Register | Login
Attackpoint - performance and training tools for orienteering athletes

Discussion: Peterdine map

in: Orienteering; General

Aug 16, 2020 1:46 PM # 
EricW:
Just spotted the Peterdine map (WA) posted on World of O.
Very impressive terrain and mapping, at least looking at it remotely.
I note more-than-expected ISOM non conformities, which strike me as well applied, along with expected Aussie quality rock terrain mapping.
I'm guessing I would support 90+% of these exceptions, if I set foot & eyes in the terrain.
I expected to see the name of some familiar AP poster,....but no credits.
Anyone here deserve credit for the mapping?
Advertisement  
Aug 16, 2020 2:22 PM # 
Juffy:
The original map was by Eric Andrews, who was never on AP - he passed away a few years ago.

Andrew Slattery updated it in 2006, and I think Peter Komyshan did some updates 2-3 years ago, but the rock mapping is largely unchanged from Eric's work.
Aug 16, 2020 2:46 PM # 
tRicky:
The yellow definitely needs updating. Many more clearings out there than shown on the map.

Eric, how did this come about? I'm ignorant about how World of O works so don't know if it's something posted recently or if you just happened across it but we had an event on Peterdine today - I was surprised to see someone other than Simmo initiate this thread for a WA event.
Aug 16, 2020 4:40 PM # 
EricW:
The Peterdine map showed up a few hours ago on http://worldofo.com/ in the section headed "Latest selected maps from..." It is now gone.
I didn't check or remember any associated course/ event info.
I don't know how this rotating map section works on the World of O, but I think it is possible somebody posted their map from this recent event, somewhere on the internet, and got enough hits to make the display.
Aug 16, 2020 5:38 PM # 
walk:
The map is still posted and can be seen by clicking on the heading of the map section. Appears to be a course from today, either H2 or H4.
Aug 17, 2020 4:32 AM # 
TheInvisibleLog:
EricW. If you like the look of Eric Andrew's mapping, take a look at Cascades. I don't know the full history of this map, but I think a lot of effort was put into the mapping because it was Eric's local terrain and it merited really good mapping. After my first run on this map I purchased a clean copy, framed it and hung it in my study as a reminder why its worth making the effort to travel to orienteering events. Whenever an event happens here, I make every effort to get there, despite the day long drive. Memories of leaving immediately after the race early Sunday afternoon, driving till the wee small hours, kipping for a few beside a side road, and driving again the next morning. Worth the effort. Copies of the map can be found on WorldofO. Here is one.

http://omaps.worldofo.com/?id=188867
Aug 17, 2020 8:03 AM # 
simmo:
Peterdine was mapped in 1992-93 and first used at the Oz 3 Days in '93. At that time Australian mappers interpreted ISOM their own way. Updates since have been minor, and no attempt has been made to conform with ISOM 17, or even with the old ISOM. I agree with IL that Cascades is Eric's best work. Of the maps he's made in WA, my impression is that he tended to rush them a little, and over-generalise as a result, but I'd have to say that Peterdine is the best of them.

One of Eric's idiosyncracies was to draw circles along a contour line. If you look carefully you'll find a number of them - there's one example just N of the dam in the W part of the map and a couple more NE and E of there. We couldn't persuade him that these were wrong. I haven't found any of them on Cascades or his NSW maps, so I don't know why he only did them in WA - they're on all of his maps here!
Aug 17, 2020 9:19 AM # 
TheInvisibleLog:
Yes, those circles jumped out at me. Even at the time they would have been highly non-compliant. But other non-compliance issues have arisen since 2017. Under-size grey areas are the obvious example. I think it wouldn't be too hard to update this map to ISOM17. Other examples of Erics work (eg Gumble) will be challenging. Still the subject of debate.
Aug 17, 2020 11:36 AM # 
tRicky:
Yeah those circles are weird and they tend to surround rock surfaces. #118 yesterday was described as "Hill and rock surface" but it didn't rate its own contour at all. Probably should have been drawn as a boulder.

I think it wouldn't be too hard to update this map to ISOM17

Thanks for volunteering!
Aug 17, 2020 12:18 PM # 
jjcote:
The circle you're referring to are these things?
Aug 17, 2020 12:33 PM # 
blairtrewin:
The original Peterdine would have been one of Eric's earlier granite maps. Agree that the Cascades was his best work (and is a superb, if tough, area to orienteer on too). Something else to note is that in 1993 Ocad wasn't yet on the scene, so the ability of mappers to shape rocks was constrained by what you could do with Letraset symbols, which definitely makes a difference in granite terrain - a comparison of the 1985 and 2002 Kooyoora maps is instructive there.

If you hit it in a dry season without much grass growth, Peterdine is seriously fast - Warren Key did 56 for something like 12.7k there in 1993. (I almost broke the hour myself, and was rather surprised to see Warren, who'd started four minutes behind me, on my shoulder at the last control).
Aug 17, 2020 1:33 PM # 
tRicky:
Now you're making me feel inadequate. Craig was planning a forecast winning time of 55mins for an 8.4km course on the weekend. Sadly none of the locals were up to the task (rain didn't help). My 9mins of errors in the first three controls didn't give me a good start.
Aug 17, 2020 1:56 PM # 
robplow:
"the ability of mappers to shape rocks was constrained by what you could do with Letraset symbols"

Letraset (rub ons) was used for standard symbols like boulders, boulder field triangles, minimum cliffs, etc. Everything else was drawn to shape, by hand, in ink. There was no constraint, other than the skill of the mapper. In some instances OCAD has led to less shaping of features. Especially for curved cliffs. In granite curved cliffs are generally lower at the ends and higher in the middle so it was common.
(in pen and ink days) to taper the ends. But that is quite hard to do elegantly in OCAD so most mappers don't bother - just draw the cliffs with uniform width using the standard line symbol.



