Hello Xhenz.

I noticed that you’ve been reclassifying trunk (highway=trunk) and other lower classification ways to motorway (highway=motorway).

Some of them such as (way/229282369) and (way/320520498) don’t fit the definition of motorway which is defined as “generally resemble a controlled-access road and are free from intersection controls (such as traffic-signals, rotaries and roundabouts, and stop signs or stop markings)” and “have access roads only for servicing purposes, such as a limited number of driveways serving non-residential property, service roads to infrastructure, or rest-area roads or bays, which may be more typical in regional or rural settings.”
We can see that while there are some controlled access ramps, there are connections to residential/industrial areas at grade and alongside the length of the motorway such as via (way/730333302) and (way/320515634). Looking at existing motorway classifications with ref=A1 from (way/1308323479) to (way/1218094435), we can see that they’re accessible only via controlled access and have higher maxspeed associated with motorways.

If you disagree I’d be more than happy to discusss with you. Thank you and happy mapping!

Hello SuadMiftari!

I previously remodeled these trunk links (way/1417425694) and (way/693033035) so they would be a separate dual carriageway. I noticed they were remodeled into overlapping ways which is triggering validation errors. Currently, it appears as if westbound drivers on (way/693033035) are driving into eastbound (way/1417425694). What do you think about separating the ways from each other like previous modeling?

Thank you and happy mapping!

Hello Sam Jayah,

It would be helpful if you could provide the specific documents you are referencing for these changes or links to them so we may align our efforts with local standards. It would also be great if these sources could be documented on the Sri Lanka OSM Wiki, as it would improve transparency and make it easier for other contributors to follow the same approach.

I would also suggest outlining your proposed classification strategy (especially for road classes and refs) on the OSM Wiki or share it on any of the main Sri Lanka OSM community channels. This will help ensure alignment and open discussion with the wider mapping community.

Thanks for your time, and I look forward to hearing from you!

Hello Sam Jayah,

I noticed you’ve been updating ref tagging in Sri Lanka and referenced the Road Development Authority, so I wanted to ask about a couple of specific edits. For way (way/151574160), the ref was changed from AA004 to A4. However, Mapillary imagery (https://www.mapillary.com/app/?pKey=459242298703748) shows the route signed as AA004 (from 2018), and the RDA website (https://rda.gov.lk/index.php?option=com_content&view=article&id=24&Itemid=123&lang=en#class-aa-roads) also lists it as AA004.

Additionally, I saw ref=B653 added along the dual carriageway from (way/258498475) to (way/884175132). However, Mapillary imagery from November 11, 2023 (https://www.mapillary.com/app/?pKey=1491813521721679) doesn’t appear to show a route number on the signage there. The RDA site (https://rda.gov.lk/index.php?option=com_content&view=article&id=24&Itemid=123&lang=en#class-b-roads) lists B653 as “Battaramulla Bypass Road,” while the ways it was applied to are tagged as “Polduwa Road.”

Would you be able to provide any insight or resources used to add these refs? Thank you and happy mapping!

Hello K1RB1L1TY,

I noticed you recently reclassified several ways from track to unclassified. I came across a few areas where some additional consistency updates, such as alignments and removing duplicate ways would help improve overall data accuracy.

This way (way/1493688077) was created as unclassified, but it appears narrower than (way/1474710013) and in my opinion would be more appropriately classified as a track. This unclassified way (way/1493688075) duplicates a portion of (way/904327943), which could be extended northward instead, as the two are partially overlapping. Additionally this unclassified way (way/464431418) does not appear to correspond to a physical road northeast of (node/13226523860), and it also duplicates the mapped road near (node/2616348509). Lastly, It looks like (way/1493688080) was created but not connected to (way/1493688081), so it’s currently disconnected from the highway network.

Some of these discrepancies may be due to differences between aerial imagery sources. In this area, Mapbox Satellite imagery appears to be more up-to-date than Bing, so aligning edits with that source may help improve accuracy. I wanted to raise these data validation issues with you before any changes were made on my end. Let me know if you can review this area or if you would like others to make changes on the features you have recently edited.

Thank you for your time and happy mapping!

Hello Sam Jayah! I noticed that you’ve reclassified some ways around Bogambara such as (way/1226673466) and (way/1484710829).

The primary road from (way/1226673466) to (way/680347317) was formerly service and seems to act more like a backroad than this definition of primary in the Sri Lanka Tagging Guidelines (osm.wiki/Sri_Lanka_Tagging_Guidelines). “The next most important roads in a country's system. (Often link larger towns.)Intra-provincial arterial roads connecting major urban areas.” Since the road’s primary purpose seems to be a bypass or back access for the nearby buildings, what do you think about changing it back to service?

For the trunk from (way/1484710829) to (way/680347316), it seems like it becomes a pedestrian area considering the widened roads with cars parked alongside them and nearby temple (way/191377765). Also, the barrier=gate tags (node/7627975830) and (node/13676033621) are on both ends of this trunk which seems unlikely for a regular trunk. What do you think about reclassifying the extent to unclassified to reflect the usage here?

Thank you and I look forward to hearing from you.

Hello K1RB1L1TY! Thank you for your input. I’ll go ahead and make edits where it seems appropriate. Thank you and happy mapping!

