I’m still in awe of the essence of the Web: connection and collaboration on a previously unimaginable scale. Yet I also feel like these connections waste my time. Not because anything in old media can provide them more effectively, but because the tools that make up the social web are still in a very early stage of evolution and they create a lot of unnecessary waste. This waste is a consequence of the Seven Digital Sins.

The first digital sin: Disorder

The absence of ordering by subject matter

disorder

The Web generates conversations between masses of people all over the world. Yet a long online discussion, whether it takes place on a forum, a mailing list or a comments thread always ends up as a huge mess. For example, take this Guardian article criticizing the value of vitamin supplements. It generated a lively debate in the comments section which highlights the value of the Internet conversations: different people from different perspectives are giving opposing, informed opinions. These give greater insight into the article and ensure the ‘other side’ has an unfiltered voice.

The problem is that, unlike the article itself, the comments have no coherent organisation. There are no dividers or subheadings, no logical progression of arguments or groupings of opinion and no distinction between unique, intelligent insights and throwaway expressions of approval and opposition. It’s painful to work out what’s going on. (And this is a relatively brief conversation! Many forum conversations go on for pages and pages.)

The second digital sin: Clutter

The existence of more posts than necessary

clutter

Online discussions are also cluttered with extraneous information. Any public discussion will usually be littered with many posts making the same basic point. For example, when Google unveiled Wave last year, tech blog comments were filled with criticisms that the interface looked too busy and complex to engage ordinary Web users. This is a worthwhile, interesting point, but there’s no value in hearing it 30 times in the same thread.

The other main cause of clutter is misunderstandings. For example, in the pictured forum, you can see ‘shyuhe’ asks a question to which ‘Abaddon’ gives an answer. Shyuhe points out that Abaddon has misunderstood his question, and creates a new post which poses the question in a different way, prompting a new reply by Abaddon. The exchange goes on in this manner for several more posts. The reader ends up going through a series of posts that are just clarifications of previous posts. In a wiki article, the original text is continually revised until it’s clear. It’s another reason why it’s so much more time-consuming and painful to absorb information from a conversation than from an article.

The third digital sin: Reinventing the Wheel

Failure to build on past discussions

Because big online discussions are so messy and cluttered, very few people have the patience to read through the entire discussion before writing a post themselves. The discussion tends to get focused entirely on the last few posts. Older posts are forgotten about. This of course means that the same conversation happens again and again within the same thread. The newcomers to the conversation don’t notice that the points they want to make have already been made. Even discussion participants who have been following it from the very start struggle to retain an understanding of everything that’s been said because there’s no way to quickly get an overview.

The Internet should be a place where people can access all the knowledge and ideas surrounding a particular subject and then say something which builds on that. In other words, discussions should progress. Instead, they resemble Groundhog Day.

The Internet should be a place where people can access all the knowledge and ideas surrounding a particular subject and then say something which builds on that. In other words, discussions should progress. Instead, they resemble Groundhog Day. As soon as a conversation has reached a certain length it is submerged into the oblivion of irredeemable messiness. Another conversation duly arises to make all the same points.

The fourth digital sin: Inconsistency

Too many competing formats within the same collaboration suite

The first three sins are most obvious on forums, comment threads and email mailing lists. But they’re also a problem on dedicated collaboration tools like Basecamp and Huddle . These tools are characterized by the stitching together different tools: typically wikis (or ‘writeboards’), forums (or ‘message boards’) and comments. Unfortunately this only adds to the general messiness of online collaboration:

Firstly, it’s often unclear whether ideas should be posted as wiki pages or forum threads. A wiki page encourages organization but a forum encourages better brainstorming (since it’s more obviously participative and conversation-like). You could start with a forum discussion and then summarize the results as a wiki page, but that splits the topic in two, making it unclear where people should make further contributions. Moreover, the act of summarizing a conversation is itself a time-consuming task.

Secondly, it’s difficult to create a coherent interface for managing all these different tools. Interaction design thrives on consistency. The more ways there are to contribute to the site and engage with other users, the more difficult it is to create a consistent interface.

These problems are exacerbated when additional collaboration formats are thrown into the mix. Blogs, social bookmarks, microblogs and Q&As all have their own unique advantages. But throw them all together and you have so many competing formats that it becomes very hard to build a clear mental picture of how things are organized and how, as a user, you should engage with the site.

