maptalk.co.nz News   |   Events   |   Forum   |   Photos   |   Links    

  Forum

Forum Home   Start New Topic   Edit Profile   Register  

1   2   3   4   5   6   7   8   9  

Mapping

Show Profile  Marquita G Posted: 16 May 2005, 9:56 AM  
I have a feeling Svend is referring to Spaghetti Soup as used for a 3 day at Queens Birthday in 2002 (?). I was responsible for the skinny contour lines on that occasion and in a sense this was an experiment on my part to try to represent the huge amount of detail on this map without the lines touching or obscurring detail. It could be argued that the area is over-mapped but as the cartographer, it is hard to know what to leave out. To my eyes the map looked great but I fully acknowledge that to older eyes it was difficult to read and I have not repeated the experiment. When the area was used again at Oceania I myself had trouble reading it due to very light printing (something beyond our control) but also the contours were too skinny again. I don't know what happened here, I wasn't involved with that event. To some degree there was a problem with "version control" of the map - the course setter had an Ocad version of it on his computer from an event he organised a year or two ago and simply used that rather than getting the correct and most updated version from the club's map custodian. Svend is also correct that most people, controllers included, don't actually know the correct specifications for symbol sizes for differing scales, something which should be rectified I guess. Another issue is that ocad is quite easy to bumble your way around and achieve what looks like a good result, but to do it properly and to get everything exactly right requires a considerable amount of attention to detail and experience of the potential pitfalls. Just this weekend I rode a MTBO event where the track symbols were far too thin on the 1:15,000 version of the map making them very difficult to read while riding - they were OK on the 1:10,000 but we forgot to un-tick the enlarge/reduce symbols box on this occasion. We live and learn!

Show Profile  Michael Posted: 16 May 2005, 2:46 PM  
These experiences confirm that the way I do it is better. I tell OCAD the map is 1:15,000, and merely print at 1:10,000. I still have to do a layout for the paper size and stick in more north lines but no chance of overlooking a tick box:-))

I suggest that event standards and controller knowledge is a subject in its own right and should be in another section of the forum. Make it under "Technical" where this one should have been:-))

Show Profile  Bryan Posted: 16 May 2005, 5:06 PM  
Michael suggests that if it is hard to map something then don't map it. This seems strange to me - maybe Michael can explain more - but if something is hard to map, to me it just means that we have to spend more time mapping and making it clear and consistent to the Orienteer. I'm very reluctant to leave something off.

In my mind there are two poles of mapping styles - those that like to generalise a lot and those that like to add in a lot of detail. I've always tended towards the 'more-detail' map as I think it creates a fairer map (as long as its clear to the Orienteer). On some maps, I've had a devil of a time trying to fathom what was left in and what was left out. (eg a 3m knoll mapped but the 1.5m knoll beside it left out)

Micheal also suggests that footpaths alongside roads should not be mapped. Again I would be reluctant to leave it out - on 1:15000 maps maybe, but on a 1:5000 map probably not. We don't want Orienteers running down urban roads when there is a footpath nearby.

I can remember mapping a Park World Tour Sprint race in Bangkok, Thailand and leaving off footpaths. The fieldchecker (and experienced Sprint Racer mapper) added on all the footpaths and curbs (some less than 1m wide) with no loss in legibility and the map was better because of it.

Show Profile  Michael Posted: 16 May 2005, 5:33 PM  
And on some maps I've had a devil of a time figuring out where a mapped feature is, only to conclude the mapper has marked a subliminal rise in the ground or an ankle-deep hollow.

I like Eric Andrews' philosophy - which (rocks, but it could be knolls and depressions) put their hands up and say "pick me!" They go on the map, other things are put in till you get down to the minimum size OR the features won't fit legibly. Then you stop. The old concept of "consistency across a map" has gone out in favour of "significance in context". A footpath close beside a road in my view is insignificant in its context even at 1:5000. I often leave off fences close beside roads on 1:15,000 mapping.

Show Profile  Jamie Posted: 16 May 2005, 8:26 PM  
Whats a footpath?

A "significance in context" philosophy leaves me uneasy as it seems to add an extra subjective element to mapping and could create more inconsistencies.

I like knowing that a knoll/rock/depression/gully/distinct tree of such and such a size will or won't be mapped and believe it aids interpretation of the map as a whole.

My preference is also for more detail and lots of balconys - as long as the map is accurate and consistent

Show Profile  Svend Posted: 16 May 2005, 9:11 PM  
Thank you Marquita for all that information. The map referred to was actually named Stags Roar. The adjacent map Woody Pot was used the next day and both had thin contours but not to the same extent as for the Oceania event and it was only in the green stripe area it was difficult to read. The Spaghetti Soup map was used at the ANZAC
4-day event in 2003 and it had excellent cartography.

Show Profile  Alistair Posted: 16 May 2005, 9:20 PM  
...also agree with Jamie on this. Generally better with more detail, with accuracy and consistency.

It's up to the course setter to avoid putting controls near or in ambiguous areas or areas where the control placement can't be descibed correctly with control descriptions.

It's up to the runner to generalise the mappers interpretation and filter extraneous information to opitmize their execution of a specific leg.

Having run on many of Bryan's maps I can say I've never ever had any problem with them - they have always been of world-class quality.

