Orienteering NZ Wiki

The need for and the techniques to make a map that is legible for the runner at speed. May impinge but does NOT include “what symbol should I use for…” And does NOT include “how can I do this in OCAD…” See also Assessment of mapping, The Scale Question.

I would love a mass debate on symbol sizes… I am concerned that mappers are being over-ruled and bullied into conforming to a set of specifications that are less than perfect. Every area we map has its own uniqueness, and when a mapper carries out his/her survey they get a feeling for that terrain and end up with a representation that should be consistant over the entire map. I understand that there are varying mapper styles and thats just human nature. We all try to create our interpretation using the same rules. An experiment was made overseas with several different mappers. Every map ended up quite different from the other, there was probably no right or wrong. ISOM 2000 was made to set some rules as required, with the emphisis on legibility which is fine. Now days Orienteering has changed with the normal map being made at 1:10000 and even larger for sprints etc. Smaller maps are being made using more complicated and interesting terrain, with the emphisis on accuracy, creating a challenging map for the runners. I'm sure everyone mostly agrees so far… The problems start to arise when we get in certain types of terrain that have complex arrangments of one or more particular feature. Such as lots of bolders, intricate vegetation boundary lines, limestone cliffs such as Pio Pio and of course sand-dune terrain with lots of knolls etc. If the ISOM 2000 rules that a 1:10000 map be drawn using 1:7500 symbol sizes was applied to some of our best detailed maps the result would be maps that would have to be ieter undermapped by excessive generalisation or the result would be an illegible, hidious to look at map. Using the theory that if an area is too complicated to be legible using these specs, is unsuitable for orienteering (as the ISOM 2000 states)is bollocks (in my opinion) therefore we have an issue to solve on these type of maps. I wish to use White Lightening as my example because it is recent and has some relivant issues. Both myself and Mike B, the other mapper, both requested that the knoll dimensions be slightly reduced, consequently this was not done by the cartographer because she had too much criticism from doing so on other occasions. This dissapointed me because it made some areas of knolls look terrible because they bled into each other and are hard to read. They also did not represent what I had mapped as the dimensions of the knoll on paper was ridiculous when applied to actual size on the ground. The white space between symbols is also very important as this gives the runner information about distance and overall feeling. When you are at these sites the map can give you the impression of expecting something larger than was intended. One way to try and remedy this is to slightly expand the mass of symbols and give cartographers the right to move contour lines etc to fit. This is currently what often happens to make a complex area easy to read, the consequence of this, however is that the end result is often not how the mapper intended the terrain to be represented. I strongly feel this should not be done unless the cartographer is also the mapper and the final result still works on the ground. In my opinion, an easy solution would be to allow slight deviation from ISOM 2000 in extreme cases. I know that rules are rules, but it does not mean they are always right. I stumbled on a website by Orest Kotylo, he is one of the worlds most prolific and best mappers and also an advisor to the IOF Technical department, it is very clear that in his respected opinion there needs to be rule changes in the future specification and, in the mean time mapper should be given some small lee-way in regard to symbol size and line thickness when required. I request that the cartographer be allowed to reduce to size of knolls on White Lightening just enough to fit without making them too small as were the previous specifications. A compromise would be that a new, slightly smaller knoll size symbol was created and used only on the problem areas where needed. Check it out http://www.o-maps.com

I've been prominent in the criticism of undersize symbols.

The problem with a “small” leeway is that there is no smallest knoll, depression or contour wrinkle in nature. Reduce the symbol size, and the mapper can put more on. And history shows they will do just that.

But there is a minimum physical size in the terrain, if this is stuck to there should be a problem with what Paul is suggesting.

I do agree with Paul to a certain extent. On occasions you do get features that are too close together. Michael I know that you stick very closely to the rules, but I do notice on a few of your maps that you do distort the map so that you can fit features, isn't this bending the rules anyway as you expand the detail compared to what it actually is in relation to the the whole area?

Greg, you are bang on with the notion of minimum size in the field!

Simon, I notice that you posted at 2:30 am. You must have been quite shit-faced by then, but still, against the odds, have come up with a valid point.

