Showing posts with label Knowledge Management. Show all posts
Showing posts with label Knowledge Management. Show all posts

Saturday, September 6, 2025

Uh, Claude, Whose Code is it Anyway?

 

 I've messing around with Claude AI for generating programming code. It has been quite useful. I wasn't looking for complete programs — just looking for examples of certain functions I wanted to use. However, Claude generates fully functional programs as part of its output. So what it generates can be run as is. In my example, when looking for Javascript code to search and display lunr.js index entries, it included all of the Javascript, HTML, and CSS styling necessary to run a complete web page.

 Which got me to thinking. Clause does not actually "write" any code, it composes code that looks like code it has seen before. So, whose code is it really? I mean, I'm not going to, but if I did take code generated by Claude and publish it as part of a product, what is to stop someone in a year or two coming and saying "hey, that code at line such-and-such is from my product which is copyrighted.  I'm going to sue." 

Now my guess is people who use Claude — and Anthropic, the people who produce it — assume the code Claude consumed in building its language model is chopped up and mutated so much there will be no recognizable "chunks" left. But what if they are wrong?  And, in particular, what about all the very repetitive parts, like CSS styling, class naming, and the like? That code is likely to be the same on every example. What distant website does that come from and how recognizable is it?

Just as a quick example, the early images generated by ChatGPT often had a weird gray bar across the bottom right, which was clearly a ghost of the Getty Images identifier. One assumes Getty Images would not look kindly on my trying to publish and copyright such an image.

 Even if the code is recognizable, Anthropic under its terms and conditions says they make no claim of "non-infringement" for the resulting output. In other words, you're on your own, legally speaking, if you use the code they generate.

 Which led to another thought: even if there is no way to identify the original code in today's output, doesn't this open up a potentially lucrative avenue for unscrupulous individuals to copyright what Claude generates today and then sue future users when they replicate the same or similar code?  What if I ask Claude to generate code for common programming tasks — creating user accounts or validating phone numbers. Then use that code in a trivial application that I copyright. Two years from now I go searching for other programmers using Claude to do the same task — and probably generating the same or similar code — then sue them for copyright infringement. There would seem to be no way to prove Claude didn't borrow the code from me over the past two years...

 Not pretty. Perhaps not practical or likely to stand up in court. Everyone is talking about the whole question of whose code... or picture... or novel... it is that AI generates. But no one seems to really expect an answer, as if it was an insolvable puzzle. And in the meantime, a dark cloud of legal risk hangs over, not the technology, but the users. At some point in the not too distant future, I expect the legal teams at larger corporations to start cracking down on the use of AI coding tools as yet another Trojan horse for legal liability. 

Sunday, March 8, 2015

I Did Something Stupid (Wearable UX)

I did something stupid today. I disagreed with one of the leading experts in interface design, mobile, and specifically the soon-to-be-released Apple Watch.

The argument is over a chart Luke Wroblewski (@lukew) posted about the distraction factor of a smart phone vs. a smart watch, indicating use of a smart watch lets you be "more connected to the people around you..."

There are two problems here:
  • First, the chart is fiction. It measures "real life" vs. "lost in phone" — concepts of psychological state which are hard to define, never mind measure.  Perhaps there is some real data behind the charts, but if so the labels do it no justice. I suspect the diagram is more a representation of belief or expectation than fact. In which, why is it a chart?
  • Second, it is not the phone that this the point of distraction. It is the information on the phone — phone, watch, iPad, eye piece, etc — and not the device itself that is the distraction. So putting the notifications on the wrist may reduce the time it takes to switch from in-the-present to in-the-data, but it does not alter the distraction factor that data provides, whether it is in my pocket or on my wrist.
One of the most disturbing studies in this area is recent research into the use of "hands free" devices while driving.  I would equate Luke's proposal that data-on-the-wrist is  less distracting from "real life" to the arguments that hands-free phone technology is less distracting to driving. However, the National Safety Council published a report arguing that hands-free does not significantly reduce the risks to distracted driving.

"Hands-free devices often are seen as a solution to the risks of driver distraction because they help eliminate two obvious risks – visual, looking away from the road and manual, removing your hands off of the steering wheel. However, a third type of distraction can occur when using cell phones while driving – cognitive, taking your mind off the road. 
The amount of exposure to each risk is key. Crashes are a function of the severity of each risk and how often the risk occurs. Most people can recognize when they are visually or mechanically distracted and seek to disengage from these activities as quickly as possible. However, people typically do not realize when they are cognitively distracted, such as taking part in a phone conversation; therefore, the risk lasts much, much longer. This likely explains why researchers have not been able to find a safety benefit to hands-free phone conversations.  "
— Understanding the distracted brain:
Why driving while using hands-free cell phones is risky behavior

National Safety Council White PaperApril 2012

One argument is that the data on the smart watch is primarily uni-directional — not requiring a response. For example, a "be there in 5 mins"  message  But if that is the case, how are those messages being generated? By smart watch wearers pulling out their phones to push the messages? And the fact that the message doesn't require a response doesn't mean it doesn't require cognitive effort — and distraction from current activities — to process it.

Wearables such as smart watches may eliminate the time and effort required to pull out the smart phone. But I have my doubts (unfounded, uninformed, as they may be) that they actually reduce the distraction that modern streaming personal data causes.




Thursday, February 20, 2014

Customer Survey Apocalypse

 I recently bought a fairly expensive item. I shopped around, decided what I wanted, and then went in to finalize the deal.

As I was completing the purchase and was shaking hands with the sales person, they leaned in and in a confidential tone told me a story. I would receive a survey through email in a week or so asking about my experience buying the item. The survey would cover a number of topics, including how clean was the showroom, was I told about all the options, was the sales person helpful, etc.

The reason for telling me this, he explained, is that the sales personnel are rated on the responses they receive. What's more, if they receive any responses less than "exceptional", they are considered to have failed. So, if I was happy with the purchase, could I please make sure I marked everything as "exceptional" on the survey?

There are so many things wrong with this experience it is hard to know where to begin. But what spurred me on to write about it is because this is the second time this has happened to me in the past year — dealing with completely different companies and different services!

So where to begin?
  • First, I actually did have a good experience with the purchase and would, without prompting, have given a good review in response to the survey.
  • But having been prompted, I am tempted to not respond to the survey at all rather than deal with the pressure to up my score.
  • I don't blame the sales person. If he is being judged on each survey, why not try to game the system and improve the scores?
  • I do blame the company. They have soiled what was a pleasant experience and made it seem seedy and somehow underhanded.
  • I understand the need to stay in touch with the customer experience. And as problematic as they are, surveys are one mechanism to achieve this. Especially for large national or multi-national corporations
  • But having said that, conflating customer awareness with personnel management is a disaster waiting to happen!
 I know people (and I am not far from being one myself) who would, under some perverted sense of justice, actually lower their scores on the survey in response to the request to score high. There are others, obviously, who would go along to be friendly and up the scores, regardless of what they might have given as ratings without external prompting.

Survey results are problematic enough to start with. 10% return on all surveys sent out is often considered a "good" percentage. Add to this limitation the fact that often those who hold the strongest opinions — either for or against — are the most eager to express their opinion. And now you further skew the results by having employees pushing participants to either extreme.

So rather than get an accurate (or semi-accurate) picture of the customer's experience,  they have invalidated the entire process and skewed the results. At the same time, they are using a completely spurious method for assessing their employees. Don't assess your customer facing employees on what the customer says! It is like telling the cleaning service their pay is dependent on how clean the bathroom sink is. You can be sure the bathroom sink will be clean -- even if the rest of the house isn't touched!

If you are going to rate customer-facing employees on what customers say, base it on the aggregate, not on individual scores. You are still going to sabotage the accuracy of your customer responses, but at least you would reduce the damage. And, possibly, not disillusion the customers you rely on to spread positive word-of-mouth about your business.



Wednesday, December 4, 2013

My letter to Pentax/Ricoh: When Customer Service Fails

I recently had a bad customer service experience.  Such things are unavoidable. In this case, the warranty repair of my brand-new camera was delayed by warehouse closures, lack of parts, etc. These are things I understand.

The bad part of the story is that — although the basic problems were unavoidable — the overall experience was eminently solvable with just the slightest bit of customer awareness and contact. Over the three and a half months my camera was gone, No one from Pentax/Ricoh contacted me. Not to tell me they had received the camera. Not to tell me it was delayed. Not to tell me they were replacing it. Not  to tell me the tracking number when they sent it back. All of these things I had to find out by calling them, repeatedly.