The fifth digital sin: Automated miscommunication

Too little or too much information about what’s happening within the collaboration suite

Auto miscommunication

The fourth sin contributes to this fifth sin. Vital to any social media application is the ability to quickly get an overview of relevant updates and new content. How do users know when a wiki article relevant to them has been updated? How do they know if someone has replied to a comment they’ve written? How do they know if there’s some new question or idea they should be responding to? Answering these questions satisfactorily becomes much more difficult when there are half a dozen different formats to keep track of. You can’t have a simple email-style inbox if each tool has its own interface and performs a radically different function.

Some collaboration tools—Wetpaint, for example—notify users of practically everything that happens on a site. This is overwhelming, and most users will respond by ignoring all notifications. Other collaboration tools go too far the other way, minimizing notifications with the consequence that some important changes have to be hunted down on the user’s own initiative. Google Docs is a surprising example of this: there is no fast, intuitive way to see exactly what new additions a collaborator has made to a document. Showing too little and showing too much both waste users’ time and make it harder to effectively coordinate your work with others.

The sixth digital sin: Aimlessness

Discussions that run off-track and waste time

The Internet has inherited this sixth sin from the offline world. We’ve all attended meetings that meander from topic to topic or get sidetracked by inconsequential details. Online discussions are no better. They should have the advantage of allowing readers to simply ignore the inconsequential parts, but lack of structure makes this difficult in practice: you need to read from top to bottom to understand what’s going on.

The simplistic comment-followed-by-reply-followed-by-reply format of today’s online discussions has the advantage of being easy to use, since it imitates how we have discussions offline. But it misses the opportunity to create a new, digital-native discussion format in which comments are not organized chronologically (or according to the person you’re responding to, as is the case with ‘threaded’ forums) but according to the specific point being discussed. Such a format would lead to claims being broken down into their component parts, leading to much clearer thinking.

We’re still relying on the mental frameworks of the old, offline world of communication. The Internet doesn’t need to be structured this way; and it shouldn’t be if we want to take full advantage of its potential for facilitating mass collaboration.

We’re still relying on the mental frameworks of the old, offline world of communication. The Internet doesn’t need to be structured this way; and it shouldn’t be if we want to take full advantage of its potential for facilitating mass collaboration.

The seventh digital sin: Incivility

Personal attacks which don’t make any constructive point

Two words: YouTube comments. How many people have been put off online discussions because of the sniping, personal attacks, pointless insults and general needless rudeness that is so common on public forums? The root of this excess is twofold. Firstly, it’s easy to over-estimate people’s hostility when you can’t see their face and so over-reactions are common. Secondly, online commentators can choose to be anonymous, and anonymous people don’t suffer any consequences for bad behavior. This has led some to demand the end of online anonymity.

Such a drastic solution would be a shame because online anonymity has a hugely positive flip side. There are many people too shy, oppressed or otherwise restricted by the duties and bonds of real life to honestly speak their mind if that involves revealing their offline identity. The anonymous Web has provided such people with a voice. Allowing secrecy of identity creates true openness of debate.

An alternative solution to incivility may lie, once again, in the discussion formats themselves. A future format might serve to undermine the one-on-one personal bickering that existing formats make so easy.

Conclusion: Sins old and new

Pride, avarice, envy, wrath, lust, gluttony and sloth: The Seven Deadly Sins are as old as humanity. Whether we blame brutal evolution or rebellious angels, they’re probably here to stay. The future of the Seven Digital Sins however, lies in our hands. They’re embedded in tools over which we have practically unlimited control. The impact that Twitter, an app whose revolutionary innovation was simply to impose a character limit on blog posts, has shown that shaping online interaction is now more an issue of design than technology. Our progress towards genuine mass collaboration is limited only by our inability to think outside the offline mental box.

The defining characteristic of social media is a revolutionary undermining of the distinction between producers and consumers of media. Instead of producing content, social media services merely facilitate user interaction. Given this, one would have thought that the tenets of User-Centred Design (UCD) would be highly pertinent to the design of social media. In fact it is very difficult to find a major social media service which was created via a genuine UCD process (almost as difficult as finding an interaction designer who does not wax lyrical about the importance of UCD). Blogger, craigslist, Delicious, eBay, Facebook, Flickr, Gaia Online, Last.fm, LiveJournal, Skype, Wikipedia: none of these pioneering legends of social media were born out of the ethnographic observation techniques, ‘personas’, ‘scenarios’, low-fidelity prototyping or constant pre-development testing that characterise UCD. Why don’t UCD techniques seem to be necessary for the creation of great social media? I propose five main reasons, not all of which are peculiar to social media.