Sorry about this… but just back-tracking a little, a third and important path sometimes taken in order to make the map clearer [enter 'C'] i forgot to expose, is simplification and generalisation. This should be and is usually used by most of our mappers with varying degrees. In my opinion (this phase appears to get you out of the deep preverbial, so I'll use it here quite liberally) there is a high risk of easily applying too much simplification, resulting in a map that people do not enjoy running on because of the bingo effect. If too much is left off (and it's a fine line in subtle terrain) the competitor can be confussed to the extent that he/she is even unsure if they are even where they think they are! Relocation can be very hard if required. The course setters can also run into problems finding suitable control sites. At least if the map errs on the overmapped, controllers can chose the more definate features. And the competitors can use their own simplification skill at will, which I think is part of the whole philosophy (that's a big word for me)of orienteering.

The mapping for the Oceania Middle Distance (gold mining terrain as complex as Naseby) was controversial. The mapper Alex Tarr hosted a mapwalk the next day for members of the Oz technical and mapping committees. None showed up, but 7 interested kiwis were treated to a valuable morning. The key techniques Alex used to maximise legibility was (as I recall) the use of the bank line without tags, strategic use of the watercourse symbol (even though there was no water), and not being afraid to leave stuff off. The bank line can be 0.18 or 0.25 and I thought the thinner line didn't stand out from contours very well.

For other recollections ask Wayne and Trish, Rob Newbrook and Sue Scott, Gavin Scott and Ian Holden. The map can be seen via the Routegadget link on the Oceania website. It was produced at 1:10,000 for the elites and 1:7500 for the rest of us.

Keep those OCAD tips and tricks coming!

After letting a bit of time go by I would like to discuss the middle distance map. I couldn't handle the contours especially in the part above Inland Road. I wonder whether it was the large number of formlines. Or perhaps wiggles have been fieldworked on top of earlier wiggles, I've found updating maps is harder than working off more generalised photogrammetry - and yet you don't like to throw away the result of previous hard work. Did I just have an off day? Could it have been produced at 2.5m?

We went out and did a relocation exercise for training afterwards and I found there was a large amount of what felt like unnecessary detail mapped. E.g. very small re-entrants that you wouldn't notice while orienteering, that probably would have made the map easier to read if they had been left off.

Michael if we had produced the map at 2.5 metres, like much of Woodhill is, the slope would have been almost unreadable particularly in the slightly steeper areas which most of it was and would hsve shown way to much unnecessary detail that didn't need to be shown. The contour lines would have been to close for comfort. I think what was done, was right.

You comment on to many formlines making it hard to read, which were basically only the chosen information at that intermediate level between 5 metre contours, is counter to you saying could it have drawn at 2.5 metres.

Re Middle Earth: In hindsight I think we did the right thing using 5m contours for a number of reasons after briefly considering the 2.5m option, which actually would have been easier to fit in with the other existing maps west of inland road. The flater part of the map near the beginning of the courses I believe would have looked clearer using 2.5m as there was a lot going on and many formlines were used anyway. But to me the vast majority of the map was represented well. It will always be a bit contentious having so many formlines on a map however I feel that the nature of this terrain type required them. You only need to look at many recent international event maps to realise we are not alone in sometimes having to depict the terrain in this way, alternatively using 2.5m contours would in some cases be clearer it is also going against the most common contour interval now used around the world even in complex terrain. It wasn't a clear decision though as the existing 2.5m version of the map on the west side of inland road (Stag's Roar)did give a very nice and perhaps better representation.

Another gray area was the decision by myself to include the small tree (green dot symbol) after a even split decision by three mappers. Some of these were a little small, however they were distinctive as being non coniforous, there weren't that many of then, and leaving some of them off may have resulted in confusion. I also felt that including them on the complex terrain would help with navigation for some grades, particularly when the oringinal version of the map had almost no tracks on, until the 4WD club merrily destroyed some of the best parts of the map with a network of brand new tracks, much to our consternation.

The thread about obtaining information about a map before a competition has got into discussion about specific mapping issues. I think those would be better talked about here where non-mappers can avoid them, and then others can concentrate on the difficult ethical questions over there.