One of the 'constraints' on the 85 Kooyoora map was that the IOF controllers only allowed Steve Key to use a single (quite small) size of boulder field triangle (scalene) . Steve got around that in many places by using the larger boulder cluster triangle (equilateral) in groups to show boulder fields with large boulders.

When I was drawing in the those days I got my own letraset made with a range of triangle sizes.
Aug 17, 2020 11:19 PM # 
TheInvisibleLog:
"Thanks for volunteering!".
Sorry but your state's border restrictions mean someone in WA will have to do it.
Seriously, my comment referred to the ratio of white to black on the map. The greater the white share, the less difficulty YOU will have. Compared to Gumble, Peterdine looks pretty white to me. Maybe I should have used the word "relative" with regards to easy.
Aug 18, 2020 12:00 AM # 
robplow:
Those 'weird circles' are just normal bare rock knolls. Breaking the contours for them was one of Eric A's idiosyncrasies. I think his logic was that these knolls sort of break the 'natural' curve of the land (more like b than a, ) so it made sense to break the contour.



I don't agree with that and would draw something like that as a smaller knoll plus spur.



I know some people got quite upset with Eric for drawing them like that but I never found that representation in any way confusing. Certainly it is less work to draw it that way.
Aug 18, 2020 12:18 AM # 
robplow:
Eric W - apart from the 'weird circles' what other ISOM deviations did you notice?. And which ones (10%) would you not approve of?
Aug 18, 2020 1:13 AM # 
EricW:
I was afraid of this.
The "90%' comment was casual eyeball based, with no measuring or consulting any version of ISOM.

I guess the short answer would be, I allowed 10% to account for something I would probably regret endorsing if I visited the terrain. I didn't notice any specific issue that I completely disagree with.

However. let's start with undersized dot knolls, although these might be a bit too undersized for my taste. I think current standard is too large for many settings, often being visually stronger than smaller contour knolls which are larger in the terrain. However, I think dot knolls should still be equal to, or a hair larger than boulders.

Double form lines, at least by strict definitions, which Australia seems to adhere to.
I usually support a few of these in certain terrain types, but agree they are generally not needed in gully/spur terrain.

Extra light stony ground, improvised scattered single dots, which used to be OK, but I believe are currently outlawed, certainly prohibited in my reading of the draft version of ISOM 2017, with which I was most familiar. The way they are used on Peterdine strike me as probably very appropriate (pending a site visit)

And yes the "weird circles" stood out. I assumed they were as you described. They strike me as a nice simple way to depict the feature, even if I wouldn't do this myself.

I didn't notice any undersized bare rock areas, but I f they exist, I don't want to endorse them. I think the min size specs for area features are appropriate.

All of these comments are all from memory, since I didn't want to get too obsessed with another look. :-)
Perhaps there was something else I'm forgetting, but I'll leave it to others to point out and debate other "violations".
Aug 18, 2020 1:48 AM # 
yurets:
Looks like a very open terrain, too much sun, no shadow, no place for kangaroos to hide and take a nap after dinner
Aug 18, 2020 1:56 AM # 
robplow:
Thanks Eric, I wasn't expecting a rigorous answer - just interested in the general observations of an experienced mapper. Always good to hear from someone who is not part of the local mapping culture - get a fresh perspective - challenge some things we might just lazily take for granted.


Double form lines, at least by strict definitions, which Australia seems to adhere to.I usually support a few of these in certain terrain types, but agree they are generally not needed in gully/spur terrain.

Actually the old ISOM forbade double form lines but the current version says something like there 'should not' be more than one. So you can have doubles but it is not recommended. Which is about the right balance I think.

Extra light stony ground, improvised scattered single dots, which used to be OK, but I believe are currently outlawed, certainly prohibited in my reading of the draft version of ISOM 2017, with which I was most familiar. The way they are used on Peterdine strike me as probably very appropriate

As I recall - in one draft of ISOM 2017 the ability to draw stony ground dot by dot was removed - you had to use the area symbols. But that was not kept. Now you can use one of three areas symbols (varying densities) or you can draw them dot by dot - with some rules about min/max distances between dots. While the area symbols seem like a good idea I find I almost never use them because it means i have no control over where the dots fall and they end up obscuring contour detail or touching or being too close to other black objects, which reduces legibility. I tend to use a variety of sizes for stony ground dots as well (ISOM has only one size) as the 'stones' can vary from fist sized to nearly 1m. Over about 1m you have to switch from stony ground dots to boulder field triangles. Which also need to come in a variety of sizes for granite terrain as the 'boulders' can vary in size from around 1m to huge. ISOM 2017 allows enlargement up to 120% to show 'height differences'. That is not enough in my opinion.

. . . undersized dot knolls, although these might be a bit too undersized for my taste. I think current standard is too large for many settings, often being visually stronger than smaller contour knolls which are larger in the terrain. However the dot knolls should still be equal to, or a hair larger than boulders.