But to make a bad story worse, I wrote to the company (see below) suggesting they would do well to keep customers (who may already be annoyed that their new equipment doesn't work) informed of basic status information, via emails, phone calls, or post cards.

That was three weeks ago. Today a representative of Ricoh called me (a first!) to explain to me:
  1. I had written to the wrong person. Their executive VP doesn't handle issues like this. (Well, duh! But since their website doesn't provide any information on who to contact, I had no choice.)
  2. Once again, he explained why the repair of my camera had been delayed. (Something I had been told, many times on the phone when I called to ask where my camera was.)
  3. Their repair department was in another state entirely. (How does that matter?) And that it was very small and the staff did not have time to be corresponding with customers.
It was that last statement, along with the reassurance that my letter would be "kept on file", that really confirmed that Ricoh, as a company is fostering a culture devoid of empathy or care for their customers. 

Well, if they are going to "file" my letter and do nothing about it, I guess I will too. So I am filing my letter here, online, as a cautionary tale for any other customers who might unfortunately ever need support from Ricoh/Pentax. Better yet, someone considering purchasing any of their equipment might also want to consider their customer service history and attitude before making a decision about what brand to buy.

November 10, 2013


James Malcolm, Executive VP
Ricoh Imaging Americas Corp.
633 17th Street, Suite 2600
Denver, CO 80202


Dear Mr. Malcolm,

Earlier this year my wife bought me a Pentax X-5 camera as a present. I was thrilled. Unfortunately, the camera had a problem that caused it to freeze randomly. What happened next makes me wonder why anyone would purchase products from Pentax or Ricoh. What is worse is that the entire affair was easily avoidable had Ricoh taken just a few simple steps to demonstrate an interest in its customers. The damage to my opinion of Pentax/Ricoh may be irreparable. But I would like to offer some suggestions to salvage your reputation with others.

First let me explain what happened. I started by going to your website to see if the issue with my camera was a known issue. I was told I should send it in for repair under warranty, which I did. Over three months ago.

Since that time, Pentax/Ricoh has never contacted me directly. No notice that the camera was received. No message that it was accepted as being under warranty. No information about when it might be returned.

In that time my wife and I have called your support center at least five times. Each time the support personnel were extremely pleasant and accommodating. Even apologetic. Each time we were told a story about how parts were ordered, supplied were delayed, or some other action was pending. As pleasant as your representatives were, they never were able to give me a clear idea of what was wrong with the camera in the first place or when I could expect to get it back.

After two and a half months, they finally told us “management” had approved replacing the camera.  (Why? Who knows.) Since then we’ve called every week to find out where the replacement is, only to be told that they couldn’t say until they got a tracking number from “the warehouse”.

It is over three months since I sent my camera in, As I said, Pentax/Ricoh has never initiated any communication with me. Everything I know about this story comes from my wife and I dogging your support department on the phone. If I had thought there was any chance of this happening, I would never have purchased a Pentax camera to begin with. And knowing what I know now, I would never recommend Pentax or Ricoh products to anyone else either.

But for the sake of your current and future customers, I would like to suggest the following small changes to your procedures that could make a tremendous impact on overall customer satisfaction:

  • When a customer sends in a product for repair, notify them (by email or postcard) that:
    1. It has been received.
    2. Whether it is accepted as a warranty repair or not.
    3. Ideally, tell them approximately how long the repair will take.
  • If there is any delay in the repairs, contact the customer again explaining the delay and giving a new estimate for completion.

What hurts most about this experience is that we deliberately chose a Pentax camera over other competitors based on the Pentax 35mm camera we have used for more than 25 years. I will never think of Pentax in the same kind light again. And that is sad because the entire episode could have been so easily avoided.



Sincerely,



Andrew Gent


Friday, July 29, 2011

Going Around in Circles

I, like several million other people, have recently been trying out Google+. G+ has received plenty of press in the past few weeks and I don't want to add to the noise. But when I started I noticed two things that I didn't see mentioned until recently:
  • All my G+ friends are KM types, or otherwise involved professionally in communication and social interaction. Few if  any of my "normal" friends are using G+ (or see why they should).
  • I don't like making circles. They require too much thinking.
The first observation were confirmed this week when the unofficial Google Plus Directory (http://findpeopleonplus.com/) posted demographic information on G+ users based on their ascribed professions. Most of the top twenty are technology or information-focused professions. And many of those that are not explicitly "in the business" are questionably tied to technology (such as writers and designers).

My second issue is around circles. I understand they sound like a good idea. My personal (and professional) relationships are more complex than Facebook's simplistic friends / non-friends model.So being able to define your relationships in more detail sounds like a positive step.

The problem is, it's far more difficult than it sounds. I have friend friends and I have professional friends. I have professional friends and professional acquaintances. Some work for my old employer; some used to; some never did. Some know I am interested in poetry and video games (among other things); some don't. A few have met my wife; some may not even know I am married.

When I start to break it down, it is not only not binary, it is more complex than even I can describe. Which is what makes Google+'s circles so frustrating. They require too much thinking. This is not a technical issue, per se, but a failure to be able to turn an implicit organic process into an explicit concrete categorization.

In other words, my friends are analog and circles are digital.

Andrew McAfee confirmed my suspicions in a blog post. He goes into far more depth and argues that it is an issue of a priori vs. a posteriori decisions.I am sure he is right from a process perspective, but I am not even sure deciding after I find an item to share is going to help that much.

Part of the joy of Twitter is that there is no decision. You post or you don't. You open yourself to anyone who chooses to listen (essentially). Oh, it has its limitations as well (starting with the length of the messages). But the freedom from thinking about who a message is intended for can be quite liberating.

However, that freedom doesn't have much to do with friends; it has more to do with publishing (or proclaiming). But it can be a useful and easier process in the digital world than trying to sort out your friends.

    Monday, July 18, 2011

    Holy Crap, Batman! The Social Business Stack

    I just read D. Hinchcliffe's () Social Business Stack at the Dachis Group and all I can say is [explicative deleted]. Dang! That's one impressive and imposing architectural diagram!

    I'm not saying the diagram is architecturally incorrect. In fact, I suspect it is accurate from a corporate IT perspective. It looks like so many other all-inclusive architectures.

    The trouble is no normal human being in their right mind could look at it and do anything but shake in their boots. This is the sort of diagram that justifies five years of intense IT investment. It also presupposes (or pre-justifies) failure since there are so many moving parts.

    The stack is accurate in that it captures all of the possible interactions and interdependencies from a KM and IT perspective. (That is, the old people/processes/technology triumvirate.) But the fact is no one really cares about anything but the top layer. (The Social/People layer.)

    So why is this so complex and social networking "in the wild" so simple? Well, it isn't that simple in real life. But:
    • On the public web people are more than willing to do things manually to "make it work", such as putting in links to blogs, etc by hand.
    • If it does become difficult, there's an app for that. People are happy to juggle 5, 10, even 20 separate apps such as bit.ly, twitpic, intagr.am, last.fm etc to achieve their goals. What's more, it is cumulative: people learn new tricks from watching their friends' posts.
    • Ultimately, the public internet is an almost limitless (since it is always growing) source of additional material, support, inspiration, or target for discussion.
     In other words, all the other layers of Dion's stack exist in the public instance but no one cares about them. Not that they aren't necessary. The next four layers (Data, Delivery, Aggregation, and Discovery) are just assumed to be there. And the critical vertical integration "glue" is heavily biased towards manual effort and simple HTTP links, rather than some complicated automation.

    The last two layers (Security and Business Model) also exist. But people are amazingly carefree about security on the public web and the Business Model is the responsibility of the technology/service providers and people simply give a yea or nay vote on the instantiation by staying with the service or moving on.

    So, what does this mean? I think the first meaning is that, as usual, corporations are taking something simple (or deeply complex but with a simple surface layer) and getting caught up in the morass that underlies it. Secondly, what the stack doesn't show is the often terribly anaemic state of the lower stacks behind corporate firewalls. The oft-repeated aphorism "If only we knew what we know" can usually be expanded to its various corellaries:
    • "If only we knew who knew what we know"
    • "If only we knew where we stored what we know"
    • "If only we could find what we know"
    • "If only I had permission to know what we know"
    • etc.
    So, I think the social business stack as represented is correct. But I am terribly concerned about what such a diagram would be used for. Because, ultimately, it is people — not technology or processes — that are the deciding factor. And people have astonishing resilience and patience for "making things work" when they have an interest in the outcome.