Ross says 1:15000 has a different symbol size and will of course affect how the mapper interprets and draws the terrain. How so Ross? I know that mappers are TRYING to fit more stuff in all the time, but 50% bigger symbols and 50% bigger scale and nothing changes. What's your interpretation of the bit in the speci section 3.1 in bold type?

Ross says that on the current “Slump” map there are many cases where there is a distinctive tree symbol 419.0 used when it should be 420.0. Where does “should” come from Ross? The speci gives us the ability to use 418, 419 and 420 as we choose. NZ adopted conventions many years ago for the green cross and circle, for an IOF event it should suffice to give the usage in the legend?

Ross says that at some point (updating old maps) you have to start fresh. That's certainly a point worth making, I've worked on many updates where I wish I had. Certainly where the geometry is bad, in spite of OCAD's rubber-sheeting tools. If the old fieldworkers made a mess of the contours it would be worth going back to the original photogrammetry (it's all looked after carefully isn't it???) It can be worth deleting and redrawing all the vegetation because the objects relate to each other and to other features. Whether that point was reached in the case of the Slump I dunno (you said it was quite good) but certainly for maps that everyone agrees are BAD, we should start again. This is the case for many “old faithfuls” up and down the country.

Hi Michael,

OK, my point about symbol sizes was a bit loose, but enlarging the map (15 to 10 thou) changes things, even when the symbol sizes are also changed. The symbols increase in size, but the gaps between everything become bigger too. This leaves more room to represent more detail which is important in middle distance orienteering. Also, any inaccuracies on a 1:15000 map are magnified when it is enlarged, regardless of symbol size increasing. On a map that stems from some very old sources I'm sure there will be at least a few cases of that happening! I'm not saying that you can't have a map that can be used for both long and middle distance (1 to 15thou and 1 to 10thou), but if you are having a World Cup (middle distance only) on detailed terrain it should be mapped specifically with that in mind. For 1:10000 There should be more line contour detail representing shape and less of the point features (small knolls and depressions) that are used more in 1:15000 for graphic generalisation (this can be achieved because of the bigger gap between symbols gained in 1:10000).

With regards to the distinctive trees, I see that you are allowed to use which ever symbol you so wish. But you have 2 different tree symbols to work with, why not use them? I don't know when it became commonplace, but from my experience in Europe, the small green dot is always used for trees that have a small trunk and little canopy (such as cabbage trees) and the ring for bigger trees with a canopy. I suppose it would suffice to give the use in the legend, but by simply changing the appropriate trees to the different symbols then map will provide the reader with a lot more information.

The current Slump map is good yes, and the terrain itself is great! Being from Hawkes Bay I can certainly visualise what will be in the terrain from the map. I'm also sure that the new version will be even better. My only concern is that others might have difficulty interpreting what's going on. I think that for a World Cup, things could have been stepped up a bit. Nit-picking elite orienteers (believe me, I'm not the worst!) will scrutinise over every small flaw they can find.

PS. The thing about the 1:15000 Slump with different symbol sizes comes from here: The ocad map is at 1:15,000 but the contour lines, form lines, slope lines and dot knolls have been reduced to 80% of standard IOF line widths. The magnetic north lines are at 250m apart as used on 1:10,000 scale maps. The map is set up for printing at 1:10,000 scale as it is easier to read the contours, knolls and depressions at this scale.

Read more here: http://www.hborienteering.com/club/history/maps/TheSlump.pdf

Interesting. I don't know that the increased gaps in 1:10,000 (in proportion to the increased symbols) are supposed to be filled up with extra detail are they? What do other mappers think?

Ross can you tell me where I can get the publication “From My Experience”? If we are going to follow that in NZ I'd like to have a copy:-))

HBOC 2007 version: Naughty!

