The Cloud Conundrum

A lot has been written about the cloud lately. Everybody's getting into it, businesses, service providers, even grandma with her iPhoto stream. The problem is, that despite many different flavors of cloud technology being out there, it's just not there yet. I'm not an expert on the implementation of cloud solutions, so I won't claim to understand the hurdles facing someone trying to set up such a thing. But I do want to take a moment to address the things that irritate me about current offerings and what I would like, no, love, to see from a cloud service.

What I Want From the Cloud

As an individual user, I have certain basic needs that need to be met before I'm ready to fully adopt cloud solutions.

Data Control and Security

I want to control my data and own it. I want this as an iron-clad guarantee. I want to have the ability to always and at all times pull my data off cloud storage and expect that data to disappear off redundancies within a reasonable period (say a few days).

Cloud storage is being pitched to people as their always on, always available hard drive away from home. At this moment, there isn't a cloud service I'd trust with pictures of my kid. Things like the Instagram debacle only hurt this cause.

I also want Data Privacy and Security. This means data encrypted before it hits the cloud, using encryption keys known only to me. I don't want anyone at the Cloud provider, or the company that eventually buys them out, or a foreign government on whose territory they have a data center, to be able to snoop through my private data. Also, obviously, this helps harden my data against eventual hacking and security breaches.

I realize the above is getting trickier to square with the legislation of many countries now requiring a service provider be able to access and display all the data they host when asked by an appropriate authority. I started writing this post before the NSA phone surveillance scandal broke and my suspicions about governments using big data techniques to watch us were rather spectacularly confirmed. I can only state that these recent stories only serve to confirm the need for data security and control over our data. Unfortunately, these very developments will likely be hindered by the powerful interests arrayed against such privacy.

Still, I can't give a cloud provider my full trust without such precautions in place.

Data Access

Part of owning your data, is being able to access it. Many cloud solutions today are silos. As an eminent example, take Apple's iCloud, where different parts of the Apple ecosystem can access iCloud in order to share settings across devices, or to recall state like 'take me to the point in my ebook that I got to when I was reading it on my iPhone earlier'. These are certainly useful functionalities. But I have no idea where on the cloud the data resides, how much of it is stored or for how long. Apps in iOS cannot freely access data of other apps even from iCloud.

Clearly, iCloud storage is more of a repository for the data necessary for apps I use rather than storage space I can access and manage on my own. It is storage that will be managed for me rather than by me.

Apple would like it if we no longer had to think of the 'file system' and could just get on with our work which is a lofty goal but not feasible yet until we find a handy way to interact with data that is not file-based.

Right now, if I want to re-use a photo I took, I still need to know where it is located so I can open it within the app I'm using to process it.

I want to know where my data is.

Furthermore, I don't like data silos. I want all my apps, in theory, to be able to see all my data.

Space and Bandwidth

Amazon Web Services has done a pretty good job setting the standard here, but they are targeted at a more technical audience. The cloud has to be virtually unlimited space, that's part of the point of the exercise. I want to forget about gigabytes, terabytes and petabytes. In the world of today, people should be able to burn through as much storage as they want.

Of course, there can and must be costs attached to it, but as Amazon has proven, those costs can be very palatable if economies of scale come into the picture.

The same applies to bandwidth, I don't want to worry that a popular video I've posted on my cloud space and made publicly available suddenly goes viral and starts drawing huge amounts of bandwidth. Here too, the same rules apply as with space, the cloud provider should be able to provide the bandwidth at a price that's not going to bankrupt you within 10 seconds (or at all, ideally).

The promise of the cloud stated in its most ideal form is: secure access to your data, accessible only to you, anywhere and at any time with no restrictions applied. Obviously, in practical terms certain restrictions will apply but from a consumer perspective, the previous statement is the goal I would like my cloud provider to be tirelessly striving towards.

One Provider to Rule Them All

In the end, I don't want to be bouncing between many different providers for my things. I'd like them all in one place. If that is asking too much, I'd like all the different places I have my data stored to behave as if it was one big data store I was accessing. That would require a degree of common interest and architecture that I don't believe companies would be in too much of a hurry to set up.

Either way, I want the Cloud to hold up to the promise of that term. It's a kind of hazy place floating lazily in the air of which I need to understand very little. I don't want to know about space, bandwidth, security, encryption or any of that technical jargon. I want to be able to grab my smartphone, and call up my data, from anywhere and at any time.

Is that asking too much?


Well. Yes. It is, actually. I have mentioned many of the hurdles above, but I'll list them out here one more time so we have a complete overview of what's standing in the way of our ideal cloud solution.

  1. Data security: there are too many vested interests against us having total security of our data. Rightly or wrongly (this is a long discussion for another time), governments and other institutions feel they need the ability to scan/collect our data for national security and/or law enforcement reasons.
  2. Data privacy: there are still too many business models predicated on the ability to collect data on users. When there is as much data collected in one place about a person as such cloud services will be able to provide, the temptation for a company to indulge in analytics is often too great to resist.
  3. Legal issues: laws about what data may be stored where, who may or may not access it, and so forth, will be difficult hurdles for a cloud provider to overcome especially as they will have to do so in each country they wish to provide their services. Many countries already have legislation in place preventing the use of certain types of cryptography, etc.
  4. User experience: so far, most companies seem not to have gotten this right. They have not yet grasped that it's all about empowering the user, making access to their data as secure but as easy as possible. No data silos, no apps that lock you in to using them, etc.


It's going to be interesting to see which way things develop. The Facebooks, Googles and Apples of the world are struggling mightily with each other to see who can provide the best ecosystem but, thus far, their focus seems to be on what they can get out of the cloud rather than what their users can get out of it. Perhaps other upstart services might eventually provide a solution, but I wouldn't hold my breath.

So far the personal 'cloud' is still a 'haze' as far as I'm concerned.

The Matrix Man Machine Perspective

The Matrix Man Machine Perspective

For My Daughter