I found it really interesting to read Alistair Hall’s article in Wilderness Magazine, which describes a recent Official Information Act Request to the Department of Conservation regarding how it estimates the track walking times that are typically advised on signs around the place. If you’ve spent any amount of time in New Zealand’s conservation estate track network, it’s likely you’ve experienced inconsistencies in the accuracy of such times. It shouldn’t really be unexpected, given that every sign presents a single time whilst people clearly walk at different rates. No stated time can be correct for everyone.
Most interesting for me from the article, I think, was not the revelation that estimated times aren’t pulled out of thin air so much as the mish-mash variety of different methods that various conservancies use. Some conservancies have slow staff walk the track and time it at a leisurely pace. In the Marlborough Souds, a fit staff member walk the tracks, then times would be weighed slightly downwards… unless they were shorter walks in which case they’d find a staff member near retirement. (No mention of how fast this older person might be—many “physically mature” people can still crack it through the outdoors pretty fast.)
Some conservancies make informal guesses and then consult staff from their experiences. Times on signs around Egmont National Park were estimated by paying a contractor to walk every track in the park and time distances between signposts, and the times were then adjusted depending on the track type. Conservancies throughout the central North Island use a formula based on the Naismith Rule of thumb, basically looking at a map and taking hill ascents into account, and then weighting the results further for the kind of track and expected type of visitor.
None of them, however, allow for variation in the type of person who might be walking on a track. They all produce a single time for the signpost next to any stretch of track, and typically don’t indicate who it’s for. Irrespective of how much calculation occurs for variations in the track, an assumption is always made about the visitor. It might be an assumption that people are very fit, or alternatively that the visitor could as easily be pushing a baby stroller, but the signs don’t say.
Reading the article provoked earlier thoughts, because for some time I’ve introspectively assumed that there’s not a linear relationship between something like “fitness” and times. It’s not always possible to just take a measurement for one person and multiply it by a constant for yourself, because you might adapt differently from that person depending on the condition. For example, I could name several people for whom I walk roughly the same speed on the flat, but as soon as we’re going up-hill, I move faster. If a DoC sign refers to a track that turns out to be all flat, we’d need to multiplying by the same constant, but if it turns out that it’s partly up-hill and partly flat, the constants will be different for part of the track and not the rest of it. Problematically, it’s not clear how much of each they should be different without knowing how much is flat and how much is up-hill.
Even if impractical, it could be a fun experiment to have sign-posts that don’t merely guess at a time, but indicate the type of track or route and how much of it there is. If I knew there would be “5 units” of the sort of hard flat track on which you’re only slowed when trapped behind a platoon of people with giant umbrellas walking their dogs, and “12 units” of a steep and slippery up-hill scramble, I could apply my own personal multipliers to each of those sections to guestimate my own personal time. Anyone else could do the same, but with their own multipliers.
I’ve run into this same sort of issue recently when trying to design the board and player characteristics for the Tararua Board Game experiment (with which I’m being eternally lazy), and how easily players could move around the board, but I’m starting to wonder if it’s not worth the bother in favour of keeping the rules simple.
I’m sure the whole thing could get complexified with many different variations of tracks and routes and rivers, adjustments for weather issues, and so on, and then there are things that can’t necessarily go onto signs at all such as a track’s current short-term condition. That said, I do still wonder if, with the right selection of variations, there could be a more practical and flexible way of estimating track times for wildly different kinds of people and groups.
There are also other approaches. When I mentioned this on twitter, Simon McAuliffe promptly responded to point out that he’s written his own personal GIS software that calculates times, optimal routes, required energy, and which can be tuned to individual people or groups with some empirical GPS input from their experiences. (Show off! :-P)
DoC’s response to the OIA request included a note that different conservancies would certainly like to work towards a more nationally consistent way for estimating track times, which is promising news. Clearly there will always be exceptions, and it might be that actually looking at a map with some presence of intelligence could still sometimes be more reliable than an indication on any sign. Also, hopefully obviously, having a more reliable estimate based on someone else’s measurement of how long it’ll take to cover a particular distance is no substitute for taking appropriate back-up measures (like portable shelter) in case it doesn’t work out.