Even though we still don’t have “wireless everywhere” (as I like to say), access to the Internet is indeed becoming more pervasive. Until the world is blanketed in wireless however, there will always be a place for offline applications. Sometimes you need to get some work done, with or without an Internet connection. Unreliable access or no access at all might have been the driving force behind offline applications in the past, but now there’s a new reason: cloud computing.
The term “cloud computing” is a bit like Web 2.0 in that it is used as a blanket term, but essentially it means accessing applications and services via the Internet (“in the cloud”) without worrying about the infrastructure that supports them. One of the best examples is GMail, Google’s email service that lets you manage your messages in any browser. It’s also a good example of why offline, synchronizing applications are so important – GMail went down completely yesterday:
Gmail is having a systemwide outage affecting multiple countries, and a whole bunch of its 100 million users are screaming about it on Twitter. Around 20 million people visit Gmail each day, according to Comscore, and they’re all seeing the same message. The first outages were reported at about 2 pm PST, 44 minutes ago.
One of the things that makes cloud computing different than services in the past is that more and more businesses rely on things like GMail to operate. When it goes down, so does a significant part of their business.
There’s a transition underway. Businesses are realizing that it doesn’t make sense to operate their own data centers and services when Google, Microsoft, and others can do it far more efficiently. But don’t let those names fool you, as GigaOm points out:
If an outage of this magnitude can strike Google, the company with a fearsome infrastructure, I wonder who — if any — can plan for the worst.
It’s extremely difficult to maintain 24×7 operations, even for a company like Google. The only reasonable thing to do is assume that service will go down at some point, and to plan accordingly.
For that reason, I think offline access and synchronization are two things that developers will need to focus on in the future. Like the other big challenge facing developers, multi-core computing, improved technologies and toolsets will be needed. Vendors are working on it, Google with Gears and Microsoft with the Sync Framework, but there’s still a lot of work to be done.
Cloud computing is great, and I’m excited about the opportunities that it provides. We have to realize that it’s only part of the equation, however. Offline access and synchronization are more important than ever.
Mack, very good points. I’d like to point out that having “wireless everywhere” will still not remove the need for offline access. Google’s issue would not have been prevented by wireless everywhere.
Also, one of the things I always try to stress to people is that on the Internet you are always dependent on others. Take the major outages in the US a few months ago when Fiber lines were severed in the Midwest. No matter what I do in my data center I have no control over the network links between the user and the data center. It is possible to mitigate this somewhat but there will always be a need for offline applications.
Hi Sean, thanks for the comment. You’re right, the same strategy should apply to connectivity, even if it were completely pervasive – assume it will go down, and plan accordingly.
Gmail and MySpace are SaaS web 2.0. Adding AJAX and mashups are 2.5 when talking to cloud services. Software + Services is 3.0 with cloud computing being only half of the software+services story. S+S has an inherent need for offline or caching capabilities to be useful and scalable.
I think of clouds computing as opaque services behind the veil, that need client software or other intermediate cloud-services to do useful things for specific problems.
Looks like Microsoft’s Live Cashback site crashed during the “Black Friday” shopping frenzy.
When the big “giants” can’t keep their sites up, it doesn’t necessarily inspire confidence in cloud computing.
On the other hand, our office relies almost entirely on online apps (Gmail, Google docs, Salesforce), and we’ve had very few problems. The occasional outage, yes, but nothing too serious.
Then again (or on the other “other” hand): I’ve been working on some WordPress sites for clients – and I can’t build those sites on the bus. I have to build them locally – running Apache and MySQL on my Mac. Very annoying – in the old days I’d build everything locally using Dreamweaver.