it or a third party website with access to your data.
    3) ability to view any angle that i choose of all my data in a simple unified way. (through a RSS viewer)
    4) doesn’t need to have a business model or profitable in any way. (GPL/BSD software on top of existing cloud platforms)

    Ori, i don’t think amazon encrypts the data in memory, and since the the memory is shared with other virtual machine, potentially a hacker can take advantage of that.

    Since it shouldn’t be so easy, it seems is pretty safe for this kind of use. someone just needs to raise the glove and code (or modify) a proper open platform to run on an AMI.

    another great post:
    http://radar.oreilly.com/2008/10/web-20-and-cloud-computing.html

  • This worries you?
    “Do no Evil”…today…

    Who guarantees what Google (a commercial body) will elect to do with all the data it has in the future?

    cheers,
    mike

Leave a Reply