Hello K1RB1L1TY! I noticed that you reclassified a lot of ways from track to unclassified in The Gambia. I see the logic in doing so when connecting between villages/towns and as residential roads for (way/370282579) and (way/370282572). This usage reflects “A road that is minor but serves as a connection in the general-purpose road network.” seen in (highway=track#How_to_decide).
Some of the unclassified ways though seem closer to the "A farm track used for access to crop fields, pastures, orchards or similar agricultural land.” and “A dirt track used for access to the mountains, desert, or other remote area.” defintions of track such as (way/951140763) and (way/951140762). These generally aren’t connecting between villages or acting as access to buildings alongside them. Since this might be an overabundance of unclassified classification, what do you think about reclassifying back to track or path to fit the intended purpose better? Thanks and happy mapping!

Hello Sam Jayah! I noticed you’ve made some adjustments to road classifications and refs around Anuradhapura (node/566574942). Some of these lead to some abrupt classification changes, and an overabundance of secondary and primary classifications which differ from local and global OSM standards.

For example, (way/671904381) to (way/44631344) as primary, transitions to a short secondary segment (way/107699958), and then to tertiary (way/672143227). According to the Sri Lanka Tagging Guidelines (osm.wiki/Sri_Lanka_Tagging_Guidelines), primary classifications are defined as “The next most important roads in a country's system. (Often link larger towns.)Intra-provincial arterial roads connecting major urban areas.”. Previously this was all tertiary classification.

The road segments (way/99072798, way/201253442, way/1480910258, way/1480910257) and (way/1482545178, way/1482636094, way/126563168) run along small residential areas and were once residential. What do you think of reclassifying them back to residential since that seems more aligned to their function?(way/443166146) to (way/44677699) were previously trunk classification with ref=AA012 and then changed to primary with ref=AB2-4 and ref=AB2-7. (osm.org/way 99072799) to (way/44631878) is currently trunk with ref=A12. Typically, trunk classification offers a more unimpeded route outside the city with primary acting as more of an arterial road.

Would you be able to point to the section of the Road Development Authority website (https://rda.gov.lk/index.php?option=com_content&view=featured&Itemid=101&lang=en) you used to adjust/add ref tagging? Thank you and happy mapping!

Hello MK 1! I noticed that you reclassified A1 from about (way/334747503) and (way/1291052713) to (way/334764342) and (way/1477325976) from motorway to trunk. This creates a gap where the trunks seem to appear abrupt between motorway classifications. In your changeset comment, you referenced limited speeds (50-70km/h), and the lack of 2+ lanes and 1 emergency lane as reasons for reclassification. Looking at 07/25/24 Mapillary (https://www.mapillary.com/app/?pKey=3902269899993845) just south of (node/3419149576), there’s road signage seemingly indicating this is a motorway. What do you think of reclassifying trunks to motorway for consistency? Thank you and happy mapping!

Hello NunoCaldeira! I apologize for any confusion. To clarify, the existing destination relations (relation/12725243) and (relation/12733136) are both for maneuvers traveling south towards Chão da Ribeira. I mention destination tagging on bidirectional (way/204537049) since if you’re driving east on it, you would be notified that you’re traveling towards Chão da Ribeira when you’re actually driving away from it. What do you think about removing destination tagging on bidirectional ways, and adding them as relations going forward? Thank you and happy mapping!

Hello NunoCaldeira! I noticed that you added “destination=Chão da Ribeira” on (way/204537049). Eastbound drivers on this way might be told they’re driving towards Chão da Ribeira, when they’re actually traveling away from it. I saw these relations (relation/12725243) and (relation/12733136) that are based on the direction of travel. What do you think about removing destination tagging on bidirectional ways, and adding them as relations going forward? Thank you and happy mapping!

Hello NunoCaldeira. Thank you for letting me know. Happy mapping!

Hello NunoCaldeira! I noticed that many of the ref tags in Azores and Madeira are being updated to remove the spacing. However, Mapillary from 07/13/24 (https://www.mapillary.com/app/?pKey=8423728917638113) shows road signage using ref formatting that includes spaces. I was wondering if there is a particular reason you could share for removing the spacing? Also, do you plan to apply this change consistently elsewhere, such as (way/26355360) where the associated relation (relation/2742099) has the spacing removed? Thank you and happy mapping!

Thank you!

Thank you!

Hello LovenOutdoors. There is a revert tool in JOSM (osm.wiki/Change_rollback) but you can also manually change the data back to its previous state. If you don’t feel comfortable doing so, I’m more than happy to do it for you. Happy mapping!

Hello MK 1, thank you for the response. Generally, it is best to tag features as they appear on signs. Are you aware of any street-level imagery (osm.wiki/Street-level_imagery_services) or government resources that verify these unique ref values? For example, Mapillary from July 2025 (https://www.mapillary.com/app/?pKey=2299428057138855) confirms the A1 ref value near (way/1352503199).

Hello LovenOutdoors, thanks for the reply! Unfortunately, Google Maps is not licensable for use in OpenStreetMap, and data from proprietary sources/maps should not be used to support edits (osm.wiki/Don%27t_copy_from_other_maps). Do you have any official government documentation or ground-level imagery from publicly citable OSM-compatible sources (osm.wiki/Street-level_imagery_services) that could confirm these ref updates? Thank you and happy mapping!

Hello, LovenOutdoors! I noticed you changed and added some ref tags in northern Cyprus. Some examples include ref=İK.05 to ref=D925 around (way/309612555), ref=D.55 to D 950 (way/1005400399 to way/965339335), ref=D.10 to D100 (way/806363590 to way/1244083914). I was wondering if you could provide some street level imagery or official documentation that reflect these changes. Thank you and happy mapping!