Twitter-Like Activity Streams: The New Center Of Gravity For The Social Enterprise
Although many social enterprise solution providers are betting their R&D fortunes on Twitter and Facebook-like activity streams, will lack of standards hasten adoption?
With the social enterprise cognoscenti due to descend on Silicon Valley next week for the Enterprise 2.0 conference (disclosure: the event is produced by InformationWeek's parent TechWeb), the debate over just what exactly is the "social enterprise" will once again be open for discussion. Most of those cognoscenti know it when they see it (or at least claim to) and many armchair quarterbacks (the millions of Twitterati whose 140 character profiles say "social media expert") will likely say, "oh, you mean Facebook for businesses, right?"
In the context of running a modern organization, the main question is whether the category is defined by a set of products (Jive, Chatter, Yammer, etc.) that attempt to democratically herd the wisdom of the organization and its constituencies (i.e.: customers, stockholders, etc.)? Or is it simply about any means to one end: efficient collaboration? If the latter, where does the line get drawn? Does email -- a decades old technology -- count? What about Lotus Notes (etymologically the progenitor of the social enterprise's ancestral "groupware")? For that matter, maybe FTP qualifies (for you twenty-somethings that see the world through a series of AJAXy web forms, that's the File Transfer Protocol - arguably the first means of electronic file sharing).
While the boundaries remain a subject for debate (even one that burns brightly within the confines of TechWeb's firewall), the center of gravity for the social enterprise became crystal clear at this past Spring's edition of E2 in Boston. It's the activity stream.
The world has Twitter to thank for the popularization of activity streams. But it alone isn't responsible for making activity streams the center of gravity for the social enterprise.
Twitter is for the most part one big massive activity stream. Networks like Friendfeed (now a part of Facebook) and Tumblr did a good job of taking it a step further by viewing all other social sites (Twitter, Flickr, YouTube, etc.) as a federation of networks from which to extract and aggregate events into silos of personal lifestreams. With IT people and the general counsels' offices equally aghast at the willingness of junior employees to collaboratively leverage these great new "tools," solution providers like Yammer (the so-called "Twitter for enterprise") stepped in to privatize the activity stream ("OK kids, go for it, but keep it inside the fiewall, will ya?"). Eventually, Facebook ripped a page out of every social network's playbook, putting the idea of federated activity streams and event aggregation on steroids. At the last E2, I lost count of the number of times I heard "Facebook for enterprise."
As far as I'm concerned, I pay little attention to what people are saying and more attention to what people are doing. Maybe it's fool's gold for them, but, on behalf of the solution providers who were in the building, the last E2 also signaled the single biggest infusion of hardcore R&D cash into social enterprise yet. The shift was conspicuous. It was as if most of the existing vendors in the space and a whole bunch of new ones woke up to idea of the business activity stream and came to E2 to see whose solution would prevail. And where certain vendors missed the boat (for example, Microsoft's Sharepoint 2010), others like Newsgator were happy to step in and complete the portfolio for them (watch my ReviewCam of how Newsgator added activity streams to Sharepoint).
Not surprisingly, no two solutions were exactly alike and often, they leveraged their creators' existing portfolios. Cisco's Quad for example was masterful at sweeping up events from Cisco's solutions and dumping them into a funnel that emptied into one or more Quad-hosted activity streams (see the demo that Cisco's Murali Sitaram gave to my colleague Fritz Nelson). That's great if you've standardized on Cisco for a lot of your infrastructure. In some cases, event aggregation was a function of the partnerships that the solution provider had forged so far. For example, by the time of the Spring E2 event, IBM's cloud-based LotusLive was intermingling its native events with events from Skype and Salesforce.com. Although it wasn't present at this spring's E2, Salesforce.com's Chatter was another example of how activity stream solutions were playing to their creator's strengths. Not only can it feed off of events coming out of Salesforce.com (the company's namesake salesforce automation and customer relationship management solution), third party developers who participate in Salesforce's AppExchange can also easily integrate the events they generate into Chatter-hosted activity streams.
The more and more I saw these enterprise activity stream solutions, the more I wanted to call them micro-event management systems. The Holy Grail for this part of the social enterprise is for any solution to be able to integrate its own event production with anything external that's also capable of producing an "event." The next step is to filter and direct the resulting intermingled streams to users on the basis of some automatic or human-programmed relevancy.