    Friday, April 1, 2011

    The Ultimate Architecture Diagram

    I carry a small notebook with me at all times to jot down ideas, reminders, fragments of thoughts, or just doodle in my spare moments. Over the years I have filled up quite a few such notebooks.

    Many of the items in them I can identify, some I cannot. They include drafts of messages to coworkers, a phrase or word I thought critical at the time, or notes for business presentations long since given and forgotten.

    I was leafing through one of my notebooks the other day when I came across a curious diagram. I don't remember where or when I drew it, but looking at it now I am struck by its simplicity, its utter honesty and completeness.



    Yes, there is stuff. And there is other stuff. Beyond that, very little matters.

    The issue, from a business, technical, and/or personal perspective, is being able to separate the right "stuff" from everything else. There lies the rub. Maybe I captured that in another diagram, but I can't find that one at the moment...

    Monday, February 21, 2011

    Whose Knowledge Is It Anyway?

    In previous posts I have discussed the shifting relationship between employer and employee in terms of the ownership and responsibility for knowledge. Many people are taking advantage of the web 2.0 revolution — through blogs, wikis, etc. — to assert the individuality of what they know and their hard-won professional experience.

    Employment always combines a certain amount of both the carrot and the stick. As much as you might enjoy what you do professionally, there are always a few things that are necessary for the company that you would choose not to do if given the option. So, the employer/employee relationship is always a collaboration, a compromise of activities that meet the needs of each.

    Salary, bonuses, and promotions are obviously "carrots". Performance reviews, management dictates, and the threat of a pink slip are part of the "stick". In balance, these two components benefit both the employer and the employee. However, when they fall out of balance, negative things start to happen.

    In the early twentieth century, when industry used the unrestrained threat of firing, low wages, and even physical violence to control the workers, the result was the labor movement and emergence of unions in the United States. As the twentieth century came to a close, the rise of the global economy and multinational corporations gave employers a new out. Not only could work be moved out of state, it could now be moved to another country entirely — leading to 10-15 years of aggressive business tactics euphemistically called downsizing, rightsizing, outsourcing, and offshoring, among other things.

    There would seem to be little the employee could do to counter this trend. Except, we are no longer in an age dominated by physical manufacturing. We are in what is referred to as the "information age". Business magazines have been touting the power and transformative capabilities of information for years now.

    And if information is the currency, ownership of information is power. So, who owns the information? Corporations would like to think they own the creative output of their employees. And it seems true enough that they rightfully own the direct output and artifacts of work done under their employ. This output may be physical products (such as tables and chairs for a furniture manufacturer), electronic products (such as software), services (such as installation, repair, or management services), processes (such as standard operating procedures or decisions trees) and any source code, documentation, or preliminary designs that led to that output.

    But do they own the knowledge and intellect used to create that output?

    They would like to think so and often lay claim to the knowledge, putting restrictions on what their employees can do with that information during — and in some cases after — their employment. But unlike the industrial age where the means of production were tangible objects (such as looms, kilns, and presses), today the means of production is knowledge.

    And knowledge, unlike a physical device such as a loom or a lathe, is not tied to a specific task. Knowledge is also not separate from the employee who possesses it. Part of the power of the web 2.0 revolution is the synergy between frictionless global communication and the realization by individuals of the importance of the knowledge they possess. And make no mistake, they possess the knowledge, not their employers.

    You cannot separate knowledge from the knower. And although you can argue that part of what they know may be a trade secret or other company-specific information, the abstract understanding of how things work and the experience of doing it, belong to the employee.

    If this sounds a little like Karl Marx revisited, it is not a surprise. The pressures being applied on employees in the late 20th/early 21st century are not unlike those of the late 19th/early 20th. The difference is that — unlike in the late 1800's —the ability to code is more important than the computer you write the code on. And as the technology itself becomes commoditized, the ability to code becomes more expensive as well.

    So what are the repercussions on employees, employers, and how you manage the knowledge between them?
    • Heavy-handed attempts to assert ownership or control over how employees use knowledge will produce resentment and resistance. This was true even before the internet age or the knowledge economy came into full bloom. But now it is brought into higher relief, especially when the employee can choose to limit use of that knowledge, either consciously or subconsciously, if they feel it is being undervalued.
    • In response, rather than unionizing, which was the approach chosen in the early 20th century, information workers in the 21st century are choosing to "socialize".
    By "socializing" I mean establishing a vibrant, dynamic community of peers where knowledge and experience is traded freely outside corporate boundaries. How is this beneficial to the employees? In several ways. Most importantly, it creates a reciprocal arrangement bartering knowledge for reputation, which works like this:
    • Individuals in need of information search the internet — including blogs and technical forums — in search of answers. If they cannot find what they need, they may ask in a forum, discussion list, or openly through services such as Quora or Twitter. By looking outside the corporate, these individuals are likely to find more unique, complete, and specific answers faster than if they stay within the firewall. In addition, they often get credit for the solution.
    • At the same time, their peers post information about their experiences — either in response to questions or as knowledge in blogs, wikis, etc. — both to help other people and to establish a reputation for themselves as knowledgeable about their field of expertise.
    Since the individual bits of knowledge being traded have minimal commercial value in and of themselves, there is no loss to the individual sharing what they know. At the same time, those knowledge tidbits can have great value to peers who are trying to solve a specific problem. As a result, a collective market of sharing and reputation building is created among practitioners completely outside of corporate boundaries.

    This knowledge sharing ecosystem is extremely loose; there are no formal definitions or boundaries. The community is composed of like-minded individuals communicating through blogs, forums, websites, and social media with no official connection beyond a commonality found in search results, comment threads, blog rolls, retweets, and the like.



    Beyond just the basic exchange of information, the blogosphere provides knowledge workers with additional benefits:
    • An outlet for ideas that are overlooked, under appreciated, or simply out of scope of their current work environment.
    • A far greater, sometimes critical but often more enthusiastic, audience for their thoughts
    Finally, the relationships established through interactions within one's profession and the reputation garnered in the open, critical eye of peers can be indispensable in the not-so-distant future. For example, peer connections made now can be indispensible when when looking for a job some time down the road.

    And smart companies are taking advantage of this change, often using blogs and forums to actively promote openings, search for good candidates, or to qualify those who apply for positions.

    The hyperbolic claims often found in resumes can be hard to verify. But an openly published and proven knowledge of the subject at hand goes a long way to convincing a potential employer that someone has what it takes.



    Of course, there are downsides. Just as participation in these extracurricular activities can help establish your reputation as a leader in the field, individuals with an aggressive, dismissive, or over-assertive personality can establish a reputation of a very different kind. When you participate in public discussion for any length of time, both the good and bad aspects of your personality will come to light.



    Ultimately, it the the individual's knowledge that counts. And the world of web 2.0 provides a vehicle for that individual to share knowledge with their employer, their fellow employees, and peers around the world to the ends he or she sees fit. Whether their employer approves or not. And, quite frankly, in many ways the world, and the individual, are better off for it.

    Sunday, January 9, 2011

    What Knowledge Management Can Learn from Small Groups

    I used to work for a large multinational corporation. I now work for a small startup consisting of 12 people who work in one room together. There is not much "knowledge management" needed with a group that size. That doesn't mean knowledge management doesn't happen, just that it happens more instinctively and with less stress around the edges.

    In fact it would seem reasonable to assume that there is little if any relationship between the two situations. But in fact there are some interesting lessons when observing small groups that can be applied to larger corporations:

    Everyone is different

    Even though my current company consists of only 12 people, there are 12 different personalities and approaches to work, communication, technology, etc. When you work in large organizations, there is a tendency to talk about how people will respond to new programs as if it were a unary decision, where all (or at least most) people respond in one way. We are then surprised by the number of people who fall outside of the defined norm.

    Every project or process has a target behavior — how you want people to use the process. But that behavior is only a target. The overall response, when all individual behaviors are taken as an average — may fall within the target range. But any single person is likely to have their own particular usage model that may well be unexpected.