Without seeing the OCAD file for that map I would guess the brown dots are the same size as the small boulder dot. That is what I usually do. Features mapped with a brown dot in that sort of terrain are typically of similar size/prominence to a small boulder, so it always felt wrong to me to use the standard (larger) dot knoll symbol - it makes them look bigger on the map than they appear in the terrain.
Aug 18, 2020 2:15 AM # 
gruver:
How can I see this wonderful map? I searched WorldofO map name=Peterdine and got nothing.
Aug 18, 2020 3:05 AM # 
robplow:
http://omaps.worldofo.com/?id=281199
Aug 18, 2020 3:06 AM # 
EricW:
It is still viewable on World of O as walk posted above.
"The map is still posted and can be seen by clicking on the heading of the map section. Appears to be a course from today, either H2 or H4".

In other words, in the section "Latest selected maps from" click on "omaps.worldofo.com" and a bunch more show up, including Peterdine.
Aug 18, 2020 3:10 AM # 
simmo:
The dot knolls are indeed 0.6, same as a small boulder. The retention of drawing stony/broken ground dot by dot was only retained in ISOM 2017 I believe at the very strong insistence of Australian mappers. At one stage I heard that IOF Map Committee also wanted a single sized boulder field drawn by area only - thankfully that didn't happen.

Steve Keys use of equilateral triangles for large boulder fields was a great innovation, which was continued by Alex Tarr among others. Eric Andrews didn't take it up, at least not on WA maps, which I think detracted from his mapping. There are a couple of areas where several boulders larger than houses are mapped with relatively small scalene triangles. Next to these are some separate 'large' (but smaller than the boulder field) boulders, so an orienteer arriving in the area is immediately drawn to the largest visual feature, but can't find it on the map!

Eric did not usually break the contour to add the circle Rob, mostly there's a continuous line from either side of the circle. If you can access the map, there's even one without any gray in it near the control at the end of the long leg.
Aug 18, 2020 3:14 AM # 
simmo:
I just followed the World of O link. Clicking on the map takes you to the Livelox for the event, the long leg i referred to above is 7-8 on course H2 and 2-3 on course H3, the circle I mentioned is a little SW of control 8/3.

Gruver you obviously weren't part of the A/NZ Challenge that happened in '93 during the week following the Peterdine race.
Aug 18, 2020 3:21 AM # 
EricW:
Dot by dot stony ground- I also lobbied for this via USA's feedback on the draft ISOM2017, for the same reasons.
Aug 18, 2020 3:28 AM # 
simmo:
There are some examples of the use of boulder field that i described above on the Peterdine map, although the worst example is on a different map. Looking at the H2 course again, on the rocky slope above controls 15 and 16 there is a small, form-line hill where the map would lead you to believe that two large boulders below the hill are the prominent feature. In fact they are of a similar size to the boulder field drawn with small scalene triangles just to the North. Further NW there are two boulders drawn to plan - which makes them quite large black blobs, but they are actually smaller than the adjacent boulder field.
Aug 18, 2020 3:39 AM # 
robplow:
Steve Keys use of equilateral triangles for large boulder fields was a great innovation

I always figured the only reason he did that was because they only had a single size (very small) scalene triangle on the letraset sheets. If there had been bigger scalene triangles available he would have used them rather than the equilaterals. If you have a range of sizes of scalene triangles there is absolutely no need to use the equilateral triangles for boulder fields - in fact it is potentially confusing. In my mind the real 'innovation' was to use more than just one size.

I tend to 'hand draw' boulder field triangles (using a black area symbol in straight line mode it takes just two 'click and drags'). I find it very hard to get the orientation of the standard OCAD triangles the way I want them - especially when you have several triangles close together. Drawing them like that also makes it easy to vary the size as appropriate.

Aug 18, 2020 3:47 AM # 
robplow:
Eric did not usually break the contour to add the circle Rob, mostly there's a continuous line from either side of the circle. If you can access the map, there's even one without any gray in it near the control at the end of the long leg.

you mean like A? The contour is still broken - an unbroken contour would look like B

I am not saying I agree with that representation - just that the meaning was clear. When running on his maps it never caused me any confusion, which is the important thing - not just blind compliance with the exact letter of ISOM rules.

presumably the one with no grey is just a rock knoll with moss or leaves etc on it.???


Aug 18, 2020 3:59 AM # 
tRicky:
What's a moss?
Aug 18, 2020 4:05 AM # 
robplow:
Kate, Elizabeth, Ian . . .
Aug 18, 2020 4:09 AM # 
robplow:
There are some examples of the use of boulder field that i described above on the Peterdine map, although the worst example is on a different map. Looking at the H2 course again, on the rocky slope above controls 15 and 16 there is a small, form-line hill where the map would lead you to believe that two large boulders below the hill are the prominent feature. In fact they are of a similar size to the boulder field drawn with small scalene triangles just to the North. Further NW there are two boulders drawn to plan - which makes them quite large black blobs, but they are actually smaller than the adjacent boulder field.

those are excellent examples of the importance of showing relative sizes. I have always thought that is one of the biggest failings of ISOM for mapping rock - not allowing for a bigger range of sizes in triangles: both boulder field and boulder cluster.

The size range of triangles should be the same as for boulders where you have small, medium, large and 'gigantic' (206). If individual boulders come in such a range of sizes then logically so do boulder fields and clusters.
Aug 18, 2020 7:12 AM # 
Jagge:
Thanks robplow for explanation. We have had similar situation with rapakivi boulders/knolls. Those are boulders that erode over time and become knolls. While still at boulder phase one can place them anywhere, also on top of a contour. But when they erode more and become sharp cone like knolls, one needs to move the contour. So, very easy to imagine those wierd rings may be rock sticking up from ground, sort of independent of land form. And some of them are mapped as boulders, some just as bare rock, some as bare rock knolls and depending on the method you may or may not need to move the contour. In addition, if you add a cliff symbol it magically becomes ok, and also if you draw same thing with form lines we see it just fine. But with full contour it is big no-no. I can imagine the mapper had plenty of arguments.

