As a follow-up to Poll: how do you use tracktype? I’d like to gather how to actually grade a sample. For the sake of this poll, lets call the pictured road a track, it might be an unclassified, but never mind.
tracktype is shown in picture above?If inclined to spell out deliberations, please do.
I am unfamiliar with this type of terrain, so it is really hard for me to answer.
It looks like hard flat baked mud, that will slowly crumble into sand and will turn into mud once rains/floods. Is it a correct description?
How often it dries/floods? Is it like depicted 5% of time? 90%? 99%? 99.9%?
Option 6: I don’t use tracktype=grade* because of this type of ambiguity and how it is completely unclear how to tag based just on reading the tags themselves, without looking up a Wiki page or Forum threads for examples 
smoothness=* was hard enough to get an understanding of, and is similarly temporal.
So am I. From looking at the picture, this looks very firm. Driving with a lorry will not leave much of a trace. perhaps like in all all of this place , from what I observe from the photo, a.k.a. a casual observer with no other local knowledge than what location presents at the very moment in time.
in such case I do not get reason for making this poll
we have enough of real tagging problems to waste time on imaginary ones
(if someone would be familiar with such place and looking for advise how to tag place they surveyed and how to adapt tagging schema to their area - it would make sense)
To me, the main point of tracktype is to let you know if your wheels will sink into it and you won’t be able to proceed easily (Or at all. Or even back up, in extreme cases).
If it varies during different times of the year, I’d mark the “worst case” scenario that you’re likely to encounter (in my little part of the Europe, it is usually heavy rain or snow thawing making ground muddy; esp. combined with a heavy vehicle with low clearance or one with few and thin tires). E.g.:
tracktype=grade1 (e.g. surface=asphalt or concrete usually behaves like that IME).surface=dirt becomes surface=mud), then it is tracktype=grade5 (that is often the case if no human intervention was done on many soft natural surfaces over here)tracktype=grade4, etc.which is IMHO why “how much hard elements are mixed in with soft ones” is mentioned in the criteria, because for very many situations, tracks with more hard materials (e.g. gravel) in the mix will produce more firm surface generally.
Of course, some things (e.g. surface=sand) might behave in other direction (i.e. be more firm when somewhat wet then when fully dry), and some (e.g. surface=gravel) are likely to be about the same in both best and worst conditions, i.e. about tracktype=grade2[1]
There is also always an option of using tracktype:conditional=* and surface:conditional=* etc. if you want to more precisely define how the value of the tag varies in different conditions. And/or add tracktype:note=*. But vast majority of mappers won’t bother, as single value is good enough for them.
Also I’d note that there is no need to obsess over tagging. If the certain tag makes a sense to you at a certain time, add it. If it doesn’t, nobody is forcing you to add it. Same with processing. If some tag does not help your use cases, you’re free to ignore them. You’re also free to make your own tags if you think they’ll fill the niche (e.g. MTB surface looseness tag. Loose sand, swamps and so on).
There is also IMHO no need to go around finding fringe cases just so one can say “Ha! I told you so, your tagging ain’t perfect, it can’t properly and fully define this example!”
We know.
supposing that the underlying support surface is firm enough. Adding a shovel full of gravel on top of mud obviously won’t help much as it would just sink – unless you add so much gravel that there is no more mud ↩︎