Proposal talk:Mtb:scale

I have added a link. German users will be very upset if we don't include that singletrail-scale somehow. It's very well established in the German speaking parts of Europe.--Extremecarver 16:52, 20 November 2008 (UTC)

What scale(s) to use?

I would suggest making two mtb scale keys: imba_scale=* and sts_scale=*. This way a trail can be classified according to both systems without conflict and without leaving one out. Alternatively, OSM should pick one and stick with it, even if it upsets a small subset of users. I'd suggest the imba_scale for that. In fact, if OSM decides to go with one or the other, it should still be referenced by name, i.e. with the imba_scale or sts_scale; similar to what is done with sac_scale=*. Creating a new hybrid scale is the worst possible option.--Hawke 18:20, 20 November 2008 (UTC)

Can you start writing the proposal somewhere and move those comments over here, while still keeping a link from this page, so we are not too much mixing things ? Sletuffe 18:22, 20 November 2008 (UTC)

Well ooops, maybe something is not clear in my mind, what is a singletrail ? and does it oppose to multipletrails ? what are you talking about ? Sletuffe 18:31, 20 November 2008 (UTC)

A singletrail is by definition of mtbikers any way where only one mtbike fits in. So you can't ride side by side. Maximum width smaller than 2m. There is no fixed width, it's rather subjective.

Well if we use two scales, then we have to define the overlap, and similarities. I think a hybrid scale is not so bad. The IMBA scale is pretty general, while the singletrail scale is restricted to singletrails. What about mtb_scale= with values easy-to-extreme (or similar - I would rather stick to numbers in any case, it's easier to remember in my opinion) and mtb_scale=S0-S5. If users want to give both value systems than they could write for example mtb_scale=S1;3 or mtb_scale=S1;intermediate. Having scales named IMBA_scale=* or sts_scale=* is even worse, because it will be harder to remember for people previously not knowing the scales. --Extremecarver 18:36, 20 November 2008 (UTC)

We don't have to define the overlap; the overlap is defined in how the scales themselves are defined.

As I see it, here are the options, in order of my preference:

• Use only one of the two scales and call it mtb_scale. This has the advantage of being well-defined, but would bother people who are greatly attached to the other system.
• Use both of the scales, on separate keys. This has the advantages of being well-defined, but would require extra tags, and extra understanding in programs wishing to use the data.
• Use a hybrid/otherwise new system. This is IMO the worst option, because it's effectively creating a whole new scale, which no one at all will know.

We don't need to worry about people who don't previously know any scale, because they're learning a new scale in any case: either the STS scale, the IMBA scale, or the OSM scale.

--Hawke 19:12, 20 November 2008 (UTC)

I'll try to work something out. We need different specifications for up and downhill anyhow! Ot the same specification of the definition but different consideration wheter it's up or downhill, as in general going downhill is easier than going uphill.--Extremecarver 19:46, 20 November 2008 (UTC)

I still think it's best to leave the defining of a difficulty scale to the (relative) experts who defined the IMBA or STS scales. We don't need to create our own system just for its own sake. And either of those other scales are likely to get wider acceptance than one created at OSM. Especially if that system is "STS, but with values for 'off the scale' in either direction". Better just to simply use STS. --Hawke 16:01, 21 November 2008 (UTC)

I agree with Hawke, above. Best to use existing scales, and to state which scale is used. Climbing guidebooks have similar issues: different locations and different climbing styles use different grading systems and it is only possible to do approximate conversions. Guidebook editors will use a mix of systems to ensure that they are using the grade assigned by the expert who graded it, they also make it clear which system is being used. The guidebook author will never invent a new system. When a reader wants to climb in another country it is up to them to understand what the different grades mean. Also, even if you invent a new scale or choose one official scale, it is still useful for people to have a means to record other scales on the mapping database because they might be necessary for some mapping applications. Bruce mcadam 18:00, 25 November 2008 (UTC)

mtb_scale VS mtb_difficulty

Since I have unilateraly changed difficulty to scale, I'm re-proposing it to stick with one or the other Sletuffe 18:41, 20 November 2008 (UTC)

I'm indifferent to this. Both have Pros and Cons.

downhill / uphill

Only one tag for both

I'm willing to merge the mtb_scale with other scales for downhill and uphill by using a name space systeme such as :

• mtb_scale:downhill=easy
• mtb_scale:uphill=more_difficult

what do you think ?