    Multiple, overlapping technologies are not a problem

    Within my small sampling, we have 12 unique sets of technologies, including different operating systems; different hardware (some laptops, some desktops, often both); different software tools; and countless communication devices. Everyone has email, everyone has an instant messaging client of some kind (or two or three), we also have wikis, blogs, forums, and an IRC channel. Not to mention smart phones, blackberries, iPads, etc.

    We occasionally have a discussion about the appropriate place to post information — especially material under review or in draft form. But I have yet to hear any complaints that there are too many choices.

    In large organizations, one of the basic requirements for any project is keeping the toolset small. If there is to be a knowledge management "system", it has to be a single system accessible by anyone in the target audience. Better yet, a single system covering multiple disciplines (KM, project management, resource management, etc.)

    The rationale for keeping the toolset small sounds good in theory:

    • Universal accessibility
    • Reduced learning curve/training costs
    • Only one place to look for information
    • Reduced IT/support costs & complexity
    But this rationale is based on 1980's computing constraints:

    • Universal accessibility — what technology, especially collaboration technology, isn't available through a web browser or across platforms (e.g. IM clients)? Selecting one tool does not make the information more or less accessible.
    • Reduced learning curve/training costs — at the same time corporations are trying to restrict the number of applications to "reduce the learning curve", their employees are busy trying out Facebook, Twitter, Skype... The main reason learning curve is a problem is because you are trying to teach people something they don't want to learn. Perhaps the issue is with the content, not a limitation of the audience's ability to multi-task.
    • One place to look for information — I have heard this argument for years, but I have yet to see a single instance where a company has successfully integrated all information into a single application. In fact, corporations seem determined to segment their knowledge into individual repositories. The closest they come to "one place for all information" is intranet search. However, they determinedly resist efforts to use generalized search engines (such as Google) and often limit what information is indexed by search under the auspices of "qualifying" content. (What happened to "one place"?)
    • Reduced IT/support costs — I used to believe this argument, because it was true. But over time, just as the locked door computer room has shrunk and more and more technology (and computing power) has migrated from a secure, air-conditioned environment.... onto the desk... out of the office... into the pocket... the role of IT in controlling — or even choosing — technology has changed significantly. But IT as an organization and as a profession seems unwilling to accept or accede to that change.
    However, individual technologies can run foul of individual preferences

    The fact that people have multiple technologies, doesn't mean they use them the same way (cf. everyone is different). Years ago, I was shocked when I answered my office phone to discover that the caller was in an office no more than thirty feet away. But I thought nothing of sending email to the person in the cube next to me.

    More recently, I was bemused the first time I received an instant message from a fellow worker two cubes down. (They didn't want to disturb the others by talking, since we work in such close proximity.)

    How and when individuals use different technologies seems like an almost limitless set of permutations. Of the 11 people I currently work with:

    • At least one answers email before IM
    • One seems to respond to both equally (and instantaneously)
    • Several answer either IMs or email, but with no clear pattern or preference
    • One will respond to IMs more often than email, but will answer the IM via email.
    • One never responds to IMs.
    Is this good or bad use of the technology? It is neither. It is how individuals work. Part of "knowledge management" is managing your sources of knowledge, your technology, and your contacts. It is not enough to know how to use the technology; you must also know how it is used by your community.

    I know this concept — the preeminence of personal choice — is an anathema to many KM practitioners. It is like trying to establish order without disturbing the chaos. How can you promote a company-wide program if each individual gets to choose for themselves?

    Well, it is not quite that bad. It is not that each individual gets to decide for themselves. You can dictate, require, or recommend specific technologies and approaches. But you need to recognize that your audience will perform those actions in the way they think is best.

    Finally...

    Have faith in people

    It is easy to see other people's behavior — when it runs counter to expectations — as stubborn or willful when it is nothing of the sort. People will be altruistic, especially when it involves assisting other individuals. However...

    KM is not their job!

    As well intentioned as they may, people have a job to do, deadlines to meet, and responsibilities to uphold. If they think of it, they are willing to share with others. But more often than not, it does not occur to them that the information they hold is valuable to others — especially if that value will not be realized until some indeterminate time in the future.

    Discussions held in hallways or decisions made over lunch are sometimes the most important events within a project. But no one thinks to capture them in a wiki or email the rest of the team. This is not knowledge hoarding, it is simply an inability to recognize that anyone else cares.

    Ultimately, perhaps the most powerful KM tool any company has is nagging repetition. When someone writes something down, remind them to post it to a forum or wiki. When they say something interesting suggest others they should tell via email. Suggest alternate ways to search for solutions to project problems.

    This sort of gentle persuasion on an individual basis can be tedious, since the scope is limited to specific situations with one or two people at a time. And I am not suggesting it alone is sufficient to make KM work on a large scale. However, it is surprising how soon you see others (who you have prodded) acting without instigation or suggesting it to others. And from such small efforts, large effects can accrue.

    Well, that is OK for a small office, but how does this apply to large corporations? The most successful KM programs I have seen, even in very large corporations,  have always had one or two advocates who were tireless in not only promoting "the program", but jumping in and helping individuals with their specific problems and demonstrating KM-ish techniques along the way. Their influence went far beyond just the person they helped, but to anyone that person then spoke to, their friends, etc... Not only their reputation preceded them, but the behavior they modeled went with it to corners of the company they might never have visited personally.

    Wednesday, May 5, 2010

    Lurking or Lost?

    Stan Garfield made an interesting comment in response to my post on lurking. Stan said:

    I assume that most people who stay subscribed to a community discussion board are usually paying attention to what is being discussed. They can benefit from doing so, even if they post infrequently (or never).

    Part of what Stan is positing is true. People who are not actively participating can -- and often do -- benefit from the discussions within the community. That is the aspect of lurking that makes it a constructive activity.

    But it is dangerous to generalize this belief, particularly if you are managing communities. Which I suspect is why Stan qualified his statement with "most". The distinction is whether people are actually lurking or have turned off completely.

    Membership, or subscriptions, are often used to measure the size of a community, using the premise that, if they are still members, they are at least "actively lurking".

    However, going back to my own experience, I currently belong to at least ten online communities where I receive email:

    • Two I actively follow, reading each email as it comes in and -- occasionally -- responding.
    • One I read thoroughly but never respond to. I consider it "keeping up" with a specific technology area.
    • One I read if I have time, but at least half the time I skip.
    • Two I browse the subjects lines but almost never actually open the email before deleting.
    • Of the remaining four, at least two I delete without even considering the subject line. One I delete angrily after seeing that their infrequent messages are wasting my time, and the last falls somewhere in between those last two categories.

    So which communities am I actively "lurking" in? If you asked me, I'd claim to be a "member" of three of the ten. Four at most. Less than 50%.

    In truth the situation is much worse than what I describe, since I am a registered user of 10 or 20 more communities which I do not receive email subscription notices from. Of these I actively visit the forums of, perhaps, 5; occasionally visit another 2; and probably can't remember my password for at least half of what remains.

    If that's the case, why don't I unsubscribe? Laziness, probably. About once a year I go through and remove myself from the most egregiously bad lists. But over the next 12 months I probably sign up for more new ones than I jettison. Some communities I signed up with for a specific purpose, which has since been satisfied. Some I joined as an experiment. And for many, there is a sense of not wanting to miss something that I might otherwise not hear about. (Although time proves that supposition wrong.)

    Whatever the cause, I am technically a member of far more communities than I am actually "involved" with, whether active or lurking. Which brings us back to the dilemma for community leaders: how do you measure the size of an online community?

    You can say for certain that people actively participating (posting, responding, or otherwise contributing) are members. But as has been said time and again, this is usually a very small percentage of the actual community size.

    On the high end (assuming your community supports or requires some form of registration) you can count all registered users. However, as I have just explained, that number is unlikely to have any true meaning. So reality falls somewhere in between.

    Second Life was roundly criticized four years ago for not being clear how they counted their "residents". (To be fair, the company improved both their methods of counting and reporting membership in response to the complaints.) But part of the confusion is not just how they were counting, but disagreements about what methods are appropriate.

    The situation is no better for internal communities within corporations. Ultimately, there is no true answer. The interest level of individual members of the community will ebb and flow constantly, often with no external indications. So there is no way to provide an precise measurement of "engagement", even as a snapshot.

    If you have subscriptions, you can report a combination of senders (unique posters) and receivers (subscribers) to give a range of possible activity. However, this leaves out anyone who reads the message online or (as is increasingly common) through an RSS feed. If you have access to web logs, you can report the number of unique visitors to the site/forum. But merging this data with email subscriptions can be difficult.

