The Cloud Promise

Having submitted a talk on html5, and calling it “The Language of Cloud Computing” (please vote for it if you are interested), I thought I should take this opportunity to discuss how I see the possibilities of the cloud. I do web development in my day job and we use some of the technology that is now discussed as cloud technologies, but this is my perspective as an end user. I will put a warning here, that there is a bit of blue-sky thinking in this post.

My original germ of this post came after reading this post by Gary Short: Cloud Computing – It’s that New Old Thing – Gary’s Blog and as I’ve been writing this post, I think that he’s the angel sitting on one shoulder telling you to be careful, to watch out for snake oil and to mind the gap, which makes me the devil on the other shoulder telling you to go on, jump in, the water’s lovely, and what could possibly go wrong? With that in mind, I recommend you read his post and keep it in mind.

There are two distinct shifts I have seen that are classed as “Cloud Computing”. On the server side, machines are becoming virtualised which allows for greater flexibility and a more efficient use of resources (go green rangers). I use VMs all the time, and I think they’re great. Of course, mine aren’t hosted across the world, like Amazon Elastic Compute Cloud (Amazon EC2) or Windows Azure. On the client side, it’s about the trend to save your data “somewhere else” so you don’t need local storage and you can access your data from everywhere. That seems to be the underlying vision of Google Chrome OS and the Apple iPad and it has a certain appeal when you don’t have to worry about backups, running low on disk space, and other maintenance tasks. Why not give it to someone else to worry about? And if that’s not enough, put your photos on Flickr!, edit them on Picnik, and then share them on Facebook, giving a lot more options than you could have on your own, and allowing others to manipulate, improve, and remix your data.

But what if you pick the wrong service? Picnik currently doesn’t work with Windows Live Photos so you miss out on that service unless you upload the photos from your computer. And what happens when you’ve been using Yahoo 360 for blogging and it disappears from the web?

What we realy need is some proper data portability. You don’t want “Facebook compatible” or “Works with Flickr”. Google Reader is great because it works with RSS, and there’s a lot of RSS out there to work with. We want something as simple as “Yes, it’s a JPG, I can work with that”, just like I can on my computer. And more than that, we need to be able to move our data from one cloud provider to another.

My ideal model for how “The Cloud” should be working, for me, would be that every cloud provider would support standard data transfer protocols, whether it’s ftp, imap for email, or something new, that supports open authentication to transfer directly between providers as well as to my local storage. The key is making it easy to synchronise my data. Yes, some of it will be world-accessable, with all the privacy issues that presents, and yes, some providers will be better at some things than others, but making it easy to transfer my data between providers means that I can move around to get the best of all worlds, instead of choosing this place for greater storage instead of that one that lets me share, or the other one with better editing. There’s loads of people who make it easy to get your data in, they support multiple clients, have a nice web uploader, annd so on, but it’s often a lot harder to get your data out. The Data Liberation Front (the Data Liberation Front) is a nice step in the right direction from Google, and I think is a good, basic philosophy for Cloud providers to follow, but that should be the bare minimum I should expect from a provider. If they don’t follow that model, they are not “The Cloud”, they are just silos full of quicksand, and the more you struggle to get you data out, the more it seems to be lost forever.

So, will html5 help with any of this? AJAX has become a reason for providers to become more open, to encourage traffic back to their own sites, and adverts, but with html5, we get, for example:

  • microdata, to help define areas of pages that other sites can understand, such as calendar events and contact details;
  • structural elements, like nav, to help parsers find the interesting data;
  • Cross-document messaging to allow pages from one domain to communicate with pages on another domain, with a security model to prevent XSS attacks; and
  • protocol handler registration to allow a website to declare that it can handle fax: or mailto: links, or jpg or apk files so I can grab a link or file from one site and sent it directly to another site that I trust.

I don’t know how many cloud providers will start to use these features, but until they do, the cloud is stunted.

Blogged with the Flock Browser

2 thoughts on “The Cloud Promise

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s