When I get back the page has expired and I have to start the form again from scratch. Same thing. I’m at a loss. Okay I’ll try again tomorrow. Same thing. Maybe it’s my browser? I change browser. Same thing.

Then I look at my file name for the PDF. Maybe it doesn’t like non-alphanumeric characters or spaces? I remove them. Same thing. Okay maybe it’s the length of the file name? Maybe it has an undisclosed character limit, and if you exceed it, it can’t read the .pdf file extension? I shorten the file name to seven characters. Boom, upload accepted. Submit application. Fail. I have exceeded the time limit and have to wait 20 minutes for another attempt. 20 minutes later I submit like a dream. I have figured out its mysteries. I burn with zeal to advise others and wonder if I should write the steps down so I don’t have to go through this again. Two days later I get an email informing me my e-visa is granted. Seven days later I receive 257 emails acknowledging receipt of my original application and assuring me it will be processed within two days.

I’m now in Delhi. I wake up in the middle of the night. I need to pee. I grope for my phone in the dark and switch on its torchlight. This helps me navigate in the dark the four meters to the light switches on the wall by the door. I need to do this because there is no way to control the lighting from the bed. I have to experiment with the switches before I find the one that turns on the light. This is because some switches control the ceiling fans and some the aircon, but there’s no way of telling which is which (even in daylight).

Eventually I have light and can get to the bathroom. This is a luxury hotel, and the bathroom is enormous. The light switches for the bathroom are another five meters away from the entrance, right next to the toilet. My phone gets me there (and back). By now I’m fully awake and I reflect that it might be simpler for me just to use my phone light and not bother with switches at all. Later I wake up hot and sweaty, and figure out that in messing with the switches I have inadvertently switched off the aircon.

Two days later I’m in another five star hotel. I’m hungry and tired. There’s a handy online room service menu accessible by QR code. I browse, select, my choice goes into the basket, I review the order, confirm the details, select “immediate delivery”, submit, and get the confirmation screen. Two hours later, there’s still no food.

Then I remember I’ve been in this situation before, in Manila. The app there was not connected to the kitchen, so you have to choose what you want on the app, and then call room service to place your actual order. Nothing in the app or the room explains this. In Manila a helpful staff member had explained it to me on the phone while also telling me my room had the QR code for an outdated menu (a new QR code was delivered to my room shortly after).

I go to the room phone. It’s an ordinary phone. There’s nothing on the phone to tell you how to call room service, front desk, anyone. I experiment with a few numbers, nothing. Zero appears to get me to an outside line. By this point I really don’t want any human interaction at all. I give up, raid the minibar.

These are all examples of what I call lumpy infrastructure. Smooth infrastructure eliminates uncertainty and gets you from purpose to outcome without interruption, undue effort or diversion. In fact, it’s so smooth you barely notice it. Lumpy infrastructure may have all the components in place and each component might be of high quality. But they are assembled wrongly, or they have missing links, or they rely on private rules or conventions that are not available to the infrastructure traveler. It is possible to traverse the infrastructure but only through exercise of imagination, experimentation, experience, or by enlisting the help of human assistants. In process and system design we use customer journeys to identify lumps and smooth them out, before they hit real people.

Now the examples above are very simple services that those of us nurtured in smooth infrastructures find deeply mysterious and often frustrating. Visa application processes must be traversed by millions of travellers – how could this not be fixed? How do other people cope? Luxury hotels that make it dangerous to get around in the dark or extremely difficult to access advertised (and revenue generating) services through the absence of one critical link – if not anticipated at the design stage, how do they not get fixed?

I am not sure of the answer. One colleague I spoke to speculated that India is a culture with a high sense of social responsibility to provide employment. It must balance that with the desire to appear modern and capable. A fully smooth infrastructure removes human agency from large sections of the process. Lumpy infrastructure requires human support to help travelers traverse critical gaps and dark sections. So you upgrade the components to look modern but you deprecate the task of joining everything up, to the extent that you have to lay people off. That seems reasonable in the room service case, but not the light switch or e-visa cases. On those, I remain in the dark.

However I think there is also a cultural payoff to this phenomenon. We believe that there is almost always a path through lumpy infrastructure, because it is reasonable to assume that many people traverse it successfully every day. In order to traverse it successfully we need to exercise our imaginations, improvise, and solicit assistance from people nearby, to an extent that is not normally required in smoother environments like Singapore or Ireland, where I spend much of my time.

When true uncertainty intrudes on our lives, it is not a bad thing to have such faculties ready to hand.

Also good news, friend and colleague Paul Corney has won CILIP’s 2019 K&IM Walford Award for his outstanding contribution in the field of Knowledge Management.

For those who are curious, the second edition of The Knowledge Manager’s Handbook involved a complete review and update of the first edition, with additional chapters on KM standards, the links between KM and digital transformation, AI and Big Data, and guidance on working externally and building professional capabilities. There are also new case studies from NASA, Public Works Department Malaysia, and PDO Oman. And if you want to get a special 20% discount, you can order directly from the Kogan Page website using the discount code FBM20 at checkout.

Nick Milton recently surveyed his KM contacts on LinkedIn and found that knowledge managers average lifespan seems to last about 6 years – which shows progress! About 60% will survive beyond 4 years, double my figure a decade ago. However, only about 25% of his subjects are likely to have had a longish (8+ years) career in KM. Progress, but slow progress!

In June this year, Patrick sat down (virtually) with Vivs Long-Ferguson of our partner Synaptica for an interview. Among the things they spoke about was how Patrick viewed the KM and taxonomy landscape. Here’s the transcript of that interview.

Many years ago, in one of our projects, we discovered that a team of three young people had been running a national event for their organisation for the preceding three years. The event was growing in popularity and profile, becoming what they termed a marquee event. When we interviewed the three young people, we discovered that they had basically figured out how to run the event through trial and error. Between them they were able to recall and explain how they managed the different aspects of the event, but all that knowledge existed only in their heads. It was clear that there was a risk of knowledge loss, more so as the organisation had a high staff turnover rate. To mitigate the risk, we embarked on a programme of documenting their method knowledge.

Here is another example, this instance of how Cargill’s global social media team manages their method knowledge. Thanks to Laura Ostenso for bringing the article to my attention.

What method knowledge do you work with? And how are you managing it?

So you have built and implemented a taxonomy but search is still not returning the desired results. What do you do?

Taxonomy alone is limited in what it can do. Search alone is also limited. Together, they become much smarter. If taxonomy and search are integrated, they can be very powerful and vastly improve the user experience.

In this article, inspired by the work of Callender, Morville and Nichani (see references in the paper) Patrick outlines 10 search patterns that taxonomy and metadata can support. He lists their benefits, dependencies, potential applications, and illustrates them with real life examples.

Taxonomy and Search Patterns v3.pdf