French company, Atos, cause a stir recently by repeating again its intention to ban internal email from 2014 that was announced earlier in the year. There has been a fair bit of misunderstanding on the Web about this so I suggest you read this BBC interview with ATOS CEO, Thierry Breton, that explains his thinking behind this strategy.

Most of the critical responses to this idea have expressed an incredulous attitude towards the idea of eliminating what most people consider to be a critical business tool (remember the reaction to the BlackBerry outage a few months ago?). In some respects, many of the arguments against banning email are reasonable:

• Can you ban email if your customers and clients are using it? (to be fair to ATOS, they aren't planning on banning external email).
• Rather than banning email, users just need to manage their inboxes better.

However, on balance I think there is good reason for aiming to effectively ban email. But rather than outlawing it, we need to reinvent how we utilise email as a software protocol and also the ageing paradigm of the inbox, particularly where the assumed effectiveness is built on false assumptions around utility and information ownership. At a software level, email offers a number of important features - such as:

• Interoperability and extendability.
• It can work offline (although this is becoming less important).
• For the sender, it costs no more to send messages to 1 or many people where ever they are.
• Both sender and receivers can store and organise copies of messages exchanged independently of each other.
• Email addresses act a simple proxies for identification.
• Email accounts can be created for individuals, groups and also non-human systems.

These features provide a great deal of utility, although we can look at each feature and find many negatives too - for example:

• Email standards and extensions aren't implemented homogeneously, so users may have problems reading or processing an email.
• When you go online after an extended absence, your inbox is flooded with new messages.
• People send many messages that for the receiver are just transient or ambient information - but your inbox treats them as all the same.
• The independent nature of email messages contributed to fragmentation of the information chain, making it hard to know who knows what and people who should know made end up getting left out of the loop.
• An email address doesn't actually tell you anything about the user, who they are or why you should trust that identity.
• High volumes of automatically generated notification emails from non-human systems contribute to information overload.

In summary, we can say that email works as a pragmatic solution, but not without creating numerous problems for individual users and organisations a like. As a result there many solutions out there that help us to deal with everything from email processing to email data management. Some solutions are technical in nature, like help desk ticketing software or records management systems, but others focus on the user as problem and attempt to fix individual behaviours. But ultimately none offer a way of making email the perfect tool and it will take a leap to improve how we communicate and collaborate.

So, what is stopping us taking this leap? Thinking about this from a social experience design perspective, I think there are four key issues that need to be addressed to create something better than the email we use today:

1. Move to open work as the default. Email provides users with a simple system of directing messages at people - this falls into a mental information sharing model of open only by exception that is the default in most organisations, but it is also supported by a false perception that email is owned by the sender and private if we restrict the names included on the To, CC and BCC lists. Of course, this doesn't mean we stop supporting some private communication entirely!
2. Everything is Miscellaneous. Centrally designed information systems that enforce fixed, common models for organising information and work process don't work. These were designed with good intentions, but they aren't effective and only encourage the use of personal information stores.
3. Collaborate by staying apart. We need the same ease of interoperability between different social business software platforms that email offers - I shouldn't be forced to use your system, when I have my own, and neither should you.
4. Who are you? We need to shift from email addresses as identifiers towards a model where organisations can offer a better user identifier and profile that will enable messaging to take place through the right channel or system.

Its entirely possible that email protocols will continue to play a role in this new environment, but I think it will also depend on other Web 2.0 protocols like Activity Streams, Open Social, and ATOM. This actually hints that the death of email will come incrementally, if we wait for the technology to rise up and present better alternatives. In practice, the tide of data created by other social business software tools will make the traditional email inbox an unsustainable proposition.

The water is rising slowly right now, but don't doubt that the inbox will need to be reinvented at some point - the question is really when and how, not if.

BTW if you found this interesting, you might enjoy this presentation, Architected for Collaboration.

Image credit: Inbox Art CC BY-SA

The Department of Education & Communities in NSW has published a range of materials during 2011 addressing various aspects of social media and how people working in this department can and should make use of it. Above is a video introduction to their internal microblogging tools, Maang.

Their social media policy has links to more resources, including an An introduction to Digital Citizenship for the workplace.

Lots of people scratch their heads about the application of social media to actually doing work, particularly with the focus on social support for people in the workplace. Some commentators have highlighted the need to integrate enterprise social software capabilities directly into employees' workflow (and ironically some are quite evangelical about it, but I digress).

I've noticed a flurry of new task-orientated tools appearing, including:

• Do.com from Salesforce (built around tight integration with Google's productivity tools).
• Strides from Socialcast.
• Asana.

(I'm sure there are more out there that I've missed)

This one isn't new, but in the same vein NationalField also offers some interesting activity rollup features to help managers.

Also, many of the social suites also offer lightweight task and project management - e.g. Jive SBS, IBM Connections and Atlassian Confluence (if you want heavy weight task management, check out Jira).

Of course, task management isn't new and won't suit every workplace culture or workflow. But these new plays in social task management may be able to achieve a beach head into organisations that have resisted enterprise microblogging tools like Yammer but by following the same basic user-driven adoption approach that can sneak over the firewall.

Would you rather be Doing and Striding than Yammer'ing? ;-)

Image credit: Task List CC BY-NC-SA

Microsoft researcher, Jonathan Grudin, wrote a paper back in 1994 that looked at eight challenges for groupware [PDF] - they were: 

1. Disparity in work and benefit.
2. Critical mass and Prisoner’s dilemma problems.
3. Disruption of social processes.
4. Exception handling.
5. Unobtrusive accessibility.
6. Difficulty of evaluation.
7. Failure of intuition.
8. Adoption process.

Its interesting to consider what the current generation of groupware, what today we call Social Business Software or "Enterprise 2.0", has done to address these problems.

I think challenges 1, 4, 5 and 7 are definitely areas where we have seen improvement - primarily through the benefits of infusing the concepts of social software into the design of the technology solutions we want to use. All sorts of design patterns have come to the fore in recent years to make collaboration software more human-centred than ever before. Plus the accidental training ground of the World Wide Web means that more and more users are ready (if not demanding) a consumer experience inside the firewall.

However, I sense that right now we are looping back into old territory by considering again the importance of embedding the technologies into workflow and most of these challenges can only improve through iteration (sorry everyone, the goal of integrating collaboration technologies into the places and tools where people are actually working isn't a new one).

But the Difficulty of evaluation and Adoption process in particular are likely to remain major challenges (Critical mass and Prisoner’s dilemma problems and Disruption of social processes are also closely associated with both). The evaluation challenge might be dealt with eventually through better analytics (Dachis Group's Social Business Index is already starting to do this), but ultimately these challenges aren't technology based:

• Organisations need to have a design mindset, in order to both implement and judge the success of such technologies; and
• We already have tried and tested methods available to support their use, we just need to use them!

If we can assume the software will continue to get better and better, moving forward why can't we focus on getting the last few points right instead?