1) Self-centred design is a valid approach
The history of social media is full of examples of designers designing for themselves and then finding that their designs are immensely popular with others. When Delicious founder Joshua Shachter solved his own information management problem, millions of others were waiting for someone to solve the same problem. When Blogger’s Meg Hourihan and Evan Williams and LiveJournal’s Brad Fitzpatrick wanted an easier way to update people about what they were doing, millions of others wanted the same thing. Because these designers were designing for their own requirements, they didn’t need to research user requirements to develop a great concept.

2) Personal experience can be an adequate substitute for user research
In those cases where a social media service was not created to serve the founder’s own requirements it can still be said that the founders had a good understanding of the user requirements, not through UCD research techniques, but through their own experience. In the cases of Facebook and Last.fm, the designers had a good understanding of their target users—students and music fans—because of the milieu in which they worked and socialised.

In the cases of EBay, Flickr, Skype and Wikipedia the focus is on activities which are common to virtually all web-users: buying and selling, photo-sharing, making phone calls and looking up articles in encyclopaedias. These activities are so well-understood and nearly-universal in the developed world that it simply wasn’t necessary for the founders to know any more than they already did through their own everyday experiences.

3) The nature of social media discourages early prototyping
One thing that is striking in the stories of social media entrepreneurs is the credit which most of them give to their users for shaping the design of the service. Social media makes it very easy for users to give feedback about the service and for their digital imprint to be monitored. This may not be as valuable as the face-to-face observation advocated by UCD, but it is much more convenient.

Moreover, it is relatively easy to redevelop social media services. UCD grew out of traditional 3D product design, where there is an immense gulf—financial and technical—between a prototype and a working product released to the public. In contrast, the Web makes the distinction between an online prototype and a finished product almost arbitrary. But while post-development prototyping is easier with social media, pre-development prototyping is actually harder because the social interactions of a community of users cannot be simulated. Thus the nature of social media tends to encourage the suspension of pre-development prototyping in the knowledge that the most valuable results issue from post-development prototypes.

4) An emphasis on simplicity offsets the need for User-Centred Design
Basecamp stands out amongst social media success stories because of its dismissive attitude to user feedback. Its nonetheless immense popularity can be partly explained by its founders’ belief ‘that there is beauty and wisdom in Web-hosted, bite-size software built to accomplish narrow tasks’. Skype cofounder Niklas Zennström shares these sentiments: ‘Skype is easy enough to use so that people don’t need to be tech savvy… If you can use a Web browser, you can use Skype’.

Anything which conscientiously focuses on providing only the broadest and most essential features in a simple, easy to grasp interface is very likely to be highly usable. It seems that an emphasis on simplicity and an awareness of usability principles can be adequate substitutes for guaranteeing usability through usability testing.

5) Implementing a User-Centred Design process is often not worth the required time and expense
None of this is to say that UCD is actually detrimental to good design. It is difficult to imagine how UCD techniques could do anything but improve design. Unfortunately, good design is not the only factor contributing to the success of a product: expenses and speed are both vital factors and both can sometimes be adversely affected by UCD processes.

A disclaimer
Of course not all social media services can be designed in a simple and straightforward form that makes usability a given; some activities are necessarily complex. Likewise, it cannot be imagined that for every potentially successful social media service there is an available designer with sufficient relevant personal experience to be able to design it without user research. The rise of niche social networking and enterprise 2.0 is likely to strengthen the need for UCD research techniques in the design of social media. Niche groups and businesses are going to require specialised, potentially complex, social media services which cannot be created by designers relying on personal experience and an emphasis on simplicity.

Finally, it is important to note that UCD is not an ‘all or nothing’ approach. The User-Centred Design process can be broken down and simplified so that only the aspects which are most needed are used. Some form of pre-development usability testing of an interface is almost always worth implementing.

With these disclaimers it is possible to conclude that while cheaper and nimbler alternatives render UCD generally unsuitable for the design of social media, UCD should never be rejected entirely and looks set to become less generally unsuitable as a result of future trends.

N.B. This is a summary of a much larger article User-centred design and the user-driven web: Is a user-centred design process suitable for social interaction design?

Top photo by: DerrickT