(I had to rejig this after reading more carefully:-))

You may have nits Ross?? I tend to agree with the concept and reality that in many terrain types a map prepared for a 1:15 and one for a 1:10 are two totally different beasts. I think it's one of the main issues that need to be resolved and catered for in ISOM201X. Long and middle course requirements are so different but worldwide it looks to me that the lines are blurred too much between the two course setting styles which in turn causes problems. However to me looking at the old Slump map and the type of terrain it is,I think you may be overly worried (admirably so) Providing of course that the line thicknesses and symbol sizes are reinstated, and since offset printing quality will allow this easily, as well as a careful and thorough remap as will be the case. I would expect if there were some point features that could be represented as contoures these would be picked up. I would think the Slump could be viewed by competitors and the IOF as an impressive model ISOM map in respect to generalization and readability. If it were to go down the path of including a lot more formlines I think the map will be too complex to run at the speeds that we will see in January. I'm sure there are areas where more detail should definately be shown, and you know the map better than me Ross, but I also think and hoping your concern is overworried about not being perfect. Maybe this comes from being such a magnificent Sprint Map specialist! My thoughts on the trees are that the green dot potentially could get mixed up with small vegetation patches, I think if it has a small trunk and not that distinct, then leave it off anyway. More frequent use of slope Tags could be useful on the Slump as also needed on many of our maps. Couple of things I don't like about the outgoing ISOM that I would love to see on the new would be to distinguish between the black of rock and the manmade black of tracks and buildings. One of them could be a dark gray. Probably the rock as it might be able to be transparent to brown contour detail. I also belive that there should be large and small dot knolls and elongated knoll point symbols, amoungst other things but thats another story. Oh yeah, the small earth bank symbol sucks out of proportion. Great to be concerned even though it's a bit late. I'm sure it will be awesome and can't wait.

Hey Ross, there again I realise that you are out the experiencing the latest developements in Europe so we mappers need to listen carefully to you and other thoughtfull elite orienteers.

Haha, I don't know if my experience is worth following. I think that looking at maps of reputable competitions in Europe would be a better guide

When I'm not sure about something in sprint mapping, and ISSOM is being useless, I just look at some old maps, or search some online, http://omaps.worldofo.com is great for that. It might help for 1:15000 & 1:10000 too. It would at-least make identifying any trends that emerge a bit easier - it's a shame they don't update ISOM every time someone starts to do something a certain way

I remember this race: http://www.woc2008.cz/files/images/scontent/maps/middleF/middleF_MEN.gif - a good example of a map that would be a bit of a nightmare if mapped the same way at 1:15000 and also one using both kinds of tree symbols, even in forest! - 'from my experience' just the ring is usually used for distinctive trees inside forested areas, the small dots are used more in open areas.

I think simplification is the most important thing for orienteers - whether competing or fieldworking. The meaning of “distinctive” is that is sticks out from the rest of lesser or other species of trees. Competitors haven't got time to analyse the type of tree. To my mind a green dot would best represent something like a gorse patch. If you intend using it for the likes of a cabbage tree (essentially an aerial feature) what do you use for a gorse patch which is essentially a ground feature?

NZ flora differs dramatically from that on European terrain. If it ain't broke why “fix it”? Next you'll be trying to differentiate between a round and a rectangular water trough!

It should be up to the runner what they to choose to simplify from what is given to them. Gorse patches just need to be drawn as a shape symbol, but keep it misshapen, then they look a lot different from a perfectly round dot-tree.

This: http://www.cornwallpark.co.nz/UserFiles/Image/38690025.jpg

Looks a lot different from this: http://www.allcountries.org/photos/new_zealand/cabbage_tree_harataonga_new_zealand_photo_scott_venning.jpg

I think it's more important in NZ, because a lot of our maps are in open land, and we have a great variety of flora of different sizes especially compared to Europe which is mostly thin pines. I'd prefer to have both on the map.

The water-trough analogy is a bit of a reach. In comparison to big canopy trees and small trees it's more like saying you should map a small water trough and a big water tank as the same feature.

A green ring is 12m in diameter and the dot is 7.5m, that's a big difference.