    But no matter how you count, there are a variety of actions community facilitators can take to "take the pulse" of the community, beyond just numbers.

    • Talk to your members. Contact people directly to ask if the community is meeting their needs or not. and if not, how it could be improved. Ask them how its going.
    • When questions are asked, forward the question to people you know are knowledgeable on the subject and suggest they post an answer. This not only keeps the conversation going, it helps gets more experienced members involved and can spur interaction between members who do not know each other yet.
    • If there aren't any questions, ask some. Not trivial questions, but questions that will get the community thinking -- and talking.
    • Thank people, either publicly or privately through email, when they contribute significantly to the community. Make them feel appreciated for their efforts.

    If this sounds more like hosting a high society soiree than facilitating a community, there's a good reason for that. They are very similar tasks. It is not enough to schedule a party, hire a caterer, and send out invitations. Once the event begins, you must play host: introduce people so no one feels left out, make sure they circulate, suggest activities... even plan party games! The exact same sort of activities that are needed to keep a community going once it has begun. What's more, being actively involved yourself gives you an intimate and immediate sense of the health and well-being of the community.

    Measurements are fine and often necessary to convince management that positive momentum is occurring. But more important is knowing for yourself, as community leader, that your members are involved and benefiting from their participation. And this is not a number, it is a state of being you can contribute to yourself.

    P.S. Stan himself has articulated many of these and other good ideas for how to actively facilitate communities in his own writing. Most recently, in his community manifesto. Recommended reading for anyone interested in the topic.

    Wednesday, March 31, 2010

    Collaboration at the Edges

    Last month there was a discussion going on in a KM group I participate in regarding rewards. The question being asked was "what incentives are in place for knowledge sharing within your organization?"

    Now, those involved in KM will recognize this discussion. It is a recurring theme, sort of like abortion, with no clear answer but lots of heated debate on both sides.

    To summarize quickly, there are those who feel that incentives (ranging from simple acknowledgments to actual physical or financial awards) applied strategically can encourage participation and knowledge sharing. While others feel any form of reward system distorts normal behavior, resulting in "false positives" as individuals game the system to collect the rewards, without any real engagement in KM practices per se.

    There seem to be valid arguments on both sides, and I have no expectations that I can sort out the issue where so many others have failed. But I do have a different perspective that may have some value. You see, over the past year I went from working for a large corporation (over 50,000 employees) to working for a company barely in the double digits (10 employees, to be exact).

    Now you might not think that the activities of ten people working together have much bearing on a large geographically distributed corporation. And, at first blush, you would be right.

    We have no trouble with knowledge sharing. We work in close proximity. Conversations break out throughout the day and anyone can and does join in. Questions are asked, frustrations expressed, solutions proposed, and problems solved with almost no conscious effort whatsoever.

    It seems almost ludicrous to ask "why don't we have issues with KM?" Isn't it obvious? We all know each other. We are working closely together on a common project.

    But we aren't all the same. I'm a writer. Most of the team are engineers. There's a manager. And regardless of our roles, we all have quite distinct personalities, which might or might not be compatible in other situations. But none of that gets in the way.

    So is it the distance, the lack of knowing each other, or the loss of a common goal that stand in the way of sharing in larger environments? The answer is probably "yes" to all three -- plus a few other characteristics of working in distributed corporations.

    But I have another experience that influences my view of this problem. Years ago (many years ago) I worked for another large corporation. However, at that time, there didn't seem to be any of the difficulties with knowledge sharing I have seen since. The company was large, distributed, and working on many different, often unrelated, products at one time. In other words, no obvious common goal.

    In many ways, the employees (particularly the engineers) of that earlier company operated as if they were a team of ten rather than an organization of 100,000. People I had never met responded to my questions openly, offered suggestions, even took time out to assist if necessary.

    Most importantly, there were no explicit incentives to encourage this behavior. It just kind of "happened".

    Of course, that's not true. It didn't just "happen". The ethos of the company, the culture of the organizations within it, and the personal dedication of the individuals the company chose to hire conspired to create that environment.* But, three things stand out about these examples in my mind:

    • If size and distance are detrimental to knowledge sharing -- as they clearly are -- what is it about incentives that would counteract that?
    • More importantly, what is it about size and distance that causes the problem?
    • And why is it the earlier company was able to overcome those obstacles without incentives?

    As I said earlier, I have no expectation that I can actually solve this dilemma. But I have a suspicion. And my suspicion is this:

    People share openly when they feel they are part of a community

    Not a member of the community, a part of the community. They share because they are assisting the community, even if the sharing is one-to-one with another member.

    Clearly in the case of small groups, it is easy for everyone to understand the common goal. The individuals cannot succeed unless the group (i.e. community) succeeds. As a consequence, they are eager to contribute where they can. Even, paradoxically, when those contributions are quite tangential to the shared goal (such as recommending good restaurants or which GPS system is best).

    Note that lurking -- a behavior specific to members of a community rather than participants -- is difficult in small physical groups. Since the group is so small, people see when you hold back and may even challenge you to open up. Such interaction is much harder in large, distributed groups.

    So getting back to the initial question: what can incentives do to counteract the negative impact of size and distance? Incentives can encourage lurkers to speak up (thereby improving whatever quantitative metrics are in place). However, incentives cannot alter the psychological affinity an individual feels towards to community. At best, the incentive may spur an initial (and temporary) jump from lurker to participant, which the individual then finds satisfying. This success may spur them to try again, and over time start to develop a sense of ownership in the group. (In other words, become part of the community.)

    This, I believe, is what advocates of incentives are aiming for.

    * For a more complete assessment of this earlier company, see Patti Anklam's excellent article The Camelot of Collaboration.

    Friday, February 26, 2010

    Lurking, a Personal Story

    I belong to several professional communities, mostly online email discussion lists. I sometimes participate in the conversations. But mostly, I lurk.

    It is not that I am a novice (I've been at my profession -- or professions -- for a number of years). It's not that I don't have anything to say. In fact, quite the opposite.

    I was reminded of this recently when a discussion came up about communities and incentives. I've worked in KM for ten years or more. I've also witnessed a number of incentive programs, both successful and not so. So I have no lack of opinions and experiential knowledge I could share.

    But I haven't said much. Why? Because these topics, communities and incentives, are areas of contention in the field of KM. What are they? How are they best employed? Do incentives help or hurt? There has already been plenty of space and time spent arguing both sides.

    So I try to pick my words carefully. I even started a blog post on the topic to try and organize my thoughts. However, while I picked and edited, others chimed in with their opinions, advice, and war stories and the conversation quickly moved on, diverging into several related threads.

    Each time someone else adds to the discussion, there is something new I feel I need to refute, promote, analyze or otherwise respond to.

    In the end I don't respond at all.

    That, in and of itself, is not interesting. But what may be interesting is what was going on in the background.

    I am not just lurking, I am also implicitly deferring to the others who are responding. Note, I don't agree with them all. And there is a certain level of angst at not responding to the comments I disagree with. Because I am emotionally involved in the topic of the conversation. But either because of the number of interrelated concepts being bandied about or the speed of the conversation, I do not feel comfortable joining in.

    As a consequence, at least to some small degree, I feel distanced from the community as a whole. And therefore lurk some more.

    This is a cyclic event. I've noticed myself do this time and again in more than one environment. I lurk, I sometimes feel better about the community, sometimes worse. Then something will spur me to respond. I'll actively participate for awhile, then go back to lurking.

    There are a number of different reasons I don't chime in:

    • I agree with what's being said and don't feel like piling on (or distracting from the points already made)
    • I disagree with what is being said but don't want to get into a pissing contest about it.
    • I disagree with what is being said but can't get my thoughts in order enough to make a cogent argument before the conversation moves on.
    • I disagree with what is being said, but it is such a minor aspect of the overall conversation that I feel arguing it would be an unnecessary quibble.

    There are probably others, but those are just some that come to mind. But more importantly, there are times I do chime in. And those times seem to fall into three categories:

    • Someone is asking for assistance and -- before anyone else responds -- I feel I have something unique or clear to offer.
    • Someone is asking for assistance and -- although I may not feel uniquely qualified -- no one else responds in a reasonable time frame (anywhere from a few hours to a day or so), so I feel my response will be of use.
    • Something is said I disgree with so much, my eagerness to correct it overcomes any of my inhibitions about going off-topic or raising an argument.

