I have been thinking recently about the power of social software at work — prompted in part by my post earlier in the week, but also by news that Cogenz, an enterprise social bookmarking tool, is now available as an on-premises version at a strikingly reasonable cost. (This may not be new news, but I only heard of it this week.) I have also been pondering the 800lb gorilla in this room: Sharepoint. After this cogitation, I have come to the view that successful enterprise social software has to put the enterprise last. This is a reversal of the traditional paradigm of business computing.
Since the birth of LEO nearly 58 years ago, computers have been part of business. By and large, their role has been to automate, speed up, replicate, organise, make more efficient, or otherwise affect work activities. That is, their primary impact has been on things that people would not do unless there was a business reason for them to be done. As a by-product in later years, people started to use business-related software to manage domestic or private activities (writing letters, making party invitations or balancing household accounts, for example), but these tended to be peripheral. During this time, if they had a home computer at all, people would expect technology at work to be ahead of what they had at home.
Over the past 5-10 years the balance between personal and business technology has changed completely. Driven by (a) the spread of internet connectivity (especially wi-fi) into the home, (b) the need to support other digital technologies (cameras, music players, gaming devices, for example), and (c) increased functionality and connectivity in small-format devices (mainly mobile phones), it is now frequently the case that people’s domestic technology outstrips that provided to them at work. Alongside this change in the hardware balance of power (and for similar reasons), software has also become much more focused on enhancing the things that people might want to do for themselves, rather than for a salary.
These changes are part and parcel of Web 2.0, social software, social networking — call it what you will. Those tools work because they serve an individual need before they do anything else. A couple of examples by way of illustration.
- Delicious works in the first instance because it helps people store pointers to web pages that they find useful. Because that storage takes place independently of the computer the user sits at, it is ideal for people who access the internet from a variety of locations (home, work, a public library, and so on). Better than that, delicious allows people to start to classify these pointers, or at least tag them with useful aides-memoire. Both of those things — location-independent storage and tagging — mean that delicious is already more useful than the alternative (browser-based bookmarks). The final piece in the jigsaw — sharing of bookmarks — is just the icing on the cake. The social aspect only comes into play once personal needs are satisfied.
- Flickr has a similar dynamic. As digital camera use spreads, people start to need different ways of showing pictures to their friends and families. It is rare that people will print all of their holiday snaps so that they can take them to work and make their colleagues jealous. Instead, they can upload them to the website and share the link. After a while, having uploaded hundreds or thousands of pictures, finding the right ones becomes difficult. But flickr offers the possibility of tagging individual pictures or grouping them in sets. That organisation makes it much easier to show them with the right people. But it also means that other people’s pictures can be discovered because they have used the same tags. Like delicious, the social aspect — sharing, commenting on, and collecting other people’s pictures — comes after the personal.
Unlike the telephone, or e-mail, which depend for their efficacy on network effects, these social tools have value at a non-networked, private, personal level. Unsurprisingly, the early adopters of those first communications technologies were large organisations. If nothing else, they were able to create small network effects internally or between each other. For example, universities were early users of e-mail because it sat well with traditional inter-institution academic collaboration. By contrast, businesses and other organisations have typically lagged behind individuals in the adoption of Web 2.0 tools. (To be clear, individuals at work may well be early users of these tools, but their employers tend to see the light much later.)
As a general rule of thumb then, technologies supporting new types of social interaction tend to be proved by use in a non-commercial context and by providing real personal value ahead of any network effect. Sometimes this doesn’t quite work out. Twitter, for example, provides little personal value without the network effect. However, I think the fact that there is such a low barrier to entry to the twitter network explains that. It also came late to the social party, and so it could piggyback on existing networks. Sometimes the social element doesn’t have a particularly great impact. Many people on flickr do not use the full range of tools (commenting, tagging, etc). I use Librarything primarily as a catalogue of my books, to make sure that I don’t duplicate them. There is a social side to the service, but I haven’t really engaged with it. That does not diminish the utility of the site for me or for anyone else using it.
This week’s McKinsey Quarterly report on “Six ways to make Web 2.0 work” includes a similar point:
2. The best uses come from users—but they require help to scale. In earlier IT campaigns, identifying and prioritizing the applications that would generate the greatest business value was relatively easy. These applications focused primarily on improving the effectiveness and efficiency of known business processes within functional silos (for example, supply-chain-management software to improve coordination across the network). By contrast, our research shows the applications that drive the most value through participatory technologies often aren’t those that management expects.
Efforts go awry when organizations try to dictate their preferred uses of the technologies—a strategy that fits applications designed specifically to improve the performance of known processes—rather than observing what works and then scaling it up. When management chooses the wrong uses, organizations often don’t regroup by switching to applications that might be successful.
