Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Mapping missing roads in cities of Malaysia, Indonesia and Vietnam #32

Open
GRABOSM opened this issue Jul 3, 2019 · 7 comments

Comments

@GRABOSM
Copy link
Owner

commented Jul 3, 2019

Objective: Grab team has mapped these regions in the past and this time we are going through the existing road geometries and add the missing ones if any. We'll be reviewing and mapping the following features: missing roads, road classification, and check for intersection configurations and alignment accuracy.

800px-NorthSouth-Expressway
Source: By mailer_diablo - Self-taken (Touched Up), CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=553058

Regions we are reviewing/mapping:

Regions we are reviewing/mapping:

  • Malaysia

    • Sandakan
    • Tawau
    • Kuantan
    • Alor Setar
    • Sibu
    • Langkawi
  • Indonesia

    • Yogyakarta
    • Kendari
    • Bengkulu
    • Makasar
    • Bandung
    • Bali
  • Vietnam

    • Khanh Hoa
    • Ho Chi Minh

Workflow: We'll be following the same workflow as mentioned in our wiki - https://github.com/GRABOSM/Grab-Data/blob/master/Data%20Improvement%20Projects and https://wiki.openstreetmap.org/wiki/Grab#Process_Flow

Validation: We'll be following the validation guidelines mentioned in our wiki: https://wiki.openstreetmap.org/wiki/Grab#Quality_Process

Changeset Comment: Mapping missing roads and road attributes https://github.com/GRABOSM/Grab-Data/issues/32

Source: Digital Globe Premium -- we would refer to others if need be.

Filter: None

Plugins: None

Edge cases: None at the moment, we'll post it as a subsequent comment if need be.

Team involved: The Grab Data team

Do reach out to us if you have any questions or suggestions on the same. Happy Mapping!

cc: @jinalfoflia

@MASangys

This comment has been minimized.

Copy link

commented Jul 4, 2019

Hope that Grab team can add Seremban in this list too because I noticed area within Seremban District lack of road data compare to other cities too.

@GRABOSM

This comment has been minimized.

Copy link
Owner Author

commented Jul 8, 2019

@MASangys thank you for the suggestion, have noted this. We'll soon be updating the list with a few more cities where the team will be mapping.

@GRABOSM

This comment has been minimized.

Copy link
Owner Author

commented Jul 11, 2019

We have updated the original list with a few more cities. Here are the new ones:

  • Indonesia
    • Makasar
    • Bandung
    • Bali
  • Vietnam
    • Ho Chi Minh

If there are any questions, suggestions or recommendations related to these regions, please do let us know.

@GRABOSM GRABOSM added the In Progress label Jul 12, 2019

@AkuAnakTimur

This comment has been minimized.

Copy link

commented Jul 30, 2019

In the state of Kedah, Malaysia; while I'm trying my best to respond to changeset discussion left by the team; I reckon a serious issue was discovered.

Outside urban areas, usually there would be a huge area of paddy fields. Sure, it would indicate areas that would be likely unmapped. Example, this place (aerial imagery available to be exploited for OpenStreetMap) which is about 30km away from the state capital, Alor Setar.

Typical highways around this area includes highway=unclassified, highway=residential, highway=service as feeder roads to one of two houses built around the area, and probably highway=track that potentially make up most (unmapped) highways in this said area.

highway=track may pose a challenging task to be mapped. It may be erroneously mapped as a highway with a higher classification or hierarchy. I've spotted this error from the mapping team, and left some discussion on a changeset as well.

These are usually can be found randomly next to canals, slithering around the paddy fields. It may be paved with rough gravels or simply just bare ground.

Hopefully these issue will be rectified by the team, because apparently I can see a lot of major highways were mapped and they started to appear quite densely on the OpenStreetMap.org website too.

P.S. I'll try my best to respond to discussion left by the team, one by one.

@GRABOSM

This comment has been minimized.

Copy link
Owner Author

commented Aug 1, 2019

Hello @AkuAnakTimur, thank you for bringing it to our attention. We mapped these roads based on the historical data, that is, previously the local mappers have added these to be unclassified roads and we did not want to abruptly change the classification.

Also, the OSM policy for an unclassified road in rural areas states this:

"In some countries, unclassified roads may be unpaved in larger, poorer or more remote/rural areas, and are typically paved in denser, richer or more central/urban areas. In this case, you should add a surface=* tag to indicate the surface.

Can you help us better understand why were they mapped/tagged as unclassified routes previously? As well as the reason they were considered as tracks now. Since one of the roads connect to major roads, we'd want to retain the connecting roads to be unclassified itself.

Thanks again for the feedback and our team really appreciate your responses and help as they are helping us better our approach.

@AkuAnakTimur

This comment has been minimized.

Copy link

commented Aug 12, 2019

Hello, thank you for the reply.

Can you help us better understand why were they mapped/tagged as unclassified routes previously? As well as the reason they were considered as tracks now. Since one of the roads connect to major roads, we'd want to retain the connecting roads to be unclassified itself.

It's undeniable that some highways across rice fields can be mapped as highway=unclassified. It's very likely that I have chosen the route with a better surface to go through (paved with asphalt) on my humble moped, surveying for OpenStreetMap as a hobby. Hence, the team may have stumbled across some highways mapped as major or significant highways (highway=unclassified and to some extent highway=residential).

image
Image source

Puzzles may appear at some certain places: some streets or paths across these agricultural landuses may not be highway=unclassified or even highway=residential.

image
Image source, with coordinates

It may appear from imageries that these highways: ways 706559610 and 708427243 ; (in OpenStreetMap terminology) appear suitable for motor vehicles. However, from ground truth (see above picture), it's not the case. It may serve as a driveway (highway=service + service=driveway) to the only house around this very small area.

Those two highways were changed back into major highways after I responded to one of your team's edit.

The two highways - ways 708700387 and 708700386 - just opposite to those first two, were mapped incorrectly. Good news, corrections were applied after I've left a discussion on one of the many changesets.

From ground truth, here's what actually the latter two look like:


Image source, with coordinates

In a nutshell: these situations show how tricky it could be to map across the paddy fields in the state of Kedah, Malaysia. I have provided some examples where the team actually would really have to seek opinion from at least a local mapper: everyone wants minimal inaccurate data in OpenStreetMap.

There may be more mistagged highways across the area of the team's editing interest, but I will be happy to let the team know sooner or later.

From the forum, if the team is somewhat rushing for a deadline, the highway=road tag is simply perfect. The team may return later to further scrutinize the data, tagging things more appropriately.

If these tricky ways need to be classified, by hook or by crook, in reality there's no shortcut for that. I might be very wrong too. Look for hints in the imagery, especially the density of any nearby dwellings. Lower the classification of any ways if it reckoned not to be very important in terms of traffic.

I will comment more on other related issues - allow me some more time.

@GRABOSM

This comment has been minimized.

Copy link
Owner Author

commented Aug 24, 2019

@AkuAnakTimur thank you so much for the detailed explanation. The team will make the changes as per your suggestions and update the data accordingly.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
3 participants
You can’t perform that action at this time.