    What really struck me was the last cause. It really goes against the common beliefs about communities.

    We (KM consultants and practitioners) tend to discuss lurking as a static state. We talk about communities being made up of 10% core contributors and 90% lurkers, as if these are permanent labels you can apply to individuals. And one of the key goals of incentives is to convince members to "cross over", stop lurking and become active contributors.

    To be fair, incentives can help. Just as taking new arrivals at a party around and introducing them to those already there can help get them to join in. Incentives done properly can help new members "break the ice".

    However, this concept of lurking as a static state is so widespread, community members themselves often adopt it and feel compelled to confess their status. At least once a month I read a post starting something like "Hi. I'm normally a lurker in this forum, but..."

    The fact is, lurking is not a static state, it is highly dynamic. The cast of active participants changes over time as people flip between lurking and leading. At the same time, lurkers are often paying close attention to what is going on, are emotionally responding to events and -- moment to moment -- adjusting their attitude towards the community based on what is said. They agree, they disagree, they get angry, insulted, surprised, or even flattered (when they feel they or their subgroup have been mentioned positively) all without breaking out of their silence.

    And, in opposition to common belief, it is not an improved attitude toward the community that is needed to get them to join in. Often it can be displeasure or disagreement that instigates a need to speak. Of course, how their contributions are received (they don't have to be agreed with, but at least considered and respected) impacts their likelihood to contribute again.

    The real difficulty for community facilitators is that this dynamic, changeable, but silent lurking makes it hard to measure the actual "health" of the community as a whole. The volume of discourse may remain high. But as with any voluntary endeavour, people can "vote with their feet". But for online communities, you can't "see" lurkers who choose to tune out. And it isn't until the conversation starts to dwindle (because there are no lurkers left to chime in or pinch hit when an active participant drops out) that you realize there is a problem.

    I can't say I have much to offer in terms of positive techniques for counteracting this situation. But I believe that having a more realistic understanding of the dynamics of lurking may help us address the needs of communities we watch over in a less back and white fashion.

    To start with, it is important not to stigmatize lurking. Everyone does it sometimes and lurking doesn't mean you aren't emotionally involved. Accepting lurking as a normal stage in a cycle may help make the distinction between active and passive particition less daunting to new members.

    And, of course, finding a way to gauge the involvement or commitment of lurkers could go a long way to understanding the actual health of the community as a whole. How to do this is still a challenge that has no clear answer...

    Sunday, February 7, 2010

    What Happened to Postcards?

    I spent much of yesterday at the Museum of Fine Arts in Boston. It was a singularly exhausting experience -- as most museum visits are. (Being a combination of exhilaration and stultification at the same time.)

    But what particularly struck me was when we visited the gift store before leaving (I can't resist tacky souvenirs) and there were no postcards for sale.

    Yes, they had one small rack of postcards of Egyptian paraphernalia, capitalizing on their current special exhibit of an Egyptian tomb and children's fascination with mummies. (Egypt is to art museums what dinosaurs are to science museums.) There was also a book of postcards depicting paintings by Monet. But there were no small mementos for sale of the individual works that may have struck a chord during your visit.

    At first I was confused. But then it occurred to me that there may be a very practical reason why postcards are missing: no one sends physical mail anymore.

    Now, this is just supposition. There are a number of different reasons why they may no longer sell postcards: too expensive to produce, take up too much space, need to constantly change stock to keep up with what is on display at any given time... But these conditions are either identical to what they were 20 years ago or easily offset by inflating prices (as they do with the trivets, posters, T-shirts, and other items that are on display).

    So I can only assume the market for postcards has itself diminished because people do not send physical mail anymore. Statistics from the US postal service verify this, indicating that personal correspondence via USPS decreased 14% between 2002 and 2008. And the drop off is expected to continue.

    The unfortunate part of this situation is that postcards play a role beyond just souvenirs and something to write "wish you were here" on. Postcards, especially postcards from places like museums and zoos that have many different exhibits, serve as mnemonic devices. These mnemonics remind us of the strong emotional experience of seeing the painting, sculpture or whatever. They also act as a surrogate of that experience that we share with those we send the postcards to.

    People do not send as much mail because email and other electronic media have replaced the need for physical letters and cards. (As well as being easier, cheaper, and more convenient.) In place of postcards, I could have taken pictures of the paintings I wanted to remember -- which I did in a few instances. But the lighting in museums is hardly conducive to photography. (In some cases, it doesn't even seem very conducive to viewing!)

    So what should be done? As much as I enjoy postcards, I recognize it is not practical to argue a return to a form of gifting that never was very practical and is now downright archaic. But it would be a loss to the patrons -- and to the museum -- if there were no form of mnemonic to help visitors retain and relive the pleasure of seeing the art in first person.

    If it is not financially viable to stock physical postcards, perhaps they can make it possible to send electronic postcards or custom "picture books" of one's favorite works, whether to yourself or to your friends?

    But, surprise surprise. They almost do...

    The museum has an searchable online catalog of many of its holdings. The catalog has an expansive advanced search capability. It even lets you send e-cards once you find a specific item. (Yes!)

    However, the catalog is only available if you are on the internet, not in the museum itself. (No!) Add to that, the catalog is really designed for those who understand how the catalog works, not the casual user. (For example, the search interface has 12 fields. Enter "Egypt" under culture and search for items on display and nothing shows up. Search for "Egypt" as a keyword and 58 pages of results are returned.)

    What would be great would be if there were monitors in each room that let you browse the items available in that room. (No painful searching.) You could select an item and send an e-card in seconds when it strikes you, rather than spending minutes (or more) searching for it later.

    Better still, for those with smart phones you could provide a simplified interface that only asks for the asset or accession number. (The accession number appears on the bottom of the placard describing each work of art.)

    (Example placard with the accession # 72.2617)


    The visitor could quickly call up an item they liked and send e-cards to others or a reminder to themselves on the spot. They could even create an e-book of their favorite works as they proceed through the collection.

    In an ideal world, the MFA could put 2D barcodes on the placards so the information could be scanned and retrieved automatically by cellphone. This would be the easiest method technically. However, it would require changes to museum itself (the placards) and a common interface on cell phones -- something available in Japan but still not standard in the US yet.

    So sticking with the first two suggestions -- and especially the suggestion for a simplified search available on smart phones -- it would be possible for the MFA to provide a thoroughly innovative and satisfying way for visitors to remember and share their experience with very little change to the existing infrastructure.

    Why should the MFA bother? Although this sort of addition would have a cost associated with it, much of the technology already exists in the electronic catalog. Simply by creating a targeted interface -- and promoting the new capability - the museum can deepen the experience for the patron as well as advertise its best qualities through the messages the patrons send.

    The word "souvenir" comes from the French for the act of remembering. Postcards are both a souvenir in the sense of a mnemonic device and a vehicle of communication. Their loss may seem minor from a commercial perspective, but they served as a valuable thread connecting the museum experience (quiet, austere, contemplative) with the visitor's regular life (loud, jumbled, exciting and excitable). A thread by which we carry the experience of art from previous generations back into our lives.

    Without it, the day at the museum is just that: a day at the museum. But museums have the opportunity to create an even stronger link, electronically. It will be interesting to see if they pick up the challenge.

    Sunday, January 10, 2010

    The Work We Do

    I recently changed jobs. As a consequence I am no longer "doing" Knowledge Management. I am reminded of this fact by friends who ask me how much KM is involved in my new job. The simple answer is none. But to be honest, that's not entirely true.

    My new role brings me back to my roots in Technical Writing, Information Architecture, and what is currently known as "Content Strategy". (Although "Content Strategy" appears to be having the same sort of identity crisis KM and IA go through on a regular basis.)

    Which may also explain why my new role doesn't feel so different. In my previous profession, people would ask me how Information Architecture relates to Knowledge Management. My pat answer was that KM is the architectural design of potential -- rather than existing -- information.

    My response was flippant, but also quite accurate. Whether you are defining the structure of a website for well-understood content or designing an interface for some as-yet-undefined content that will be chosen by people in the future... the tools, the methods, and the experience you use are much the same.

    Ditto technical writing, which is very much like information architecture except on a much smaller scale. (What goes in this document vs. another? Where do we put the introductory information so the user can't help tripping over it? etc.)