ISOM2000 used to allow 120% enlargement of triangles, but around here here mappers deliberately misunderstood ISOM document to get even more flexibility: The picture in document states range 0.5-1.0 by the triangle. I guess that means height and width, but here interpretation was allowed height being anything between 0.5 - 1.0 mm, so triangles of half of the size were common. Robplow's suggestion of triangle range being same as for individual boulders sounds a lot how some mappers used to use that added flexibility. You will still see maps like that, proper conversion to 2017 would need remapping of rock features. Personally I don't much like with those stony ground limitations either (not arranged in line and min 3 dots) and these maps may be fine examples.

BTW, Kari S. made WOC 85 maps famous here and I suspect Australian maps had major impact on rock mapping here. Not immediately and there is still long way to go. And 1993 WOC maps changed the way vegatation mapping was seen here. I can still remembe a pro mapper showing the detailed green mapping and saying that's the way of the future. Here maps used to have only quite large patches of green, couple of years laters maps with attempts to map green in a detailed way appeared.
Aug 18, 2020 10:35 AM # 
robplow:
detailed green mapping and saying that's the way of the future

That's sort of funny. I remember attending an ICOM in the 00's where someone gave a presentation on generalization. He showed lots of examples of Swedish maps with similarly detailed green mapping and said "we don't want maps like that here" *. He blamed the trend on more detailed green mapping on the influx of central European mappers.

*After the presentation I talked to a well known Swedish mapper who told me "if I made maps like he (the presenter) wants I would be out of work within a year" - ie: his customers want maps with detailed green mapping.

Ever since I started mapping seriously in the late 80's I have used varying sized triangles for boulder fields and clusters - inspired by the WOC 85 maps. It has never been technically 'compliant' with any version of ISOM. No one has ever complained about that - not to my face anyway. It just makes sense to map that way. It makes the maps better and fairer because you can show relative sizes properly - you don't get the sort of anomalies that Simmo has described above. I have converted some of my older maps to ISOM 2017.ie: I changed the formlines to the new thinner version, changed brown crosses to triangles, etc. But I am not about to convert all triangles to one standard size. And I am not about to stop using vary sizes of triangles.

In ISOM there is a minimum size symbol for a boulder but no maximum (206 Gigantic boulder can be drawn as big you want). The same should apply to boulder fields and clusters.


stony ground . . . not arranged in line ISOM 2017 says: "To avoid confusion with symbol Distinct vegetation boundary (416), the dots should not be arranged to form a line". "Should" not "shall". So you can do that if you really want. i believe that was something Australian mappers also demanded. Although I have never felt the need to put stony ground in a straight line in any terrain - Australian or elsewhere.
Aug 18, 2020 10:50 AM # 
blairtrewin:
Long narrow lines of stony ground (typically on top of spurs) are characteristic of some terrain types in the drier parts of Australia (South Australia is where they most commonly turn up on orienteering maps).
Aug 18, 2020 11:01 AM # 
robplow:
Yes there are - and I have mapped many terrains like that. I have always mapped them as a narrow band of stony of ground - not a single line of dots.



I have never seen an example where space was so tight on the map there was only room for a single line of dots. You don't have to offset the dots very much.
Aug 18, 2020 11:13 AM # 
Jagge:
That Cascades map has (short) sections of dots in line. I can see it is possible to map it like robplow suggest, but mapper might think it makes it look too stony for added density and/or take space from the non stony ground part. But I can guess it is possible to find a work around the issue. So not conviced line arrangement needs to be stated in our standard.

Jukola map from the same era. Varying robplow style triangles, pretty Australian style stony ground use I think. No varying dot size, but dots arent actually round to make them differ from vegetation boundary dots. Highly non-compliant.
https://doma.martinivarsson.se/show_map.php?user=m...
Aug 18, 2020 11:20 AM # 
blairtrewin:
I certainly prefer Rob's version of the linear stony ground. (I suspect this type of formation is a characteristic of desert/semi-desert terrains, but not too many other orienteering countries have these - the western US and parts of southern Spain come to mind).
Aug 18, 2020 12:00 PM # 
robplow:
That Cascades map has (short) sections of dots in line

That is "granite" terrain - the rocks generally don't tend to form in lines in granite.

It's not so much 'semi desert' - it is more the type of rock surely. And the fact that it was not glaciated in the last ice age. Although I guess the semi- desert lack of undergrowth makes the rocky ground more visible/relevant than in other areas.

Interesting Jukola map - the stony ground looks like very small triangles - that seems like a logical thing to try out in order to distinguish from veg boundaries. But in reality I doubt anyone would notice the different shapes when running - they both look like small black dots until you study the map very closely at home afterwards. Nowadays there is the alternative veg boundary symbol of dark green dashes - introduced in ISOM 2017 to try to avoid confusion with stony ground.

Otherwise it looks like a pretty standard Finnish map to me - lots of bare rock, very readable. Index contours would have been good.
Aug 18, 2020 1:02 PM # 
manudona:
This map is another example of linear black dots along spurs. Classic semi desert terrain in Alicante, SE Spain. Mapper was V. Dobretsov, Russia.