In practice, I suspect this means that corporate information is less likely to lead to social interactions (even inside the firewall) than personal content is (such as collections of links, and views expressed in blogs). People are more likely to appreciate the value of other people’s personal content than anonymous material, no matter how relevant the latter is supposed to be to their work. More importantly, when someone appreciates the value of being able to create their own content by using a tool or system provided by their employer, they are more likely to support and promote the use of that tool or system amongst their colleagues. That way success lies.
But what of existing corporate systems? Can they have social elements successfully grafted onto them? This question is most commonly asked of Sharepoint because, as Andrew Gent has put it “Is SharePoint the Lotus Notes of the 21st Century?“. He starts with praise.
The result is a very powerful collaboration, simple document management, and web space management system. It didn’t hurt that V2 of the team collaboration portion of the product (known at the time as Windows SharePoint Services) was “free” for most enterprise Office customers. SharePoint essentially invented a market segment which until that point had been occupied by “integrated” combinations of large and/or complex product sets. Just as Lotus Notes did 20 years ago.
Another similarity is the limitations of the basic architectural design of the product. All products have what could be called a “design center” — a focal point — an ideal business problem that the product tries to solve. The design center defines the core architectural goals of the product. SharePoint’s design center is flexible collaborative functionality centered around light-weight document management and customizable portals.
And the fact is SharePoint’s design center hit a bull’s eye. The need for easy-to-use collaboration spaces and web sites that don’t require web programming — that work well with Microsoft Office and the Microsoft security model — has been a big hit inside corporations. As a salesman for a competing product once told me, his job is not so much selling their own product, but explaining why customers shouldn’t use SharePoint.
But then things get ugly:
SharePoint is designed with flexibility at the space or site level. It allows individuals to take responsibility for managing their own sites and collections of sites. But if — from a corporate or even a divisional level — you want to manage the larger collection, SharePoint becomes resistant — almost belligerent — to control.
The inability to create even simple relationships between lists in different spaces (beyond simple filtered aggregation) without programming is the first sign of strain in SharePoint’s design. Then there are site columns. Site columns let you — ostensibly — define common metadata for multiple lists or libraries. However, you cannot enforce the use of site columns and site columns only work within a single site collection. There is no metadata control across multiple site collections. In other words, simplified control within the sites leads to lack of control at the macro level.
These are all just symptoms of a larger systemic issue: SharePoint is designed around the site. In Version 3 (also know as MOSS 2007) site collections have been introduced to provide some limited amount of cross-site control. But the underlying design principles of SharePoint (ie. user control and customization) work against control at the higher level.
So there is a fundamental reason why Sharepoint will not be able to move from the merely collaborative to the genuinely social. It is driven by the need to support existing business structures and pre-defined designs. Sharepoint uses cannot be emergent, a key feature of Enterprise 2.0 tools (as explained by Andrew McAfee and Rob Salkowitz), because they need to be planned from the outset. In David Weinberger’s terms, the filtering takes place on the way in, not on the way out. As JP Rangaswami suggests, filtering on the way out provides opportunities for more interesting knowledge management.
1. In order to filter on the way in, we need to have filters, filters which can act as anchors and frames and thereby corrupt the flow of information. We’ve learnt a lot about anchors and frames and their effect on predilections and prejudices and decision-making. With David’s first principle, we reduce the risk of this bias entering our classification processes too early.
2. I think it was economist Mihaly Polanyi who talked about things that we know we know, things that we know we don’t know and things that we don’t know we don’t know. Again, filtering on the way in prevents us gathering the things that we don’t know we don’t know.
3. The act of filtering is itself considered necessary to solve a scale problem. We can’t process infinite volumes of things. But maybe now it’s okay to be a digital squirrel, given the trends in the costs of storage. [Sometimes I wonder why we ever delete things, since we can now store snapshots every time something changes. We need never throw away information]. Filtering on the way out becomes something that happens in a natural-selection way, based on people using some element of information, tagging it, collaboratively filtering it.
Thanks to Euan Semple, we do at least know that Microsoft’s heart is in the right place:
…the highlight so far for me of FASTForward ’09 has been getting to know Christian Finn, director for SharePoint product management at Microsoft. Christian is a really nice guy who has been going out of his way to spend time with the bloggers from the FASTForward blog and myself getting his head around the social computing world we all get so excited about.
I am interested to see how this engagement works out for Microsoft and for us. Especially because I think one of the underpinnings of the Microsoft/Apple dichotomy is the two companies’ different approaches to the corporate and the personal. Apple has always been more focused on the personal, while Microsoft concentrated on enterprise needs. This nearly killed Apple in the years when “personal computers” were really more likely to be desktop enterprise systems. Apple has made a comeback on the back of increased personalisation of technology. Can Microsoft work out how that is done?