We currently mostly map lands, but still forget to map the navigation on rivers and near coastlines with enough attributes.

Notably, there are missing objects : rocks, bridge pillars, man-made islands, separation dikes, pontoons built over waters, cables, even if we still have fences (like those used as antipollution-filters), and all other kind of objects that make obstacle to navigation.

As well, there's no way to clearly indicate the direction of navigation (notably on rivers, or under the pillars of bridges, or near sluice doors), except if we trace maritime routes (which are only appropriate over the sea, but not really for rivers, except possibly under bridges on large rivers to separate the trafics).

Could these be mapped correctly ?

Note: islets in the middle of a river have a coastline, but the default mapping treats closed contours of these types to be filled as if they were lakes, even if they are already on the middle of a lake or river.

We need a way to map areas in the middle of waters excluded from navigation (such as shallows, sandbox under water by tides, or submarine installations, or other invisible obstacles, or regulatory restrictions).

Should we map them ? How can we do that ? What is the support for these type of objects in existing map styles for generating the tile bitmaps, starting at zoom level 15 or higher ?

asked 12 Jun '11, 04:20

Verdy_p's gravatar image

Verdy_p
14181115
accept rate: 0%

edited 27 Nov '12, 10:03

When further discussion can be helpful you can also consider to open a thread about that routing topic at http://forum.openstreetmap.org with better possibilities than here (FAQ site) to ask and answer in detail step by step.

(15 Jun '11, 19:08) stephan75

Your question has a lot of components, so I'm going to do my best to tackle each part.

Could these be mapped correctly ?

One can certainly map the maritime routes, but doing routing over open water, I think, would be as complex as mapping routing of a person walking in an open field. There's really not a good solution to that, but that doesn't mean we can't (and shouldn't) map the features you mention.

I suspect the reason they're not mapped much right now is the limited number of OSMers with a boat.

We need a way to map areas in the middle of waters excluded from navigation (such as shallows, sandbox under water by tides, or submarine installations, or other invisible obstacles, or regulatory restrictions).

This sounds like it deserves a longer, more in depth discussion, such as one might find on the various OSM mailing lists, such as talk, or tagging. That's a better forum for these kinds of open ended discussions.

Should we map them ?

If they're generally useful, and "ground verifiable", yes.

I think the "how" might be a bit more complex, but OSM certainly has some maritime features, such as boundaries, docks, slipways, moorings, etc.

What is the support for these type of objects in existing map styles for generating the tile bitmaps, starting at zoom level 15 or higher ?

The question of whether to map something should be kept separate from the methods of rendering them. We map things we don't render, or don't always render. The strength of the project lies in this separation between data that's collected and data that's displayed.

permanent link

answered 12 Jun '11, 06:55

emacsen's gravatar image

emacsen
1.2k1423
accept rate: 13%

2

... and of course this sort of question has been done to death on the mailing lists in the past. Try searching for e.g.:

seamap +site:lists.openstreetmap.org/pipermail/talk

(12 Jun '11, 09:49) SomeoneElse ♦

Your reply was much too vague to be usable, if all it indicates is that we need to search within thousands of unsorted messages (with new ones appearing at all times). The pipermail interface does not allow easy seraching in its huge collection. If these questions have been replied many times, it was because it was not easily solved, requiring us to complicate the geometries with arbitrary segments added in order to build adjascent polygons with the same tags, only because of a small islet or a bridge pilar in the middle of a river...

(28 May '12, 08:08) Verdy_p

The "e.g." was supposed to be a clue that you could try using Google or similar to search for whatever you want that has been discussed on the mailing lists. For example, searching for "pontoon" in this way gets a page and a half of hits on 3 or 4 lists - probably half a dozen conversations at most. Is that really so difficult?

(28 May '12, 15:45) SomeoneElse ♦
1

You might be better taking such a wide-ranging question to the forum or mailing lists where they can be discussed. This site is intended for straightforward, single questions that can answered clearly. This will help the questioner and others looking for an answer too.