Another solution might be to simplify greatly by supposing that, if it's easy to go downhill, then it is more_difficult to go uphill. Things would be more simple, but I'm not sure that we can take this shortcut Sletuffe 18:45, 20 November 2008 (UTC)

I don't think we can take that shortcut. Some ways are easy to go down, but nearly impossible to cycle up. Others are hard going up and down... Also we have to cater for mtbiking in fat regions, or for flat :sections, that might be equally challenging. While merging downhill and flat makes some sense to me, I can'think of merging uphill with flat or downhill. Regarding mtb_scale vs mtb_difficulty I'm indifferent.

We would off course define, and that might be best:

• mtb_scale=easy (flat) (for a way that is flat),
• mtb_scale:uphill=easy (uphill easy) + mtb_scale:downhill(moderate) (for a way that is not flat).

The difference between flat and not flat should be used from that moment where it makes significant difference.--Extremecarver 19:42, 20 November 2008 (UTC)

I start to like this one very much, I'll modify the proposal so people can say what they think Sletuffe 23:39, 20 November 2008 (UTC)

IMO, it's a very bad idea. The considerations by which we measure trail difficulty must be the same whether going up- or down-hill. e.g. "obstacles up to 10cm in diameter, loose soil, 500% grade". Those will not change based on the direction you're going, unless the trail takes a different route up vs. down. And if that's the case, you must simply tag the separate routes anyway. --Hawke 15:58, 21 November 2008 (UTC)

No the way itself won't change of course, but the importance of the objects regarding to the difficulty does. The German Singletrail-Skala which I used for mtb:scale:downhill and mtb:scale is not considering any uphill parts - they said that they see no need for it, as uphill will be largely influenced by your fitness. I stuck to the STS as well as possible, but felt the need to put in one category more for the beginning (which is thereby also closer to the mtb:scale:uphill or IMBA), and put one category on top which is "carry your bike" because we have no other way to express it.

As for the mtb:scale:uphill I tried to copy the IMBA as far as it makes sense. The problem with that IMBA scale is however in my eyes, that it is not very well thought at all. They completely leave out any difficult downhill trails from their rating. And no, why should the rating be the same for uphill and downhill, or the way categories. Any downhill rated mtb:scale:downhill grade4 or higher is in my eyes impossible for cycling up the hill. loose soil downhill doesn't matter very much, you just have to be a bit slower. Loose soil uphill at 20% is the end, because you're rear tire whill spin and you have to step of your bike. Anything steeper than 30-35% is simply not bikeable uphill, but downhill 70-100% is with some experience not so difficult. Or if there is a 2m high step going downhill a good biker will simply take it, going uphill it's unclear wheter it's possible to pass at all or not? We could try to get the values of downhill and uphill a bit closer together however.--Extremecarver 16:27, 21 November 2008 (UTC)

Hrm...It seems to me that there are different goals here: On the one hand, there's "how hard is it to use" which seems to be what you're going for, and on the other there's "what's the trail like" which is IMO more important. The latter is also why I feel that it's better to use an existing scale that some people will be familiar with, rather than creating our own. Taking your example of loose soil: If I know that, say, an STS scale of "S2" means that there's probably loose soil, I know how difficult it will be: downwards, I just need to go a bit slower, uphill it will be much more difficult. Even worse, there's the fact that there's no good way to portray "uphill" and "downhill" in OSM, unless you suggest that we should have the way always pointing uphill or downhill (bad) and have to split the way every time it changes from up to down, and vice versa (very bad). All in all, it greatly complicates the tagging for very little benefit. --Hawke 16:46, 21 November 2008 (UTC)

Well I don't think you need to define what's uphill or downhill. You should just put an overlay of SRTM lines over it, and you can easily see whether it's up or downhill. The advantage of one tag for both is that we would simply need to put mtb:scale=gradeX but we then would not know how easy or difficult it is to go up or down. mtb_scale should IMHO be a difficulty rating, where the difficulty is based on the respective surface. I first tried to put in one value to have them all. But when I used the STS scale there was simply no equivalent in the IMBA scale for going uphill, and vice versa. Try to create an alternative system based on the STS and include difficulty ratings for going uphill. I can't see this to work. There's simply too much difference. Let me give some pictured examples.

