You are here
How Will Data "In The Cloud" Be Stored and Accessed?
Cloud Computing has been a popular buzzword in high tech as of late. The actual meaning of the term is murky; for background on the its meaning see Infoworld's fantastic overview. For myself, Cloud Computing is primarily about developer access to cheap utility computing (as great as that is!). Rather, it is about consumer's transition to software as a service (SaaS) and the shift towards storing data "in the cloud".
To keep up with developments in Cloud Computing I've become a reader of Elastic Vapor, the blog of Reuven Cohen, co-Founder & CTO of Toronto based Enomaly. Last week Reuven posted about the offline cloud, which I think of the "airplane problem" due to a memorable rant by 37signals. Reuven's approach to offline access is less opinionated, ending with a simple question of how the access problem will evolve:
My question is; are we moving toward a future where the desktop is nothing more then a local cache? Or will we soon reach a point where technologies like Wi-max, 3G and wifi on planes make the internet a ubiquitous part of our everyday life where local storage isn't needed at all?
To respond to Reuven's question directly, I think that we will see increasing use of caching even as connectivity increases. This combination (more caching, even if it isn't needed as badly) will eventually make most connectivity issues moot. To separately comment on web applications in general, I do not think that the traditional desktop application will be completely replaced by the web applications over the next 20 years, just as TV has not eliminated radio (I know, the comparison is weak at best). However, I do believe that the new paradigm (web applications and online storage) will eventually become dominant. Browsers will become more powerful, gaining traditionally desktop features through technologies such as Gears caching content locally and web browsers develop site specific features.
Beyond browsers themselves, open standards such as WebDav, CalDAV, IMAP, SyncML and even LDAP are already allowing various desktop applications to push, pull and cache data from the cloud. WiFi is increasingly ubiquitous and quick, decreasing reliance on stale cached data. The end result is a quicker, more powerful, (almost) always on online experience.
Despite all of its problems, the recent launch of Apple's Mobile Me is a clear signal of where Cloud Computing is trending. For the most part, the service is a actually a re-packaging of existing .Mac services, with the marketing message being the largest change. .Mac was known for keeping multiple Macs in sync through the web via IMAP, App Syncing and iDisk. The marketing focus was on the Mac Desktop experience. With Me.com the emphasis is clearly on the cloud itself, with access through any type of device you may have. The paradigm has shifted: online data, not the desktop, is now the master copy. What was called Desktop synchronization is now labeled as pushing and pulling from the cloud.
Besides Apple, Google is the other prominent example of a hybrid desktop/web application model with centralized online storage. Google's apps all started life accessible through the browser only based but have become less dependent in the past year. Gmail now supports IMAP and has an API for your contact system, including automatic synchronization with Apple Address Book, if you're lucky enough to use an iPod Touch or iPhone. Besides launching Outlook support, Google Calendar has added CalDav integration so that iCal syncs perfectly with Google Calendar, with the Cloud once again being the master copy.
I've already been witnessing this trend with my own computing habits in both personal and professional realms. For our business, almost all of the applications used live on the web if there is a suitable solution. However, in cases where we have a choice between accessing these through a traditional desktop application or via a web browser we tend to favor desktop access. Expect more on our own cloud based systems in another post.