Hard to distinguish vegetation boundary and stony ground. Russian School tends to overmap certain features.

http://org.ntnu.no/ogruppa/sportslig/doma/show_map...
Aug 18, 2020 7:46 PM # 
EricW:
"And 1993 WOC maps changed the way vegatation mapping was seen here. I can still remembe a pro mapper showing the detailed green mapping and saying that's the way of the future."

For those unfamiliar with the Harriman/WOC93 maps, the green is mostly Mountain Laurel which is unusually distinct, compared to other thick vegetation types. To Steve K's credit, he picked up on the value of drawing this unfamiliar-to-him vegetation in great detail.

I think the important point is that the level of green detail should be relative to the vegetation type, Most Nordic thickets are not nearly as distinct. Some planted/managed vegetation types in continental terrrain are probably more comparable.

Linear stony ground- One common application in my region is to use this symbol for indistinct stone walls, about as narrow as you can get. I happen to believe this application is overused, but I have never seen a case where it was not possible to offset the dots.
Aug 18, 2020 9:35 PM # 
jjcote:
And as a further clarification, the basemaps that Steve Key was given to start with were not just photogrammetry, but had already been fieldchecked by people who were very familiar with mountain laurel.
Aug 19, 2020 12:59 AM # 
TheInvisibleLog:
Re: black dots in straight lines. "I believe that was something Australian mappers also demanded."
My memory of that time is that we strongly requested that individual dots were allowed as an alternative to a dot screen for both rocky and broken ground. I think the straight line constraint was a reasonable compromise from the MC. Being a small orienteering nation we were in no position to make demands. We can hope for constructive collaboration. If we "demanded" anything, it was that the specification had applicability to orienteering terrains beyond Europe. We actually had gold mining terrain in mind when making that demand/request. The challenges with converting existing granite terrain maps (conversion of boulder fields and cliff edges) only became fully apparent later with the arrival of compliance testing tools and our first regional carnival.

But then, my memory may be influenced by later events, such as the development of compliance testing tools and our first regional carnival.
Aug 19, 2020 2:10 AM # 
robplow:
Ok perhaps 'demand' was the wrong the word. I remember reading an Australian submission for what became ISOM 2017 which argued very strongly that it was essential to allow stony ground dots in a straight line. I remember because there were several map samples supplied as evidence of this (all spur/gully, not mining as I recall) and when I looked at them I thought that in all cases it would have been no problem at all to draw these thin strips of stony ground as I have shown above with dots offset a bit,

But I agree the current wording (should not as opposed to shall not) seems reasonable to me. If you really feel making the dots into a straight line is the best way in some cases that's fine. So long as there is no chance of confusion with veg boundaries. But that is mostly unlikely in that sort of terrain. On the other hand - the fact that stony ground dots have been arranged in lines in one place on map may mean the mapper decides not to draw a distinct vegetation boundary in another place where otherwise it would be normal to use that symbol.
Aug 19, 2020 2:17 AM # 
robplow:
I was in NY n 92 training on maps in and around Harriman. These were pre the WOC 93 maps. Yet they all had accurate mapping of the mountain laurel. I don't think that was a Steve Key innovation - with vegetation that thick and with such definite boundaries and those boundaries interacting with very detailed contour and rock features, it is just obvious it has to be mapped accurately and had been mapped that way before he arrived - as far I could tell.
Aug 19, 2020 2:46 AM # 
TheInvisibleLog:
Just being politic Rob. Looking back, the main lesson I take from the experience of consultation was that we did not really have much idea of the implications of the new specification when we were commenting. I know I didn't. Most comments focussed on the symbology, the obvious. We remained silent on the issues of legibility and scale that arise from defined minimum separations, out of our inexperience. To my mind that was the heart of the reworking of the specification. Knowledge came with experience of using the specification and using digital testing, in my case in mining terrain. Although the specification was still officially a "living document", I think that our (my?) experience came too late. More recently there were plans to jointly explore the issues with the MC through a funded field visit, but the virus put an end to all that. Just one of the millions of unanticipated impacts of Covid.
Aug 19, 2020 6:14 AM # 
Jagge:
To me the confusion with veg boundaries is mostly caused by dots being evenly spaced. An area of dots aways has an edge, and dots at that edge always forms a line. To me the key is variation in distances between points at teh edge. Evenly spaced dots at that edge makes it look like vegetation boundary. Even robplow's nice ilustration would look a lot less like veg boundary by adding some variation to distances between points. I believe Manudona's example suffers from lack of this variation, not points forming lines.
Aug 19, 2020 10:53 AM # 
LOST_Richard:
Interesting discussion,

As a point of history I "discovered" Peterdine and negotiated access and supervised and checked Eric's mapping at the time. Also set the courses with Noel Schoknecht and Dave Pass as Controller and we are all still active orienteers.

It is too long ago to recall all the detailed changes but we worked well together and also created the Ngangaguringguring Hill map for the Easter 93 Carnival.

https://gracemolloyorienteering.wordpress.com/page...

Great to run there again last Sunday

I have yet to trip over the funky contour line :-)
Aug 19, 2020 10:58 AM # 
robplow:
Interesting Neil - I sort of look at all of that from a different perspective.

I realise all this compliance stuff is something you have a particular interest in, which is good, but I think that stuff can go too far and become counter productive.