(27 Nov '12, 10:12) ChrisH

For islands / rocks in rivers or lakes, these need to be mapped using a multipolygon relation. See the instructions on this page for Creating islands in lakes. Its the same for rivers, but add the waterway=riverbank ways as the outer instead.

Then the islands will be shown correctly on most renderers.

permanent link

answered 12 Jun '11, 12:24

Vclaw's gravatar image

Vclaw
9.1k891140
accept rate: 22%

Yes of course we cannot map all routes possible on the sea and rivers, unless this is a regular public or commercial line such as regional or cross-border ferries or "river buses" in large cities (most often, even if thesse services are commercialn they are operating with public subsidies and under regional supervision for the concessed public domain on sea borders and rivers), because they complete the terrestrial network for the general public, where this is the only available or most practical route.

But areas excluded to navigation could easily be mapped, because there are either physical elements (natural rocks visible by satellite, man-made banks for bridge pillars on large rivers) or created by legal restrictions (e.g. on the English Channel, and West of Britanny, for oil/chemical tankers and cargos, or due to maritime natural reserves) or regulation restrictions (direction of navigation under bridges).

Other things we still don't map (and that are present on most maritime and river maps) are the indication of the signal type, angles and range of coverage of lighthouses, oyster parcs, areas dicovered on low tides, and other sand banks, areas reserved for swimming and limits for permitted accostages on beaches.

Should OpenStreetMap cover the navigation conditions on seas, lakes and rivers ? Just like we propose extensive ground coverage for motorized vehicles, info for public transportation, some info for bicycles and locating shops/services, and even fewer info for pedestrians.

Initially I gave tis question because of islets and rocks, which are immediately verifiable, and currently not easily mapable on rivers. This s still difficult to represent with the proposed items.

And marsh areas often include large areas that are not navigatable at all. They also ocur in cities, within parks near many lakes, and along rivers. They are still interesting to map due to the unique natural landscape they provide (and often these wet areas are protected by laws or regional and international programs). These areas could be specially hatched be ome pattern on top of water areas, even if we can still, right now mark a single point of interest for selected (and named) natural reserves, using an icon.

permanent link

answered 12 Jun '11, 09:39

Verdy_p's gravatar image

Verdy_p
14181115
accept rate: 0%

Initially I gave tis question because of islets and rocks, which are immediately verifiable, and currently not easily mapable on rivers. This s still difficult to represent with the proposed items.

Why are they difficult to represent? What proposed items?

(12 Jun '11, 13:19) emacsen

The items documented in OSM, and either proposed in the JSM interface, or manually addable as "custom" properties supported by a documentation and by some graphical representation (coloring, icon) in renderers. There's still nothing representing these rocks or man-made constructions, and nothing documented about navigation regulation.

(15 Jun '11, 12:33) Verdy_p

We are currently proposed to use multipolygons sharing some common paths to join them. It would be much simpler to just draw these items in the middle of a river without having to split it at arbitrary positions in multiple paths, with the implicit effect on the containing water area to make these elements as inner ways, without using complex markup, just like we can add items in the middle of a land area (or may be a bot could add the implicitly implied properties to make them inner ways in an outer way for water areas)

(15 Jun '11, 12:35) Verdy_p

I consider this subject solved. Navigation subjects are now very well handled and demonstrated in the OpenSeaMap projet, which also documents many tagging practices on OSM. It requires many specific tags not used on land (up to various details of the seamarks, navigation areas, security restrictions, protection of the environment, regulation of anchorages, speeed limits, in natural parks or in cultural heritage areas like Venice).

It's curious that nobody replied with OpenSeaMap here.

permanent link

answered 27 Nov '12, 09:57

Verdy_p's gravatar image

Verdy_p
14181115
accept rate: 0%

Your answer
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here

By RSS:

Answers

Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text](http://url.com/ "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Question tags:

×39
×10
×3
×2

question asked: 12 Jun '11, 04:20

question was seen: 6,083 times

last updated: 27 Nov '12, 10:12

powered by OSQA