Generalisation is such a controversial topic. For those who want to keep it simple and follow the IOF Map Commission guidelines and ISOM2000 with zero tolerance of straying from the rules, this has resulted in rejection of larger scale maps for middle events on a large number of high profile events, there is a discussion going on where many of the world's top mappers are claiming if they were to generalize as expected of them they would be out of work, such is the customer demand for detail. Many of the most popular world O multi-day meets can no longer get WRE status. Hence decisions on ISOM201X appear to have stalled.

http://web.entrebalizas.org/Doma/show_map.php?user=imingorance&map=174

…a better example of green dots and circles

Wow Alistair!! That map looks like it's been attacked by green measles. One wonders about the reliability of those green spots. There are so many that it is impossible to navigate using most of them. Even the Planner has restricted the control sites to obvious features that probably wouldn't normally be used on a M21A course.

The fieldworker must have had a nightmares recording them - presumably most came off aerial photography???

With boulder fields we have a specific symbol whereby we don't attempt to plot every boulder - just the “pick me” ones which stand out. Perhaps a green dotted screen would have been easier to use, such has been used on the Whakaipo Bay map where the gorse patches are so numerous it would take forever to map the small ones, yet it is possible to run through the these area with minor weaving around individual bushes.

I was browsing the RouteGadget Facebook pages, which is not so much about RG but Pullautin, the software for automatically generating maps from LIDAR data. But this comment was not even about Pullautin, it was mapping philosophy. Consider this.

“For the best athletes the leg is the task, not the control - a control is never genuinely challenging if there is lots of details. It means putting control to a really really detailed place usually makes no sense. It is just easy, because from the edge of the control ring one can see all details. But if same details are spread to four times larger area it becomes more challenging because you can't see it all at once. The only difficulty with detailed controls locations is legibility, one may not be able to read the map well - and that is not the kind of challenge we are supposed to give atheltes.”

I guess we should do this kind of thinking also when planning how to map an area and to figure the right generalizaton level. And what ISOM should really be like. Do we or have we done that?“

This is really interesting. We have a fairly blunt instrument for controlling generalisation, and that is the scale and symbol sizes. A call for larger scales can be seen as a shift in the type of orienteering, the long becomes a bit more like the middle, the middle becomes a bit more like the sprint. We should instead back off and consider what types of orienteering we want to have.

Malcolm asks an interesting question in his report on the NZOF website/FB page. The public race on the middle distance area had termite mounds marked, while the WOC race did not.

Seems wrong to me to leave off the mounds if they are distinct. I would think any significant ones would stand out as much as some knolls and rocks, if not more than. Generalisation of rock to only the big one's can totally depend upon the terrain. With the undergrowth and moss covered rocks shown on the tv coverage small rock wouldn't stand out well in the terrain hence a higher level of generalisation. I am certain this level would be reduced if the forest floor was clean. So it's all relative with no fixed rule. Mappers discretion rules sometimes.

More of a topical discussion would be around the overuse of formlines, Long and Middle, especially for a WOC map under IOF scrutiny. They were happy to bag France about their mapping.

On a philosophical level, we include distinct point features that have the right density. When there are too many (a) we can't fit them in and (b) they are no use for nav anyway. Silly example, trees in a forest. More pertinent example, rock features, termite mounds. You're right it depends, we would have to be there to judge.

Yes I've often felt a clean forest floor brings problems. It makes all features (incl contour wiggles) much more distinct. And then there become too many, see above. I think the blueberry ground cover, fallen trees, young saplings etc that we saw, may make the generalisation easier. Including the contours, the number of formlines is surprising. Possibly its the “LiDAR effect” that has been talked about at mapping conferences - if its on the basemap, its actually very time-consuming to decide which to smooth out. You can even get this on Stuart Hyslop photogrammetry when he has good photos.

They usually have a mapping conference at WOCs. I don't recall seeing any proceedings from France or Switzerland. We should ask.

I can't help bringing a bit of bio-pedantry to the brown cross part of the conversattion. I'm assuming these were ant hills rather than termite mounds. These can get as big as the sort of knolls that we'd expect to see mapped in eg sand dune terrain, so in that repsect I can understand them being mapped, but rather than the big “clay-brick” towers that the Aussies and kiwis would expect to see, I'm guessing they would look more like a mound of whatever the typical leaf-litter/forest floor material is. So from an Australasian orienteer's perspective I can well imagine it would have taken a bit of adjusting to. Hopefully they were present and mapped in the model areas and training maps.