• 1. Based on STS this is S5, there is no IMBA definition for this picture, uphill even carrying your bike becomes very difficult : :::http://www.trailhunter.de/gallery/img5671.html
• 2. I would put this picture as STS S4 (again no IMBA definiton matches properly) http://www.trailhunter.de/gallery/img5663.html
• 3. This would be no more than STS S1 or STS2 but going uphill is nevertheless impossible without stepptin of your bike http://www.trailhunter.de/gallery/img5675.html
• 4. This again is really easy on downhill, but on uphill carry your bike STS1 or STS2 http://www.trailhunter.de/gallery/img5678.html

I don't see how any system based on "what's the trail like" can classify those ways in such a way that I know how it will be like mtbiking there. Personally for me with the current grading scale system I would would understand it however. For going uphill I wouldn't see any difference from 2.-4. however. On each of these ways I have to carry my bike. Did you understand te concept that for any way that declines one should give BOTH values i.e. mtb:scale:uphill=grade4 AND mtb:scale:downhill=grade2.? Or take a more extreme example. A graveled forest rode with a slope of 30%. That's a S0 or S1. However cycling that very easy way uphill is impossible. Now take a way that's just flat. It might be STS 2-3 but you may still be going in both directions, the steeper it becomes the harder it's to go up, while going level or going down you would not increase the difficulty rating.--Extremecarver 17:34, 21 November 2008 (UTC)

I'd mark them as: S3 or S4 for the first two, and S2 for the second two. IMBA scale "double-black" for the first two, and "blue square" for the latter two. The classification is not affected by whether it's up-or down-hill. You have phrases like "Stairs and flat stages are to be expected" and and "unavoidable obstacles 8 inches tall or less". This means that it will be somewhat easy going down, and quite hard going up. So you do know what it would be like mtbiking there. (IMBA scale is not as helpful in that regard though, because the stairs are unavoidable obstacles that only apply in one direction)

Hrm, thinking about this further, it strikes me that the STS and IMBA scales are kind of for totally different purposes. IMBA seems to be more geared towards mountain-bike obstacle course things (note how "TTFs" are rather a big feature of their scale, and see [1].) STS is more geared towards more natural trails. This suggests even more that it would be good to use sts_scale=* as well as imba_scale=*, because either system is poor for rating the other kind of trail. --Hawke 18:33, 21 November 2008 (UTC)

Your IMBA assessement is exactly why I'm thinking it's not really good to have one tag for both directions. From my viewpoint when looking at the map I want to know wheter I can go uphill or not, or wheter I can go down or not, and wheter this will be difficult for me (I would like that) or easy (this might become boring). I don't actually mind what the trail is made up from exactly. Seing IMBA blue square definition only I would think, yeah maybe I can still go up here. I don't think we need TTFs really. One should better make a own bikepark difficulty scale for them. I mean a TTF won't matter me in my decision wheter I can go that trail over a mountain or not, it will only exist in purposebuilt tracks, and those usually are well documented either on a webpage of the bikeparkorganiser, or even distributed on paper map to bikepark users, documented on blackboards..... TTFs are more a thing needed in a guide, not in a map. Also TTFs might be taken down in winter, changing place rather often,.... That's why I only partly copied things from IMBA for mtb:scale:uphill.

IMBA vs. STS has nothing to do with up/downhill. They're just different from each other, for rating different things. IMBA scale is for something closer to trials bike than mountain bike. Personally, I would find the IMBA scale to be nearly useless, since anything over "easy" is too difficult for me (8-inch unavoidable obstacles!?) and "TTFs" are not something I've ever used, nor do I have any interest in them. You can determine whether you can go uphill or not purely by using the STS scale -- though of course it depends on skill. S0 and S1 are probably navigable by a not-particularly-skilled mountain biker, up or down hill. S2 should be doable by a skilled rider uphill, and by a moderately-skilled rider downhill. But my point is this: You don't need a different rating on the up- and down-hill. You always know that it will be more difficult to go up than down. --Hawke 19:37, 21 November 2008 (UTC)