    Now, I know there are technical writers that do nothing but write and are affronted if you suggest they "design" things. Just as there are Information Architects that design taxonomies and not much else. But the fields themselves are much bigger than that. Which is where the confusion and bickering comes in.

    I seem to be constantly in the middle of one battle or another no matter which of my various "professions" I am practicing. The reason for that is because the boundaries are very fuzzy. And ambiguity makes people uncomfortable.

    So they try to delineate their roles. On the one hand, practitioners try to define the field by how they currently practice it: the tools they use or the methodology they have adopted. While others with a more philosophical bent try to expand the scope, often treading on the toes of their neighboring professions.

    Currently, information architects are arguing whether IA even exists or if it (and several other forms of design) are all just different flavors of a new profession, user experience design. Or is it interaction design? Or is it...

    Similarly, KM practitioners (and pundits from other fields) are trying to decide if there is a battle going on between KM and social media. Excuse me? There is no battle unless you assume KM as a field of study has to be practiced in a specific way or within a predefined, limited field of vision.

    I feel like channeling a famous ex-wrestler and shouting "It doesn't matter what your profession is!"

    The names we make up for our jobs help avoid conflicts when working with others by divvying up the territory. They also give us a ready answer to social situations when someone asks "what is it you do?"

    Unfortunately, these names also create unnecessary barriers to getting work done. If you define your role by specific tools or tasks, you are also defining the boundaries for the solutions you can provide. You doom yourself to repeating the same work over and over... even when the environment around you changes, as it inevitably will.

    The fact is that all of my professions are variations on addressing the traditional dilemma of communications theory. Whether it is communication between management and employees, among the employees themselves, between the company and customers, or among the current and potential customers the company seeks, the professions I travel with are all trying to resolve the problem of getting information into the right hands at the right time.

    In knowledge management it is creating channels for the ambient knowledge within whatever community you support. In this case, your audience is often both content providers and consumers at different times. You don't control the content, you try to maximize the channel.

    In information architecture it is sorting, defining, and providing a clear logical structure to at least that part of the information space you have control over (whether that be a web site, marketing, promotion, community facilitation, or whatever). You manage the content, somewhat like an orchestra conductor. But you are not the composer.

    In user experience and usability design it is tuning the communication channel to be as effective as possible. You don't control the content or the structure, but you control how the audience interacts with it. (Of course, this is untrue in actuality because the interface becomes part of the message. Wasn't it McLuhan who said the medium is the message? But this is just another example of how the professions overlap...)

    And in technical writing it is creating the perfect communication, including content and structure, but within the limited scope of the channel you have available to you (whether that be books, online help, training modules, or a web site). You have complete control over the content, but less over the channel and still less over the audience.

    They all address the issue of communication from different angles. And taken together, provide an endless array of solutions and approaches when faced with any problem related to information.

    That's why I like it.

    And that's what I do.

    Saturday, November 14, 2009

    Approaches to Sustainability: Design to Zero


    [Continued from Approaches to Sustainable KM: Embedded KM]


    Another approach to sustainable KM is "design to zero".

    It's impossible to start a business initiative with no resources. Even if it is only your own time and attention, there is some expenditure required. And usually there is a lot more than just that.

    Any new project requires a "bump" to get it started. This might include training, hardware and software expenditures, project management, etc. Any number of capital or resource costs are needed to get things going.

    The problem is that budget planning often only accounts for the short term (2, 3 or perhaps 5 years at most). For projects with a defined endpoint, this is OK. However, almost all KM projects are intended to run indefinitely. (It doesn't make sense to stop sharing knowledge after 3 years, does it?)

    This means that, although it may not show up on the plan of record, the KM program must account for the ongoing maintenance of long-term projects. If you load up your KM program with management of ongoing initiatives, there are two negative consequences:

    • If budgets and headcount are cut (or worse, eliminated), you have no choice but to abandon one or more of the initiatives, usually bringing the program to a screeching halt. Without the expected leadership and constant "push", non-sustainable programs fail when the budget stops.
    • Even if budgets stay the same, after a while, you have no spare resources to start new programs. Even if a good idea comes along (such as Enterprise 2.0) your KM team is fully booked and you do not have sufficient resources to start anything new without impacting existing programs.

    How do you avoid this dilemma? The key is to design each project -- from the beginning -- to reach zero cost.

    That doesn't mean management goes to zero or that residual effort goes to zero, but that the program is designed to become self-supporting at a set point in time.

    So rather than planning for the the first year "bump" and letting maintenance trail on indefinitely, plan from the beginning that management of the program and responsibility for its ongoing success will be transferred to the appropriate people within the organization. Sometimes this means the project may need a bigger expense up front (as soon in the graph below). But the benefit is that the project then becomes self-sustaining and the KM team can move on to tackle other tasks.


    Going back to our example of embedded KM, it is not sufficient to have the idea to invite architects from other disciplines to the project reviews. You need to make sure that the organization running the reviews understands that their success depends on this outside participation and, therefore, they are responsible for making sure it continues once the program is off the ground.

    As with any sustainability practice, not all projects are suited for design to zero. But far more projects are than you might expect. The trick is to look for the part of the organization (usually lines of business) that will benefit most from the effort. Engage them early in the planning, so they feel responsible for its success. Then get them to commit to ongoing management as part of their regular business cycle.

    [To be continued]

    Friday, July 10, 2009

    Approaches to Sustainability: Embedded KM


    [Continued from Sustainable KM: Principles & Approaches]



    The first approach to sustainable KM is fairly obvious: embedded KM. This is where you embed knowledge management practices into the existing business processes. Some of the benefits to embedded KM are:

    • Rapid adoption
    • Tied to business metrics

    An added advantage is that -- if done properly -- there is little or no need for training since there is little or no change to behavior. If you place the knowledge management procedures in the existing operational processes, they will be completed by people as they perform their day-to-day work.

    This is best explained by example:

    • Say your business process already includes reviews at key milestones. It is likely that certain documents are required for the review: project overview, costs, schedule, etc. There may already be a template for these documents.
    • If one of your KM goals is to make employees more aware of other projects to increase shared resources and reduce overlap, an easy way to do this would be to simply collect the project overviews from each review to create a project catalog.

    First, let's look at an embedded KM solution with little or no change to the process. Since the documents are already being created and the review is an existing milestone in the process, the only work needed to implement the KM catalog is creating the repository (a one-time event) and influencing the review managers to post the appropriate documents in it. Working with the managers responsible for the reviews, it should not be a major hurdle to get this slight modification to the standard process in place.

    However, collecting information is only part of the solution. To have any impact at all, the information has to be used. And, quite frankly, a directory full of unsorted documents is neither particularly useful nor appealing to anyone. So to complete the circle there are at least two additional steps:

    • Adding steps to the project lifecycle to remind team members to use the repository. (Again, this is best done by tweeking existing process steps or milestones, such as the steps for starting the requirements gathering, design, and/or implementation stages.)
    • Making the repository easy to use.

    The latter step sounds deceptively simple, but is really the crux of where KM projects go astray from a sustainability perspective. For the repository to be useful, it needs to be searchable/sortable in some meaningful way. The primary way to do this is to provide metadata about the documents; classifications such as industry, country, client name, etc.

    Rarely is the interesting information clearly or consistently labeled in business documents. So the only way to effectively add the metadata is to:

    • Require those contributing to the repository to fill out the metadata for each item
    • Modify the templates used to include specific fields for the metadata (and get the authors to use them correctly)

    Suddenly our simple update to an existing process has become far more complex. Getting people to fill out input forms including metadata is extremely hard; they are either put off by the form and so resist contributing or they fill out the form with incomplete or misleading information. The same problem occurs when templates contain embedded fields. Making sure everyone uses the new templates and uses them correctly is a significant training and communication task, besides the effort required to maintain and periodically update the templates. Finally, the only way to make sure the forms or templates are being filled out correctly is to monitor the submissions to make sure they are complete and meaningful.

    As you can see, a simple "improvement" to the system quickly burgeons into extended activities required to maintain and support the process. Conceptually, you can view a "capture & reuse" program such as described as being positioned somewhere on a curve from no modification to optimized for reuse.

    When there is little or no modification to the content, very little management needs to be applied and the process is highly sustainable. Unfortunately, this also leaves the work of providing relevance up to the users of the content, which is often too much to make the effort worthwhile. (In other words, the system may go unused and have little value.)