I made this comment to another mapper in discussing generalisation, wonder what others think. Specially anyone who ran on both these maps.

Interesting to compare the approaches at JWOC in Czech and WOC in Finland (middle distance maps I think). As I recall in JWOC they used a lot of stony ground dots and I found it hard to make out the boulders on the map. In WOC they didnt use stony ground, the map looked very clear and they even had enough white space for some vege boundary dots. But Malcolm (Ingham) commented that only the big rock was mapped. I wonder which was easiest to navigate in? The map that had everything on it but was hard to read, or the map that had less on it but some people expected more?

I found the WOC maps easier to read than the JWOC middle maps. But the rock on each of the terrains was very different. The JWOC rock was all broken and more like a boulder field, but the Finnish rock was more like bedrock poking out of the ground

Interesting discussion: http://news.worldofo.com/2013/10/20/sandvik-after-25manna-we-need-good-legible-maps/

Thanks for the link rossmaxmo. To me, it's reassuring that the rest of the orienteering world is debating the same issues that are concerning us in NZ with respect to IOF's ISOM revision.

We were allowed by the IOF for WMOC 2000 Harakeke a 9% reduction and we could get all the detail we wanted to map in.

APOC 1994 Knottingley was printed at 1:15000 for elites which was readable with offset printing (but also used a reduction in symbol size).

For World Cup 1 Oceania 2014 we used the standard symbol sizes (we would not have got clearance to use a reduced symbol set) - I was still mapping too fine and after draft fieldwork, the map was unreadable using standard symbols. I then went out and visited everything again and had to think what should be taken off. Generalisation (and some more detail added on) went on for several months. For a World Cup event (or any large international event) factor in a long period required for fieldcheck and map refinement (almost the same time as for the initial fieldwork) - also required endless visits of planner/controller/National controller and a few IOF controller visits followed by mapper going back to fix/correct things. Generalisation can be done but takes a lot of extra time. The whole process to create an IOF compliant map in detailed sand dune terrain probably ends up with a more accurate map as everything is checked and rechecked - the style of the map is different to many of the sanddune maps we are used to which have all detail mapped.

Thanks for posting that Bryan. It's very timely as clubs head towards mapping or remapping old terrain for WMOC in the next couple of years. Pretty certain some of our old styles won't be up to scratch. Also as Selwyn wisely suggested, mappers should keep their ear to the ground on any changes to ISOM which may slightly alter things as ISOM201x looks to become ISOM2017.

I found this interesting: https://www.facebook.com/177518995597572/photos/a.180287198654085.49663.177518995597572/1134109763271819/?type=1

He talks about recognising tops of steep slopes with formlines, I often feel that the bottom of a steep slope is worth a formline, especially the edge of flat land.

The post has nothing to do with Route Gadget, I think that's just the author's Facebook name.

I revisited that link from two posts ago to see which school of formline mapping the JWOC maps followed. I think it was a transient link, try https://www.facebook.com/pages/RouteGadget/177518995597572?fref=photo and go to the posting on 26 June, also 28 June is relevant. Which do you think? Lots to be learned by studying the championship maps at this time of year.

Some good discussion about use of formlines on RouteGadget's Facebook page https://www.facebook.com/pages/RouteGadget/177518995597572?fref=ts There are examples from O-Ringen.

Note its not about RG, it's about mapping. I think the RG guy is also a mapper, and the author of the automatic map-generator Karttapullautin

On Maptalk today, davenev (David Nevin Whangaraei) posted : I will update a ridge this summer, it's steep - you can't run up the sides, semi-open, with rocks from 0.5m to 5.0m and a few small cliffs. ISOM says, over 1m. Would we measure that from the downhill side, the uphill side, or take an average, or just pick those that say, Pick me ? Which is pretty much what I did 4 years ago…

My view is that these things that have minimum heights need to be that height from whichever way you approach. So knolls and boulders need to be up to your waist all round including the uphill side. A too-small boulder on a steep slope could be a rockface, though there are minimum lengths to observe as well as height. “Pick me” is a very useful philosophy when there are too many big-enough features. (Thank you Eric…)