The sts_scale defines the difficulty of a path in flat or downhill direction (see discription on the sts website). The MTB Rider itself can decide if he can ride a S2 uphill or not by his experience with S2 ways down or by the scale itself. That would work for most. But the MTB rider must be able to find out somehow if a path goes up or downhill in the direction he wants to use. How is that do be defined (only needed if we add a uphill scale) ? The conclusion is that the sts scale or the discussion about it is somewhat independend from to be discussed uphill scale - if needed. There doesn'T seem to be one existing except the IMBA scale. About splitting up STS0 I think that doesn'T make sense. because STS0 is also a track of tracktype grade 2. http://www.singletrail-skala.de/images/s-grade/s0_3.jpg ) the picture of the different tracktype speak for themself. So something below mtb scale grade 2 would be a tracktype grade 1 (paved or cobblestone). So useless. If it is somehow possible to start the mtb scale (i would suggest: mtb_stscale instead also ) with 0 like S0 it is easier to adopt the numbers without causing confusionb etween the grade x and stsx number. --mightym 12:45, 22 November 2008 (UTC)

I suppose that every one who wants to setup a mtb map will also add contour lines, so direction of the slope might be guessed by this mean.

Your sentence "because S0 is also a track of tracktype grade 2" might be wrong, because track implies a minimum width (well it's not writen, but it is to me, since it is a track) while the sts scale seams to cover "single trails". And also because a grade2/3/4/5 tracktype is also a "kind of STS0". There are no bijection here.

we are proposing mtb:scale=value, but that's true that "scale" might not be self explanatory maybe mtb:sts_scale=value would be much consistent to sac_scale. I don't mind very much. But if we end with only one scale for mtb, I'll be happy to save 4 characters by having scale instead of sts_scale, but well, I can go over that and get more precise by adding just 4 characters Sletuffe 12:18, 22 November 2008 (UTC)

STS is not exklusive for single trails. But only STS0 could be used for highway=tracks of tracktype grade 2-5. But tracktype 2-5 automatically implies STS0 (see the linked picture). So if someone doesn't add the mtb scale, or however it is gonna be called, to any highway=track tracktype grade 2-5. It won'T care for MTB Riders cause they know it is STS0. I would also be happy if we save 4 characters. the wiki explaning the grades will be self explaining. If someone adds another scale sometime in the future he will have to use x addtional characters. ;) So conclusion: No grade below STS0 and no additional direction information --mightym 14:00, 22 November 2008 (UTC)

From a "to be simple" point of view

Looks like you've come to a level I can't help anymore since I'm a "sunday mountain biker" and will probably never tag anything after grade2. While hiking, I won't tag it either (for now) because I don't have the level to make a classification while not being on a bike. So I'll be tagging rather flat tracks or slightly steep path with mtb:scale= Then, having 2 scales for up/down will probably make me lazy and I will probably just tag : mtb:scale=grade1 since I don't see the point in tagging mtb:scale:downhill=grade1 + mtb:scale:uhill=grade1 which would be the case for 95% track/path I'll be biking on. So my point is, that if we also want to attract "sunday mountain bikers" like me, we'd better find a uniq scale for both up/down even if that leave a few information appart. And why not, create a second proposal for downhill being more specific.

So to say, something like :

• highway=track
• mtb:scale=grade1

AND optionnaly, if you want to be more specific add :

• mtb:downhill:scale=grade1 (but part of another separated proposal that we could link to from this page )

Sletuffe 10:53, 22 November 2008 (UTC)

While re-reading me, I fear that my needs are not covered by this tag, but still, my point stands ! I am just remembering the flat track I recently mapped, even if not usable by a racing bike, it is obviously usable by a mtb, and that is an information I have allready put in the smoothness=bad tag, or worst, that is allready implied in a track, so I probably won't tag mtb:scale=grade1 since it is obviously the case allready. So I come to the point : maybe we are constructing a extrem scale for mtb but not a scale for my current use. (And all this is not only a question of "Me" because we have to suppose many others are in my case ) Sletuffe 11:01, 22 November 2008 (UTC)

using a full name space system

I have transformed all pages related to mtb into a "name space system" with ":" separators to group all in a mtb:XX:YY=value.

