28 November 2008

Why and When a Wiki Won't Work Weally Well!

Wiki's have been a success for several years now in the public space, using the wisdom of crowds and collaborative knowledge creation.

For the last few years wikis have also been making their way into the enterprise space as well. Solutions like Socialtext, Atlassian's Confluence, MindTouch Deki and Traction TeamPage are now being used by many different sized business and government departments.

The phenomenon is now attracting researchers, interested in understanding how wikis are being used and adapted in this area.
  • My own research is looking at what sort of corporate cultures assist in wiki use & growth.
  • A german team of researchers at the University of Bamberg are looking into organisation wiki use from a SNA point of view using Social Systems Theory.
  • Researchers Janus Boye and Dorthe R. Jesperse published a report earlier this year showing that wikis are taking off in the corporate world, but care should be taken in the areas of culture and structure to avoid some of the pitfalls of an open system when used for content management.
Many commentators are also discussing how they can be improved upon, either through growing the technology itself, or by better implementation, as Stewart Mader's site often covers.

Simon McDonald posted a nice article summarizing his negative (and positive) experiences with wikis on evolt.org.

Today I found a blog post by the "Wikis in Organisations" team talking about what mistakes to avoid when implementing a wiki. Great advice and follows the idea of The Mistake Bank I came across a little while back.

All of this fits with Dave Snowden's constant reminder that we learn more from our mistakes than our successes. I am starting to agree.

24 November 2008

Weaving a model for E2.0 adoption

Given the Greek heritage of thought in Western Culture, it's not surprising that we tend to struggle with managing the intangibles of business life. It's not that we can't handle them. We deal with intangibles in other parts of our life, relationships, even the quality of the wine with last night's meal.

But in the business world, where our actions need to be justified, usually with a dollar figure, it is simply easier to do. It is easier for us to have nice, orderly, process-driven business lives. There's just one problem. Businesses are full of people. Social people, imperfect people, obnoxious people, human people. People that live lives full of intangibles, and we have seen many stupendous failures and staggering success stories where intuition and social pressures played a far larger part than the logic of the situation.

Enterprise 2.0 technologies are designed to take advantage of the social nature of the people who work in our businesses. It allows them to connect, interact, and exploit the knowledge within (and sometimes without) the business in a more effective ways than could have been imagined by the promoters of scientific management 50-odd years ago at the boom of the industrial era.

We still live with the vestiges of the industrial era and in some sectors many of the lessons learnt are still relevant and valuable. However. As Drucker pointed out:
“Every knowledge worker in modern organization is an "executive" if, by virtue of his position or knowledge, he is responsible for a contribution that materially affects the capacity of the organization to perform and to obtain results.”
Source: The Effective Executive - 1966
So when implementing E2.0 solutions, the intangibles need to be taken into account. Not just in the initial justification for expenditure, but more importantly in the selection and implementation of these tools.

There are several people talking about this at the moment.
I do a lot of reading and have thoughts like those above in my head as I review my research data. That is how qualitative research works. It is immersive. Unlike sterile surveys summarised by countless statistical analysis, exploratory social research involves getting involved in the complex interplays between people and teasing out the cultural factors that are impacting their success or failure.

Matt's work on the two-domain model has been exiciting to watch and encourages me to finish my thesis so I can enjoy putting it to use by building the implementation programs and principles that Stephen Collins says are so necessary. Matt initially moved away from my focus of "Participation" and it's two forms, Mandated and Spontaneous, however in his last post he used instead the term "Interaction". I like this term. It speaks at a higher level than participation, and also includes the concept of reciprical involvement between human and technology that my data seems to indicate is going on.

Patrick suggested last night that I consider instead a layered model where the intangible and tangible domains overlap the same space. This is a nice idea and I am already thinking about how I can visualise that to better display the interactions in the data.

Finally, I discovered a new idea while writing about how the wikis were used in SMEs yesterday. I share it here in the hope I will get some feedback, either critical or helpful, I don't mind. It's a new thought for me and I want to sound it out here before expanding on it in the thesis in the hope it will be more rigorous for the exposure.