Look at this section of Kooyoora (ISOM 2000). How far would go to make that ISOM 2017 compliant? You would have to all change all the boulder field triangles to just two the sizes (standard and 120%). That would be a lot of work and is not. It is not like the old map was ever ISOM 2000 compliant in that respect - there was only a single size of triangle in that as well. Also I am guessing there would be some minimum size violations : too small bare rock areas, the very short cliff on a brown dot knoll in the light green in the eastern part of that clip - that has to be under the minimum length. None of those seem grievous enough to worry about in my book.


(The map clip I carefully selected gets cropped when it displays in AP - if you click on it you can see the whole clip and the full range of triangle sizes I was trying to demonstrate - there are some very large triangles in the SW corner)

The important thing is good map that is accurate, understandable, legible, etc. The rules are intended to help with that but they are not perfect and cannot cover every eventuality. Sometimes there are contradictions and unintended consequences that mean blind adherence would be detrimental to the aim of the best possible map. Some flexibility is essential.

That would be my main criticism of the changes made in ISOM 2017: this tendency to stricter and stricter rules rather than giving smart experienced mappers some flexibility to exercise their skill and experience. Instead of focusing on blind compliance we should be able to look at map and say : ' Yeah that makes sense to me.The rules have been bent in a few places but that was done to make the map better so lets not worry too much'
Aug 19, 2020 2:08 PM # 
EricW:
Re credit for WOC93 Mountain Laurel mapping-
In case there is any doubt, I am quite happy to see other maps and mappers in the discussion, especially Harriman related.
Aug 19, 2020 2:11 PM # 
EricW:
"... main criticism of the changes made in ISOM 2017: this tendency to stricter and stricter rules rather than giving smart experienced mappers some flexibility to exercise their skill and experience. Instead of focusing on blind compliance...

Amen
I hope some relevant people are listening.
Aug 19, 2020 6:10 PM # 
yurets:
There's something about Peterdine...
Aug 19, 2020 10:29 PM # 
tRicky:
I doubt that the blind would do very well at orienteering.
Aug 20, 2020 12:36 AM # 
TheInvisibleLog:
I am not going to take issue with any of the arguments you have made Rob. First, my interest in compliance is driven by my "position" as mapping convener. I ended up in the position through the "last man/woman standing" principle and it took me a year of being requested before I agreed.

Besides organising a seminar every now and then, the main role seems to be to approve or reject requests for map variations from specification for major carnivals. I decided if anyone needed to understand the ins and outs of compliance and conversion, it needs to be me, until I wriggle out of the role somehow. Believe me, if someone contacted me today and offered to take on the role I would hand over in an instant. But I feel it would be poor form to resign of my own accord before doing my share of tenure, if only in recognition of those who held the role before me for quite a few years each (looking at you Noel, Adrian). And in the few years in the role I have learnt the need to be diplomatic on these compliance issues. So what follows is a form of walking on eggshells.

Compliance in Australia is legalistically problematic. There are many more maps like Kooyoora that are legible, but were not even compliant with the previous standard. I am struggling to think of a carnival since my tenure began where I have not been asked to approve a variation. Sometimes its minor such as keeping square end cliffs and non standard scalenes. Sometimes its larger, such as a map having essentially been mapped at 10k and needing to be extra enlarged to be legible. Generally there are a few months until the carnival, and my choice is to approve and take the criticism that I allowed a non compliant map to be used against the rules, or to reject and throw the carnival schedule into disarray. So far I have always chosen the former. The closest I have come to rejection is with a recently postponed carnival where I have said that my approval at the early date did not apply to the new carnival date because there was now time to improve the map.

You could argue that my situation is a result of poor process and that Australia needs a pre-approval process well before the carnival date. I think Sweden has one. But I don't think this is a simple administrative change. Map conversion is a rules issue, but also a financial issue, particularly for smaller orienteering communities. And it is my belief that most who maintain our rules do not yet have a sufficient understanding of issues with map conversion. Likewise, those who organise carnivals are generally also unaware of the issues. I thinks its unfair to ask a sole mapping convener to make a series of these decisions without an informed consensus on variations in the wider orienteering community.

My response to this situation has been to push the implications of conversion so that we can have an informed conversation about when deviations are acceptable. There are some in the room who maintain that none (or very few) are acceptable, and I suspect they view my work with objective compliance checking with suspicion. There are others who take a view that Australia rewrote the rules with granite mapping in 1985 so who are these Europeans telling us what to do. We know how to map granite etc. They may well view my interest in objective testing as a waste of time.

So far I have submitted segments of 40 maps of varying terrain and complexity through objective testing using OCAD and CheckOMap. I have remapped a gold mining area with as close to ISOM compliance as I can get (multiple rounds of testing and adjustment). I am in the process of mapping another sluice mining area at 15k and 10k for instructive comparison purposes. I have come to a few conclusions that not everyone agrees with.

1. The heart of the new specification is the legibility requirements with digital printing. Minimum separations are critical. The denser the features on the map, the more critical. I see few (maybe no) reasons why one should not comply with this requirement. Minimum areas and lengths are also important but raise some challenges with some mappable features - short but obvious water courses in mining terrain and smaller but very obvious bare rock areas in complex granite where there isn't room for exaggeration. The MC compliance test using OCAD looks at the minimum area and length compliance, but as far as I understand, is subjective on the separation issue. I am pretty happy with the latter position.

2. I am less concerned about some of the variations to symbol size such as boulder fields as mentioned in this discussion. The meaning is obvious. The main issue is that variations are legible. But variations in symbol size are quickly picked up in objective testing processes. You can't slide under the radar here like you can with separations.