As for footpaths/sidewalks/pavements/pedestrian ways (...) - on 1:15000 - no; but on 1:5000 it should depend on what is appropriate - if they are a separate entity to the road then maybe they should be on, but if they are just a strip of asfalt beside the roadway then maybe they should be left off, or included as part of the road-way.

Curbs?! Must be a pretty big curb!

Show Profile  Martin Posted: 16 May 2005, 10:44 PM  
I was referring to when a concrete footpath runs parallel to a sealed road, and there is no grass verge. and on a 1:5000 or 1:4000 map

On the Carrington map I made up my mind that anything that was two pavements (eg footpath + sealed road) side by side would just be mapped as one for ease of reading it. anything with a grass strip in between I mapped separately.

It's interesting to read the differing opinions of some of NZs top mappers

Show Profile  Michael Posted: 18 May 2005, 1:18 PM  
There aren't two poles of mapping styles, there's a continuum. But whatever their style mappers must confront two realities:

In areas of intense detail, the diameter of a small boulder symbol is (for non-sprint mapping) 6m on the ground, a knoll symbol is 7.5m on the ground, a depression (the most awkward) 12m on the ground. In other words several double paces across. If the features are closer than that, you have to leave some things out. (There are a few opportunities for "exploding" the positions outwards.) Eric's "pick me" principle is a useful guide to me when I'm mapping AND when I'm competing.

In areas where there isn't intense detail, there are minimum sizes - 1m high/deep for boulders, rock faces, knolls and depressions. I lower the thresholds a bit in the case of particularly distinctive features in the middle of not much else.

These two principles are what I mean by significance in context. I agree Jamie that we should adhere fairly rigorously to these minimum sizes. NZ maps have thousands of boulders, rock faces, knolls and depressions that are nowhere near 1m high or deep.

Now back to the canopy thing. No matter where on the style continuum, and how long you spend with your pencil and rubber, there are some places where the mapper must conclude that its impossible to depict the terrain legibly and unambiguously. Places you yourself would not put a control. But down the track an inexperienced planner is going to delight in using the "nice tricky control site". Such places are better with less detail to forestall this.

Because the sprint specification is new, and because the multilevel issue is a curly cartographic problem, I would expect to find these difficult situations more commonly in park and campus maps. Yes its worth trying to find ways to depict them. And leaving them off, generalising them into a building etc is also an option. That quote in the ISSOM: "A map with few well chosen features will give a much better map than a map cluttered with many insignificant features."



Show Profile  Svend Posted: 22 May 2005, 10:08 AM  
Some mappers may not be aware that OCAD9 Beta is now available for download at http:/www.ocad.com/beta9/

There are some interesting hints on cartography at
http://olles.bloguje.cz/1_tema.php

Show Profile  Michael Posted: 21 April 2006, 2:00 PM  
In the thread on the nationals Bryan reminded me of this one.

I don't leap straight in after events because without exception there has been a huge amount of work and the organisers need to know that it is appreciated.

But some earlier events now deserve comment. I am sick of over-detailed mapping, done by using standard symbol sizes at 1:10,000. Mapping (other than for the sprint) should be done "as if for 1:15,000. I think that the mapping at Mauriceville (WOA OY2) may be a culprit, the Katoa Po map. I seem to run at Taupo mostly at night now, and the night relay has become a matter of stop; study the map; and then run on memory. I don't enjoy that type of orienteering.

Mapping for 1:15,000 CAN be done - see Naseby.

Show Profile  Jamie Posted: 21 April 2006, 2:43 PM  
yeah good point, then again it is tempting when thinking about mapping in this way not to map detail and therefore removing the fine navigational challenge...ala nationals classic 2006. Full kudos to those involved for finding and mapping a fantastic area, but imagine if there was $10, 000 more to play with...

Show Profile  Nah Posted: 21 April 2006, 4:25 PM  
Nah!

"These experiences confirm that the way I do it is better." That's modest, Michael. I for one don't like your mapping style, especially in Woodhill, and most people I know don't. Of course you're going to complain about over-detailed mapping - you're Mr. Leave-It-Off-If-At-All-Possible himself.

Leaving detail off is stupid unless it does not hinder orienteering. There is almost always room for more detail on New Zealand maps when compared with the features on the ground. Some people seem to have found a balance of what to show and what not to show, others have not...

And as Jamie said, how easy was the Classic map at Nationals. That's not to detract from a great event but I didn't leave much room to make mistakes - it was almost impossible to do so.

So Nah!

Show Profile  Jamie Posted: 21 April 2006, 7:02 PM  
yeah, I didn't say it was easy, it might have been for everyone else but I lost 15 minutes. I think I light detail

Show Profile  rob.g Posted: 21 April 2006, 7:19 PM  
Hey nah thats way over the top with the criticism of Michaels mapping. The problem with the Goblin type maps is they are at 5 metre contour interval which Michael was asked to do, and the maps were made when the viz was extremely low.
With the Kareta area Michael mapped mostly a depression style which was not poular, and he knows it.
Naseby was excellent except for the part that wasn't remapped due to funding limits.
Overall we are lucky to have the mappers we have, but one problem is they are all ageing.


1   2   3   4   5   6   7   8   9  


Ruffneck Productions © Ruffneck Productions maptalk.co.nz