This is just an idea, don't flame me, I just thought it would be more consistent to the Ski proposal, and since I find their work very great, we could do the same. However feel free to explain me why you wouldn't like it Sletuffe 00:46, 21 November 2008 (UTC)

Choice of values

I have no good explanations about best values to use, but I'm a bit worried with gradeX (Am I too much against tracktype and it makes me think too much about it ? ) but I'm not in favor of : numbers.

• If, for any reasons someone comes later with an in-between grade5 and grade6 we cannot do much about it (grade5.5 ?)
• harder to remember, if editors deals with it, no problems, but since we are mostly doing it on a human way it is just a little better to remember IMHO that easy is.... well, easy ! than grade3 is easy.

But I don't have much better counter examples Sletuffe 14:01, 21 November 2008 (UTC)

I prefer gradeX because it seems to work quite well for tracktype (well in the sense that many people use it). I also though about using numbers only instead of gradeX just X. As the key is mtb:scale(:uphill/:downhil)=* it is quite understandable with a number alone I think. With tracktype=* it wasn't, because why not put tracktype=footpath etc... A scale is a scale. I would agree that if you only have easy, intermediate, advanced, expert than that makes more sense, but putting 8 different levels of difficulty into words would mean we are running out of words, and I don't like qualifiers like very. For inbetween we could eiter use grade5.5 or grade5:+ grade5:- or grade5+ grade+- (with grade5:+ or grade5:5 being much better identifiable for the renderers). With the current system of mtb_scale I think it's safe to drop mtb=yes and mtb_uphill=*. I think with numbers giving inbetween values is much easier than with words. Because what would you want to put inbetween easiest and easy? very easy? --Extremecarver 14:24, 21 November 2008 (UTC)

I knew you would say that, and that you would point me smoothness=* to talk about very_* . Just a fast word of history: I have proposed the very_ but I wasn't in favor of it. I asked english speakers for words, but english is poor as if I had used french, I would have provided the words. But we came across the same problem you are facing now in germany with mtb=yes : people have used the scale BEFORE it was approved, and for sake of simplicity we kept the original tagging in wich we included the very_* stuff. Now that I have the chance to start over, I will refuse very_* or more_* or extremly_*. Now about words or numbers, I just have repeated stupidly what has been in used in many place (smoothness, sac_scale, trail_visibility ) but I have to go over that, and think for myself. In the real world of scales things, at least in france, we use numbers for climbing, paragliding, hiking but we also use words for skiing, via-ferrata, hiking. In every case, the numbering schemas are used for high accuracy and precision, and words for simpified scales for unused people to it. In osm data base we should stick to one. And the answer comes to my mind by what I have said in smoothness=* and by what others have told me :

We can convert a high accuracy scale to a lower accuracy scale, but not the opposite

This definetly gives my answer, you are right, let's use numbers Sletuffe 14:55, 21 November 2008 (UTC)

I would choose the numbers. It is more 'universal'. If the STS scale (www.singletrail-skala.de) is adopted in principle, where should numbers start to count with in OSM? I think it makes sense to start with S0 = grade 0 (or only "0" it the word grade is not used). so that numbers of the STS are identical to the OSM numbers. --mightym 14:30, 22 November 2008 (UTC)

I'm in favor of dropping the "grade" just 0 to 7 might be enough and ok to start from 0 (and because humans are lazy) Sletuffe 00:40, 25 November 2008 (UTC)

impossibility value?

Should we include mtb:scale:uphill=8 or mtb:scale:uphill:no/unclassified to have a proposed value for a way that can't be passed with a mountainbike at all, like a place where you have to climb and use both hands and have no possibility to carry your mtb?

Mapping to sac_scale

Ouch, looks to me you are a bit extrem !! what sort of biker are you ? You wrote : "grade7" "Pushing the bike is largely impossible, the mtb has to be carried" = sac_scale="demanding_mountain_hiking" T3 and upwards. But demanding_mountain_hiking said : "exposed sites may be secured with ropes or chains, possible need to use hands for balance"

Can you still carry your bike when you need your hands for balance and you need your hand to grab ropes or chains ?

I am possibly missusing the sac_scale then, but I am tagging path with "demanding_mountain_hiking" where you definitly need both your hands to grab a chain, or stairs, or rope.