3. Conversion of some of our more complex maps is not really possible if one is to strictly adhere to specification. No-intervention CRT conversion will produce thousands of non compliance points per square kilometer. Conversion actually means remapping, and that becomes a major financial decision. The example Rob gives of Kooyoora is instructive. It was first mapped for a WOC and remapped for a WMOC. I can't see any national carnival funding a remap. I doubt Australia is the only nation facing this challenge. If we continue to use this and other maps we are accepting a de-facto non-compliance. At least we can acknowledge this by deciding on the extent of compliance that can be afforded with conversions.

4. There are some terrains mapped in the past using symbol size reduction. My calculations are that the reductions equate to a mapping scale of about 12,000. Remapping to 15k will significantly change the map that orienteers see. Lots of detail will be replaced by brown dots. Mapping is not funded by the elites of the sport, but by regular carnival attenders. Their preferences should considered in the choice as to whether to adhere to the specification. Hence my test mapping of a complex sluice and dredging area at 15k and 10k. And no, its not Rowdy Flat.

I think my conclusions above are based on some solid investigation. I have been careful not to definitively say where the sport in Australia should place compliance in its rules. I only ask that any decision be informed, inclusive and convergent with reality. At the moment we are seeing an emergent compliance position determined by the hiccups of carnival organisation, the limited understanding of ISOM compliance implications and the unwillingness of this mapping convener to be prescriptive. Maybe one step is replacing the convener. Any volunteers?
Aug 20, 2020 2:01 AM # 
jjcote:
In contrast, in the USA there is no official oversight of mapping, or requirement that specs are complied with, whatsoever. If it's a national level meet, the sanctioning committee does have to give an okay if you use a nonstandard scale, but that's it.
Aug 20, 2020 2:26 AM # 
TheInvisibleLog:
And WRE?
Aug 20, 2020 3:32 AM # 
simmo:
Invis. is it the case that IOF checks all of our Group A event maps for compliance, or just the ones that are WREs and/or Regional (ie Oceania)?

If the latter, why not approve them all without too much bother? We could then use areas like Kooyoora and classic goldmining for national carnivals, with WREs held separately on nearby compliant maps. There would be no need for expensive remapping, and the 2017 conversion applied sensibly, largely retaining the original 'Australian-ness' of the granite and mining features.

My reading of the OA Competition Rules would seem to allow this, because Group A events are basically at your position's discretion.
Aug 20, 2020 3:37 AM # 
TheInvisibleLog:
WREs are the responsibility of the IOF adviser to submit the map to the Map Commission for review. If its not an IOF event then the MC has no role.

This means that what you suggest is indeed one approach, and you could argue that this is in fact what is happening. But as I said, its an emergent decision, partly dependent on who is the mapping convener. Perhaps leaving it to the somewhat random appointment of the convener is not the best way to come to this position. Or maybe its an inevitable outcome forced by cost.

It seemed sensible that the Australian rules say that all new maps for championships be ISOM compliant. If you are mapping new terrain, it makes sense in 99% of situations to try an comply as much as possible with ISOM. But we have more championships and fewer competitors than in the past and fewer quality new areas, so most carnival maps will be already mapped areas as a result of opportunity and cost pressures. Lets say Victoria chose to use Kooyoora at the next carnival here, in say 2032 when we have a vaccine. Do you use the existing version, update basics such as track and veg, or convert partially or fully to the new symbology and standards? When does the map become "new"? Lots of grey in there.
Aug 20, 2020 3:42 AM # 
simmo:
Well it sounds like you are there for a couple more years at least, then we need to find a like-minded replacement. Perhaps robplow is returning to Oz?
Aug 20, 2020 3:47 AM # 
TheInvisibleLog:
Yes, about as long as it will take for all State border quarantine controls to be lifted I suspect. Perhaps the next time we really have to face this is with the next Oceania Carnival. Maybe the NZ travel bubble will be even longer to appear. ;-)
Aug 20, 2020 11:34 AM # 
jjcote:
Honestly, I don't know what the USA does about WRE maps, because I haven't been directly involved with that part of organizing a WRE, at least not recently. Maybe the map gets submitted to somebody at the IOF? But if there's a compliance officer here domestically, I don't know who it is, and that seems unlikely, although I can think of a few people who would be qualified for the position. We used to have a Mapping Committee, or at least a Mapping Committee Chairman, and a registry of maps (well, a list, anyway), but the last chairman abdicated about 30 years ago, saying that the position was not needed.
Aug 20, 2020 12:13 PM # 
blairtrewin:
Unless things have changed, maps for high-level events (Regional Championships, World Cups and above) have to be submitted to the Map Commission for compliance testing - although, as we found out the hard way last year, just because something gets through the tool doesn't mean everything has been picked up - but "ordinary" WREs are the responsibility of the Event Adviser.

It would be a while since the US has staged a WRE outside of NAOC, wouldn't it?
Aug 20, 2020 10:39 PM # 
TheInvisibleLog:
I think that depends upon the tool you use. One tool would have picked that up (assuming you are referring to a thin olive green shape).
Aug 21, 2020 6:47 AM # 
gruver:
Around here there's no talk of compliance or checking. A concern of the Mapping Committee is slavish adherence to the sprint specification for steep, green club events within city limits. Some maps are unreadable. Isnt life funny.
Aug 21, 2020 8:31 AM # 
robplow:
unwillingness of this mapping convener to be prescriptive. Maybe one step is replacing the convener
Don't sell yourself short - I think what you have just described is a very well thought out, reasonable and pragmatic attitude to it all. Far preferable to the hardline ideology of 100% compliance, no exceptions. Which, quite apart from the arrogance of it, is highly impractical as you have eloquently explained.

approve and take the criticism

My guess is that criticism comes mostly from other mappers not the "regular carnival attenders". That has certainly my personal experience: The general O public are happy with a map but other mappers just love to pick up minor 'deviations'

There are some terrains mapped in the past using symbol size reduction

That is certainly true of the much lauded Cascades map - which was mapped at 10000 (ie when printed at 10000 the symbols are not enlarged by 150%). No one seems to have ever demanded that map be redone at 15000 to be 'compliant'. And It remains (quite rightly) a favorite map/area of many.

