Magic Bus

Note: If you think Twitter is a joke, or that Microsoft has no chance of being trusted with anything, or that Apple will fall to the surging citizen militia, or any other common wisdom of the Internets, please move along, I beg you.

I thought I’d take a little more time to explore the reasons why I think Microsoft’s Live Mesh can emerge as the host platform for a large percentage of Twitter-like behavior, and why this is so important to the various strategies of all the players in the enterprise space. To begin, let’s go back to the previous mindset for using personal computers, the one where there were things called applications and operating systems and monopolies and such.

These monopolies (Windows and Office) derived from the transition from the previous era of computing, where IT maintained control over the local area network and the applications companies allowed their workers to use. Microsoft’s support of TCP-IP in Windows for Workgroups and Windows NT when the consumer and professional streams of the operating system converged transferred control from IT to a loose federation of internal and external administrators of directory services and network permissions.

Soon an average civilian could set up tunnels between remote and home offices to operate (albeit slowly via ISDN) client server software such as Lotus Notes, Exchange Server, and the beginnings of virtualization solutions via Citrix. From a consultant perspective, the first of what now has become cloud services was barely but strategically useful. You could bootstrap LAN-based applications over tunneled connections and begin to develop the concept of a cross-domain enterprise tied together not by location but by business process.

As these two contiguous networks built out – the public Web and private corporate hub-spoke cells – individuals became more autonomous in their mobility and responsibilities. To IT, this seemed relatively marginal in terms of their control of the corporate infrastructure; mobile devices were nowhere near small or secure enough to have much impact, and laptops could be controlled via Notes replication and ActiveX sandboxing of company workflow and data.

Microsoft sowed the seeds of IT’s loosening of control with Y2K’s battle with IBM for control of the messaging market. Notes’ Domino transfer of access to messages to a Web interface drove Microsoft to develop Outlook Web Access on top of what became the Ajax framework. Even though that effort was slowed and wrested away from the Exchange team by the emergence of .Net, the Ajax genie was out of the bottle and adopted by a succession of competitors, most decisively Google with Gmail.

These on-demand applications still had limited security and richness compared to behind-the-firewall apps, but businesses soon realized that they were competing for their worker’s time against the rapid growth of the Internet as a source of information and communications. Much as IT would prefer to maintain control at the intranet level, Web 2.0-aware CIOs and CTOs started breaking the company’s information flow up into eyes-only, company-only, and everybody-only (or marketing) streams.

This brought the opportunities and liabilities of Scoble at Microsoft, blogging as the new media, and the new IT jungle of social media, where everyone’s identity is time-sliced up according to these very different but parallel-running information streams. Facebook’s attempts at canonizing behavioral signals by product and event tags (Beacon) were poorly received, and Google’s attempts to harness these gestural signals have met with resistance over mis-use of a social graph obtained under separate contracts with users.

While Facebook taught entrepreneurs the value of an API strategy, Apple used Microsoft’s disarray over the Gates transition and a succession of OS delivery failures around Longhorn to finesse the transition to Intel and into the iPhone era. Seemingly the opposite of an open API approach, the iPhone took down the same Ajax space with a usable Web experience and pivoted to it with its early support for browser apps while the native App devkit and Store were being built.

While Apple went after Microsoft with Intel and iPhone, Microsoft began building Mesh, essentially a rewrite of Notes replication over open protocols with FeedSync combined with an atomization of social media primitives into a new platform on which to build applications that are identity rather than hardware or native OS-centric. Today, we see Live Mesh as about virtualizing files from the containing device over a Web hub, but at a deeper level, the Mesh is as much an information router as a bit traffic cop. How to act on the data becomes more strategic than the underlying job of moving things around to follow the user.

What’s missing in that new equation is a guiding mechanism for tracking user intentions. In a desktop centric environment, calendars, documents, and email capture most of the decisions about who does what, when, and how. In a mobile network, it’s the intersections between those stores that produce change on an ongoing basis – a dynamic group calendar and the threaded conversations that inform it.

And it’s here that simple wins – a lowest common denominator approach to playing the underlying notes, the rhythm section if you will. Recall the Watergate question: What did you know and when did you know it? The payload of the canonical message: Who, What, When, Where. Who is the social graph – who is the source of the information and how do I rate it? What do I need to know, when did this happen, and where do I find more? Yes, dear friends – it’s a tweet.

I’ll leave the breastbeating and rumination to the FriendFeeders, but the synchronous convergence of 1) a device to both capture and transmit this data anywhere at anytime (iPhone), 2) a swarming attractor to form the dynamic of social graph and dynamic discovery in real time (Twitter/Track), and 3) an organizing framework for building a reliable transport and applications that take advantage of dynamic social signals (Live Mesh) – together represent a storm system that is moving fast.

Take the speed with which Twitter has absorbed Summize and Gnip, recalling aspects of Google’s harnessing of relevance-based search and the notion of the enterprise service bus as a way of accelerating enterprise application integration. It’s not difficult to imagine passing aggregated Gnip streams along over Amazon’s Simple Queue Service to extend enterprise-grade reliability. Or use Mesh’s new ability to bypass cloud storage and go peer-to-peer using the cloud for encryption and NAT transitions.

The iPhone’s growing pains aside, the rollout of the App Store and Mobile Me foreshadow how Mesh will work at the user level and soon under IT management between cooperating corporate clouds. With the genie out of the Twitter bottle and client services like Twhirl hosted on AIR and soon Silverlight, we should see consoles emerging that make real-time information management less stressful than air traffic control.

We’re seeing social services merging and learning how to talk over this new social service bus, whether by acquisition, standards negotiation, or amazingly, complete leopardization, i.e. changing of spots. As Pete Townshend wrote:

Every day I get in the queue (Too much, the Magic Bus)
To get on the bus that takes me to you (Too much, the Magic Bus)

By the way, the name of The Who’s record company? Track.