The biggest obstacle to reaching this Holy Grail is a lack of standards. Events from one system should plug-and-play into any activity stream host. But such interoperability is a pipe dream today (Make sure you read my update below, regarding http://activitystrea.ms. Although they'd never say it, my sense is that most solution providers selfishly prefer that there be no standards during this embryonic stage of the industry. God forbid there should be a standard that neutralizes any handful of vendors' opportunity to dominate the niche, and wipe the others out. This lack of industry unity is tragic for at least a couple of reasons.
First, there are few if any organizations that don't collaborate with other organizations. If the promise of these social enterprise tools is really to take the friction out of collaboration, then, presumably, organizations would want that efficiency for inter-organization collaborations as much as they'd want it for intra-organization collaborations. But imagine being a lawyer, or an ad agency, or public relations firm who collaborates with hundreds of clients. Because there are no standards, should you be expected to tune into hundreds of your clients' activity streams, all on different platforms? The platforms they picked? Or should you be able to take your activity stream host of choice, and point it at your clients' systems -- gracefully collecting and organizing the events from each for your consumption? The same of course is true in reverse. If you're a client to that lawyer, ad agency or PR firm, do you even want to have to deal with the dissimilarities between systems to successfully collaborate?
Another reason this lack of unity is tragic is that the industry seems incapable of learning from its past. History has proven over and over again that the odds are against proprietary walled-gardens where open standards could blaze a path to interoperability. In contrast, adoption typically shoots through the roof when standards are in play, improving the business opportunities for all members of the ecosystem.
Philosophically, Novell is the closest to having the right vision. It's just that it may have placed its bet on the wrong horse. At the last E2, Novell demonstrated how, using Google's open Wave Federation Protocol (WFP), the same activity stream could be hosted in both Novell's Pulse and Google's Wave. The idea being that if you picked Google's Wave as your company's activity stream host of choice and your ad agency picked Novell's Pulse as its activity stream host of choice, neither of you would have to compromise. Both of you could have visibility into the same events. But despite its open source nature, WFP is by no means a standard. Not only don't any other activity stream hosts or "event providers" support it, Google discontinued development on Google Wave itself earlier this year. Theoretically, work on WFP can carry on. But so far, there's been no rallying cry from the industry to get behind it. For all its good intentions, even Novell's Pulse is an island of technology.
The editors at InformationWeek see a lot of promise for this category of social enterprise solutions. But in the same breath, the lack of standards is enough cause for concern that I suggest you proceed cautiously as you evaluate your options. Nothing motivates solutions providers more than your money which is why I encourage you to use your buying power to raise the standards issue. On the industry side, if you're a solution provider and you're ready to have the standards conversation, there's no better time than at E2 to have that conversation and we (TechWeb) would be more than willing to organize and host it. Just let me know. I'm easy to find.
Update (11/5/2010): In response to this column, SkypeJournal.com editor Phil Wolff pinged me on Twitter to ask if I had heard of the http://activitystrea.ms standard. In that tweet, he says that it's "embraced" by Facebook, MySpace, Microsoft, and Google. I responded that I had heard of it, but in a context that it was a market failure. I was referring to Activate co-founder Anil Dash's on-stage interview of Twitter Platform/API Product Manager Ryan Sarver and Facebook CTO Bret Taylor at Web 2 Expo in New York which took place last month (video embedded below).
At 3:53 into the interview, Dash confronts Sarver and Taylor with a question about the need for standards and the failure of http://activitystrea.ms. Sarver responds first, elaborating on why the idea is too ambitious and Taylor follows up by agreeing with everything Sarver said. In other words, "Facebook" also agrees that a standard for activity streams is too ambitious.
On Twitter, Dash joined the conversation, tweeting that "All those companies "embrace" [http://activitystrea.ms], but there are no apps that use/depend on [it]." Dash goes on to say (in response to my question about the difference between "embrace" and "use") that "Embrace costs nothing and, usually, means nothing." As more information comes to light, it may become fodder for another one of my columns here on InformationWeek. Here's the video:
David Berlind is the chief content officer of TechWeb and editor-in-chief of TechWeb.com. He can be reached at email@example.com and you also can find him on Twitter and other social networks (see the list below).
IT Service Management Must EvolveThe idea of technology being delivered as a service appeals to the 409 IT pros responding to our Service-Oriented IT Survey. But cloud providers are competing for that work, and CIOs are being selective.