    To make the information usable, it needs to be modified to provide relevance. However, doing this requires changes to the process by which it is captured, adding new challenges in communicating the new process, training the users, and encouraging submissions. The content becomes more usable, but significant, ongoing effort is required to enforce and monitor the submissions, since the effort is now put on the contributors rather than the end users. In other words, an unsustainable process.

    The trick is to find the middle ground (indicated by the blue box in the following diagram), where the content is usable enough and the contribution process simple enough that people can and will manage their own use of the system. Experience teaches us that this "sweet spot" can be an extremely narrow segment of the curve and very hard to hit.



    Now, I must confess. I cheated in my previous example. Creating a project catalog from existing documents is often the first instinct. But as I point out, processes managing this sort of explicit knowledge quickly evolve into complex, unsustainable programs from only minor "improvements".

    So let's consider an alternative. Rather than trying to make all project knowledge available to anyone, what if we simply try to expand the current knowledge base incrementally over time? Rather than collecting the review documents, why not include at least one reviewer from an unrelated project to each review? This could be an architect, implementer, or project manager as long as that person can provide an objective, outside view of the project progress.

    This approach has numerous benefits, but two in particular are related to our example:

    • First, from a project management perspective, the outside reviewer helps to keep the project team "honest". It is easy for internal reviews to become formulaic rubber stamp events if those involved are all working on the project.They do not have enough distance to see hidden pitfalls and will resist calling foul on people they have to work with on a daily basis.
    • Second, from a KM perspective, including outsiders gives at least one person a much more indepth and personal knowledge than could ever be gained by reading a set of historical documents with no one to explain them. Another value from a KM perspective is the opportunity the reviewer and the project team have to exchange knowledge, hints, and tips on the fly and in context of the discussion.

    The outside reviewer will take this knowledge back to their own project where it may or may not be used immediately. But it will stick with that person for a long, long time due to their intimate interaction with the other team.

    What is required to make this program work? Initiating the program may be difficult because it requires diplomacy. On the other hand, it involves only a limited number of people. What is needed is to convince management (either of the project teams or of the review process itself) of the efficacy of including outside reviewers. Although there are KM benefits, the real advantage is that the proposed process has management benefits, as described, and will make the reviews more meaningful.

    Once you succeed at convincing them to make the modification to the review process, the program then becomes essentially self-managing from a KM perspective. The project management teams are responsible for ensuring outside reviewers are included and with each review, little by little, knowledge is shared across the organization.

    Again, there are a number of details I have left out that will impact the efficacy of such a program. How outside reviewers are selected and/or rewarded for their efforts can significantly impact the extent to which knowledge spreads as well as the reviewers' willingness to participate. But all of these factors can be addressed either up front or at a periodic (annual?) review of the program, with little impact to the individuals who carry out the plan. The program will continue to expand the knowledge base over time with little or no input from the KM team. The KM team can move on to addressing other issues without being tied up in maintaining the review process, in the best sense of sustainable KM.


    Thursday, June 18, 2009

    The Art of Managing Knowledge Management Programs

    I recently gave a presentation on adaptive knowledge architectures (slides and audio). The presentation was more a case study than anything else. I ended with two slides of lessons learned -- what I would do differently in hindsight to avoid the difficulties we encountered.

    The original slides (and afterthought) are more conceptual than practical. For example, my primary insight was "beware of success". What I meant by that was that, if you succeed, others will not only want to jump on the bandwagon, they will try to take control and alter both the goals and practices to match their needs rather than the original principles.

    There were two things I left out of my slides. One I intentionally left out because it is a more generalized issue about strategy, KM or otherwise. The other I simply forgot until people started asking me questions. They are both practical considerations when managing knowledge management programs.

    The general rule is: communicate continually and repeatedly.

    I must have presented the original architecture (the 3-tier model) 50-60 times when we started. And my boss at the time did as well. It even appeared in a case study written by Microsoft.

    I would then include the 3-tier architecture diagram at the beginning of every KM presentation I gave -- explaining how the new features or changes fit into the architecture. Inevitably someone would ask about the diagram as if they had never seen it. Even people I know I had presented it to within the previous year.

    Richard Saul Wurman makes the point in one of his books that you only learn what you are ready to learn. This is particularly true of strategies and architectures. You have to repeat it over and over again -- until you are bored with it! Because there will be people who haven't absorbed it yet.

    I think this applies particularly when you move into Enterprise 2.0 and web 2.0 where there is a seismic shift of intent and responsibilities. Managers just don't get it. They say they do, but they don't. They are hearing part of it (the rapid adoption part) but not the self-managed part. They think they can pick and choose the attributes without damaging the system. Sorry, it doesn't work that way. In the case study I used for my adaptive architecture presentation, that is exactly what they did to our communities.

    Which brings me to my second point (the one I had forgotten). The reason they took over our communities was not so much that we had succeeded at KM, but that we had succeeded at content management. At the beginning of my presentation I mentioned that the 3-tier diagram includes the top tier (the intranet) so I could dismiss it and say it is not part of the KM environment.

    What I didn't count on was the fact that we built an infrastructure (based on SharePoint) that was so much easier to use and and manage than what was being used for the intranet, management would want it for their "portals".

    Essentially, they were responsible for the last branch of the intranet hierarchy and it took weeks (and several employees) to get content posted. Ironically, I had worked with the organization's IT team trying to sell them on using SharePoint to manage their intranet sites (separate from our KM infrastructure), but it was rejected. (I won't go into that here, but that was an entertaining episode in its own right.)

    But what happened was management saw that our "communities" had all the attributes of their portals but little of the pain. So they jumped on it.

    So what would I do differently? I wouldn't ignore the top layer. I would set aside part of the infrastructure just for them. Even though it isn't really KM, I would do it to create a manageable buffer zone between their activities and our KM processes.

    How does this apply to Enterprise 2.0? I think the same thing will happen to people trying to implement web 2.0 internally. Part of the attraction of social software is its ease of use. And if the technology catches on, people will jump on it for purposes not intended by the software vendors or the sponsors. And once they do, they will try to apply their traditional "management" thinking to it.

    Two small examples I saw at a large corporation:

    • The internal corporate "wikipedia" included pages describing, among other things, each of the organizations: what they did, who ran them, their relationships. Someone discovered this and complained that the entries did not match the description on the managed intranet pages. It was not a suggestion or a demand. It was simply stated as what they saw as a fact; that the wiki pages should be made to match the intranet pages. In other words, selected content should be controlled.
    • On our social networking site we got management to promote the site to our organization. It was then pointed out that the managers themselves should create profiles. Several did. But upper management insisted that their executive assistants draft the content and that we should have a way to post these "ghosted" profiles (completely in opposition to the basic model and implementation of the application). We did change the software to permit this, but as innocuous as this sounds it does create a philosophical schism in the application: who gets to have "ghosted" profiles? How can people tell real from edited profiles? etc.

    These may seem trivial, but our apps were still in birth mode, without widespread adoption. So these were just the first signs of the problem.

    The question is: what would I do about it? Like I said in the presentation, I don't think there is a generic answer. It depends on the corporate culture and, I'm afraid, the individuals involved. You can try to second guess the culture. So, for example, in hindsight I would have created spaces for the HQ intranet sites to try to alleviate the pressure on the communities. Would that have worked? Possibly. But it is just as likely that they would then argue that the communities are unnecessary since their portals provide all the information consultants need. (Which, in fact they did when they argued that we should dissolve all of the communities that they didn't control. But we managed to stop that effort...)

    In the case of Enterprise 2.0, I would have suggested creating one or more executive wikis, secured for use by managers of the individual organizations. This may have satisfied their need for secrecy, ease of use, and taught them a little bit about the operating principles of web 2.0. Would that have sufficiently distracted them from messing with the employee wiki (which was the real goal)? Perhaps, perhaps not. But it would be worth a try.

    Another recommendation would be to have a very clear business objective for E2.0. For example, internal development blogs for each project/product. You can allow other uses of blogs, but by having a clear, measurable, but not ROI-based, objective and repeatedly stating it (i.e. constant communication as stated above) you may be able to deflect people trying to commandeer the program.

    I could go on, but I better stop before I end up writing War & Peace...


    [Many thanks to Steve Ardire for inspiring this post and Stan Garfield for teaching me much of what I know about managing KM programs.]