The quote looks at another possibility when implementing E2.0 tools. Often we hear the catch-cry: match the tools to the culture of the business. My question: is this entirely correct?
It should be noted that many of the uses were only successful within individual divisions of the companies, and suited the cultural constraints of those divisions. This suggests that it may be the business use itself that needs to be aligned with the local group’s culture, rather than the tool.
Please post your replies written on the back of $100 notes! I'm going to have to do something pretty special for my poor wife after all this to convince her I'm not just a part of the office furniture :-)

17 November 2008

The beginnings of a theory of participation

After several days of being immersed in my research data I am starting to see patterns appearing around the use of wikis in Small to Medium Enterprises.

One of the repeated findings in my study was the user's awareness that the wiki was only going to be valuable if everybody was participating. They spoke of an awareness that a solo effort was a waste of time and effort.

Below is my first try at mapping the interactions between these concepts and uses the actual labels used by the participants in the study. There is minimal theory in this framework. Each link is based on evidence in the data. I share it because I am interested in feedback, both from practitioners and academics. Not just on the model itself, but also if people have seen similar frameworks published. I'm sure I cannot be the only one seeing these interactions in the wild?


The key for my study seems to hinge around group participation in the wiki. This participation generates both tangible and intangible benefits which can boost trust and encourage further spontaneous participation which was the hallmark of all the successful wiki implementations in the study.

Many people currently talking about Enterprise 2.0, including Andrew McAfee himself, mention the process of evolution where both the functions and even the tools themselves are modified or replaced over time to best suit the needs of the business. Often the idea of success or failure is based on the notion of participation. For example, Safe-Fail can be useful to detect the success or failure of systems based on the concept that participation in these systems is an emergent behaviour of a complex system, so it is unreasonable to be able to predict beforehand which tools will gain participation simply by comparing software features with business requirements.

I am interested in three areas surrounding this:
  1. What drives an employee to use or not use a tool?
  2. What part does the culture of the organisation play in this process?
  3. What can be done to kick start a successful adoption?
I hope you will be kind enough to take a few minutes to consider these questions, the model and how these factors might be improved based on your theories.

06 November 2008

Having a sticky in your wiki!

A few people lately have asked about my Master's research project and I realised that while I talk about it a lot I haven't really posted what I am up to, so for anyone who is interested, I would summarise it this way:
The corporate Wiki seems to be highly suitable given the fluid nature of distributed cognition within the Small to Medium Enterprise environment. The limited and usually grass-roots adoption of Wikis in this sector suggests that the organisation's culture effects why and how they are implemented due to it's strong influence on Knowledge sharing in general.

To prove this hypothesis, my research will study Wiki use in SMEs and build a model of what cultural factors play a part in their uptake and success.
After researching 5 companies in Melborne and Sydney, the thesis is coming together now and I hope to make it available to interested parties early in the new year. I am also looking at publishing it on the Digital Thesis Database or the ProQuest Dissertations & Theses register if you have access.

02 November 2008

Enterprise 2.0 & it's effect on Organisational Culture

Back in 2007, Tom Davenport posted a critical article on Andrew McAffee's Enterprise 2.0 idea. He put forward the view that while Prof. McAfee is a nice guy, and his ideas are very interesting, they were "not going to become the next big thing", and certainly "won't make organizational hierarchy and politics go away".