I remember seeing a presentation at the beginning of the revision process that lead to ISOM 2017 which said quite clearly there was a problem with map quality and the way to fix it was to make ISOM very much stricter. I think is a big mistake. Certainly you need robust rules but we have had perfectly adequate rules for a long time now. The rules do need updating regularly of course but making the rules stricter and stricter does not necessarily make the maps any better. IN fact I think it can have the opposite effect because it takes away the ability of a good mapper exercise their skill and discretion. IF want to improve the quality of maps and events simply tightening the rules is not going to do it. The effort should be put into education/training.

Some suggestions:

1. more mapping workshops, - and those workshops need to focus on field survey not just drawing. At the moment all this talk of 'compliance' focuses only on the drawing - it is possible to send a map to the the MC and it come back as 100% 'compliant' and yet it could make no sense to anyone in the terrain. The field survey is typically over 67% of the work for a normal forest map yet all of this Compliance talk totally ignores all of that. Obviously checking of the map in the terrain is the responsibility of the controller/event advisor. Hence

2. better training of event advisers/controllers. It is very rare in my experience, even at international level, to have an event adviser/controller who is an experienced mapper. And often I have found their knowledge of the the intricacies of mapping worryingly inadequate. So I think event advisers/controllers need better knowledge about mapping issues. Perhaps the best way to tackle this would be to split up the the role of event adviser/controller between several people: make sure there is a mapping controller who is an experienced mapper and responsible only for the maps (and maybe courses) and have someone else responsible for controlling other aspects of the event. A major event these days is a pretty significant undertaking and it is unreasonable to expect any one person to have the competency and experience to adequately cover all aspects. I would feel quite comfortable as a controller responsible for mapping and courses but far less so for things like timing systems, general admin etc.
Aug 21, 2020 1:00 PM # 
robplow:
I think that depends upon the tool you use. One tool would have picked that up (assuming you are referring to a thin olive green shape).

We are talking about that Chinese WC middle map from last year.

http://news.worldofo.com/2019/10/26/world-cup-chin...

The thin strip of olive next to a path under the number '5'

Lets assume the strip of olive was 0.38mm - not the regulation 0.40mm. And lets assume the compliance software picked that up and so it was enlarged slightly to be 0.40mm. Now the map would be 'compliant' but would the problem really be solved by that minor adjustment? Would a 0.4mm strip be significantly more visible on the map than 0.38mm? Would that very minor change have made it any clearer to the runners in that race and mean that none of them would have inadvertently run across the olive green. OK if it had been 0.4mm it would have been possible to disqualify anyone who ran across it but that is not really the point. The point is to have maps that are not ambiguous, not having to disqualify people inadvertently breaking the rules in the first place.

Here is similar example:



A few observations:

The olive strip in my example is 0.4mm wide but the path is 0.18mm so it effectively covers 0.09mm leaving only 0.31mm. Is it still 'compliant'?

Where the contour crosses the olive green even less of it is visible. If it had been a index contour (as on the right hand side) it could almost completely obscure the olive.

The fact that the olive is bordered by green also makes it harder to see - if it had been white either side (as on the right hand side) the olive would have stood out better.

ISOM has nothing to say about subtleties like that. Just that the olive has to be 0.4mm wide. If ISOM did try to cover all eventualities like that it would just become bogged down in detail that no one ever reads.

While those compliance algorithms are certainly useful tools in helping to make a map fair and legible; they are in no way a substitute for human judgement, experience, common sense, etc.

And worse, I think there is a risk that if someone from the MC runs the compliance software and pronounces a map 'compliant', subconsciously the organisers/event advisers might think there is no longer anything to worry about regarding the map and let their guard down.
Aug 21, 2020 4:14 PM # 
jjcote:
My opinion, which I think Rob likely shares, is that compliance should be determined by an experienced person, and that the automated tools are useful to that person by indicating things to take a look at. But not all things that are so flagged need to be fixed, and some things that are not flagged do need to be fixed.
Aug 21, 2020 6:18 PM # 
EricW:
Second this as well
Again, is anybody listening?
Aug 21, 2020 11:42 PM # 
tRicky:
I am.
Aug 22, 2020 1:16 AM # 
TheInvisibleLog:
To be fair to the MC, their compliance checking process is only partly objective... scale, symbol size, minimum lengths and areas. The rest is a subjective assessment.

My preference would be for a more stringent objective compliance test as a first step, together with a subjective assessment that is prepared to accept non-compliance where non-compliance is judged to not impact the legibility of the map, or in some cases improve legibility.

Having been through a really stringent self-administered compliance process for one map (repeated submission to CheckOMap), my map legibility was improved. That might be because of my lesser experience (30 maps). I suspect there are many mappers with a similar level of experience so I am confident the process would be instructive for others. I also learnt where compliance made the map less legible. Its a very useful tool, but a guide, not the be all and end all. It should not be dismissed though.

Please login to add a message.