Here is a foot path I have tagged with "demanding_mountain_hiking" [2] Sletuffe 14:12, 21 November 2008 (UTC)

I myself have carried my bike up things like on that picture. If it's only for 10-20m I feel quite allright in doing so. I have once even seen someone with the mtb attached to the backpack in parts climbing up a climging grade 5 for about 300m vertical. I think the two highest grading should be: 1. impossible to bike but you still may carry your bike. 2. Carrying is impossible. I wasn't that happy about the comparison with T3, but T2 was too easy. So like T2+.

Then, you are an extrem biker, carrying a bike is not biking to my mind, and if the guys you are talking about did climb with a bike in his back, why souldn't we say grade7 is "like" "difficult_alpine_hiking" (T6) or climbing after all ? If I'm climbing with my son on my back, I wouldn't consider that this climbing wall is level 0 for babies, so no, I don't agree with you comparison to T3. Back to your very important statement about smoothness : nothing is ever impassable (for bikes), but we need to define a "stop" and that stop, is, to me whether or not you can be on your bike to pass this way or part of this way.

As of T2, I have mapped trails with it that I think no one will pass with a bike uphill, and would seriously take risks by using it downhill, but I'm not an extrem biker Sletuffe

I'm neither. But I like alpine mtbiking. Most summers I spent in the Valais, where I often go to about 3000m with my bike. I can't bike down more than mtb:downhill:grade4 myself, but I know there are people who can so I don't want to discrinimate against them. Examples for the higher up downhill categories like in the pictures [3]. Off course most mountainbiking takes place in the easier grades. Of course I said before that no way is impassable, but that was more ment for the ways regarding the smoothness discussion. I think it does make sense to say here you can still bike, here you have to carry your bike, and here you can't pass anymore because you need both hands for holding onto rock etc....

In Austria you will find most maps classifying trails as , Wirtschaftswege (highway=track), easy Wanderwege/hiking trails (highway=path , foot=designated), difficult Wanderwege/hiking trails (highway=path , foot=designated) - still no hands needed but unafraid of heights, Klettersteige (graded A-F in guides though not in maps) (fixed rope tour / via Feratta) hands are needed - but I've passed some grade A and A-B with my bike which was rather difficult, and finally Kletterrouten / climbing routes (only for very few maps climbing routes are shown) - usually you would need to fix your bike to your backpack if at all you want to have a chance to get up (graded with many different systems), howevever some climbing routes of grade I-II UIAA - [4] are so easy that you could still go downhill on a bike (or go uphill carrying your bike if your surefooted). --Extremecarver 16:27, 21 November 2008 (UTC)

I don't think we need to map to sac_scale. The SAC scale does not take into account the added difficulty of carrying your bike. Though I think the original point was just that "anything lower than SAC scale T3 can simply be bicycled by a sufficiently skilled mountain biker". No need to worry about general mapping between scales. --Hawke 15:44, 21 November 2008 (UTC)

I just put this in as comparison.

Split STS grade S0?

I see no need to split the STS grade S0 in two. If the only difference is whether it can be used by cars, then there's no need to split it here -- highway=path and highway=track should do that. --Hawke 19:51, 21 November 2008 (UTC)

Describe scale levels with words instead of numbers?

I'd prefer to describe the trail or the skill required to ride it rather than using numbers, as it is easier to memorize and remember while editing. Avoid the same problem that came with tracktype=*. -- vibrog 11:59, 26 November 2008 (UTC)

take important keywords out of the description, a scheme (flat/dh) could be:

• 0: mtb:scale=hardpack
• 1 =small_obstacles
• 2 =rocky,loose
• 3 =boulders,hairpinturns
• 4 =steep,steps
• 5 =very_steep,landslide
• 6 =extreme

It is important to note that splitting of ways are discourged. This means that the relative frequency of difficult turns, obstacles and/or needed dabs are relevant in the description of these levels as well.

That would be part of the discription on the wiki. MTB riders know what an S0 or S3 is. Because most do have problems with S2 already. So those won't ride S5. Using names is difficult. SAC scale also uses numbers and words in OSM which i find confusing because i use the numbers when talking to someone.

--Mightym 14:46, 26 November 2008 (UTC)