In commenting on the article, David Weinberger wrote the following which I think touches on my own thoughts of the article quite nicely:
Tom, technology has never done anything on its own. So, I agree: the tools themselves won't bring about Enterprise 2.0. Tools by themselves can't even assemble a cabinet from Ikea. (On the other hand, can anything assemble a cabinet from Ikea?)
But, our tools when taken up do have an effect on how we interact. Some of the effects are direct -- e.g., wikis enable a type of asynchronous collaborative, negotiated writing not exactly like anything before them -- and some of the effects will undoubtedly be indirect. E.g., once we've used wikis productively, perhaps our attitude toward the nature of authority will change a little. Maybe not; the effects of technology on attitudes and expectations are hard to observe much less predict.
But it's reasonable to think that the technology, when taken up and used, will affect enterprises directly and indirectly...and (I suspect) in the direction E2.0 adumbrates.
Or not.
This is the crux. Most people who are either hyping Enterprise 2.0 or criticising it seem to be taking extreme positions. Not only that, they are taking quite black and white definitions of the issue of culture and Enterprise 2.0. Often they see this aspect as a chicken and egg dilemma and think that because certain cultural factors are necessary for Enterprise 2.0 tools to be successfully implemented (the egg comes from the chicken) that these tools then cannot modify the culture of the organisation further once implemented (so therefore chickens cannot possibly come from eggs).

My friend Matthew Hodgson from SMS, who I have quickly grown to respect, recently blogged about the above article and a response to it regarding Enterprise 2.0 and culture change. While I am sometimes scared by how similar Matt and my views are, I found some of his points hard to integrate with the recent views of culture. Hannerz, in discussing the term Cultural Complexity says:
"The term 'complex' may in itself be about as intellectually attractive as the word 'messy,' but one of its virtues in this context is precisely its sober insistence that we should think twice before accepting any simple characterization of the cultures in question in terms of a single essence."
The definition of culture I am using for my thesis is based upon Strauss & Quinn from Duke University:
"Culture … consists of regular occurrences in the humanly created world, in the schemas people share as a result of these, and in the interactions between these schemas and this world. When we speak of culture, then, we do so only to summarize such regularities"
As Matt points out, "Tools don't change people, people change people". But just like stone axes, bronze swords, the steamship, coal & oil technologies and the telephone before them, tools effect the way people can manipulate their environment AND interact with one another, and in the area of communications tools, culture is effected even more dramatically. My family's relaxation culture was changed irreversibly with the purchase of a television with a built-in hard-drive that allows us to pause live TV. Kids now come to dinner mid-TV shows, Movies can be paused to sooth a child's nightmare. Stress levels have never been so low!

Responding to Tom Davenport's article, Dion Hinchcliffe wrote a ZDNet article suggesting that Enterprise 2.0 tools weren't the cause of change, but rather a catalyst to speed up change via the cultural modifications I just touched on, but also that the fundamentally social nature and ease of use of these tools makes them their own change agent, making the users their own change advocates which helps overcome many of the cultural barriers that previous technologies have run in to.

I am not sure I agree with everything he had to say and he admits to being biased, however I think he is on the right track. Matt wasn't quite so enthused though, and in his blog post this week, he attacks the article from the viewpoint of Organisation Psychology by pointing out that based on a study by three HCI academics, Culture effects wiki use, but wiki use does not effect culture. I have several concerns with Matts summary and the study it was based upon, so please forgive me Matt as I dissect both in an attempt to get my thoughts straight. I hope it comes across as criticism, not cynicism.

1. The wikipedia study consists of a statistical analysis searching for correlations between Hofstede's cultural metrics (Power Distance, Collectivism versus Individualism, Femininity versus Masculinity, and Uncertainty Avoidance) and certain types of editing actions on Wikipedia pages in several languages.
A) Hofstede himself admits that these metrics are to be used at national levels and can not be transferred to smaller entities such as corporations,
B) He also admits that that should only be used in relation to one another. In other words one country against another country, not one member of a race versus a member of another race,
C) Misuse of Hofstede's work often leads to ecological fallacies where individuals are tarred with the same brush as the average of the group to which they belong. This happens in the study where they suggest that managers should take into account that all French workers "are likely to feel uncomfortable about deleting others' work. It is therefore advisable not to expect or require it of them in collaborative online work." This sort of statement from senior lecturers and PhD students is dissapointing. The fact it made it past peer review, even more so.

2. The paper only declares that culture does effect wiki use (just like previous studies have shown it does on WWW use). It does not follow that the opposite is untrue and this is a one way street.

3. Matt goes on to summarise five types of corporate culture and how they effect information behaviour. Here Matt shows his hand, assuming two things that I believe the Neural Cognition literature does not agree with, but are often seen from the psychological view of culture:
A) he presents the culture as an object in it's own right rather than a collective term for the complex overlapping neural schemas held individually by each member of the group,
B) he sees this culture as the essence of the group's motivation to act and the difference between it and groups with different cultures.
In plain english what I am saying is that everybody is partially the same and partially individual. We are all dissenters at some level. Smaller sections of the group may not exhibit the same "culture" as the main group, and certainly individuals may dissent entirely and often for quite small and personal reasons, such as the sudden ability to have a say in corporate policy.
4. The study itself does not use a longitudinal methodology in order to ascertain if any changes to the culture took place before and after the use of Wikipedia, in fact the study didn't even identify individual editors, but grouped them buy the language they were speaking (to their credit they mention this in their limitations section).

5. And finally, Hofstede's metrics were measured in the 60's and 70's within the IBM corporation's offices around the world. This present several shortcomings:
A) It could easily be argued that IBM's national staff do not properly represent the countries cultures as a whole (and many have argued this),
B) Even if you did try to measure that change, these are national averages and Hofstede's study would have to be conducted all over again to determine a shift.

So, can a tool change a culture?
*YES!* I believe so. They do all the time and have done for centuries.

Can a tool be implemented in a culture if there is a mismatch between the tool and the needs of that culture?
Not likely and if it did, it would be an uphill battle. However, I have evidence of wikis working in divisions, but not successfully transferring up to corporate-wide use. This data suggests that an entire enterprise doesn't need to be exhibiting the flat-wide, open frameworks that Matt speaks of (or the Search, Links, Authoring, Tags, Extensions and Signals (SLATES) that James Dellow quoted of Andrew McAfee recently) for individual groups to be evolving toward this style of operation and therefore being able to successfully implement these tools.

Does this mean these tools cannot change a culture once implemented?
NO, this simply does not follow. Not from Matts logic and not from the study he quotes. In fact, both my research and personal corporate experience have revealed changes in the way businesses operate when knowledge workers encounter new communications possibilities through the use of Enterprise 2.0 tools.

Does that mean Enterprise 2.0 tools are a new breed of tools that will tap into a deep human need, transforming business and corporate structures around the world like we have never seen before?
Of course not! Look I realise I'm pushing against the hype-cycle here, I believe these technologies allow a cultural change that we have not seen before through - amongst other things - the introduction of community into our communications tools which until now have been largely didactic in nature, but on this point Tom Davenport is probably spot-on when he says,
"They won't make the ideas of the front-line worker in corporations as influential as those of the CEO. Most of the barriers that prevent knowledge from flowing freely in organizations – power differentials, lack of trust, missing incentives, unsupportive cultures, and the general busyness of employees today – won't be addressed or substantially changed by technology alone".
But please hear me! That doesn't mean they won't be changed at all. These technologies are simply enablers - powerful ones because they tap in to the cultural norms that in the past were usually experienced socially within family, recreational and religious groups. They don't "cause" organisational cultures to change any more than radio technologies caused war fleets to become more coordinated, but they do act as a catalyst to change. Radio didn't just bring over-the-horizon coordination, extending the distance of the semaphore flags. Once the enabler was there, it's convenience and instantaneous nature caused a shift in the command structure of naval fleets which flowed on to Naval strategy and the ways governments could wield that power. It is in these new uses of the technology that the pro-hype people seem to be basing it's success and more-critical people like James seem to be searching for. However, I hate to be boring, but in most cases it will be the little cultural and communications issues being overcome that will probably be the biggest benefit; and for that we simply have to implement them before we will see if solutions emerge. And that is what the Evolutionary aspect of Enterprise 2.0 is all about.

So in conclusion, Matt I will let you modify the "if you build it, they will come" statement if you let me modify yours from "it’s only people who change people’s behaviour" to "it’s only people AND THE WAY THEY INTERACT THAT changeS people’s behaviour".