Gregory Peony's Comments
| Changeset | When | Comment |
|---|---|---|
| 172021440 | I understand. I can be difficult to map when other data is obscuring your view. Thank you for thinking about how you could restore the map's existing data. Copy and pasting was a interesting idea and can be really useful when you're mapping features that have the exact same shape, though there are some draw backs to this method and better ways of achieving your objective. You can press `w` to toggle wireframe mode. This will reduce the visibility of areas (the highlight/fill). You can filter out data that's either in your way or you don't want to edit via the map panel on right. Press `u` to open map data, then scroll down and toggle the check boxes to select what data you want to see and be able to interact with. Warnings about hidden data will still appear. Hope this helps. You can check out osm.wiki/ID |
|
| 172093417 | Hi, the footpritns you mapped here represent buildings. You correctly excluded what appear to be (white) tarps from some building footprints. For my feedback I will focus on https://pewu.github.io/osm-history/#/way/1430792920 . You outlined both the roof and walls of this building resulting in it being oversized: 192 square meters compared to 150 the way I mapped it. Here the North and Eastern walls are visible and shadows are cast NW. Please read osm.wiki/Roof_modelling and watch https://youtu.be/JAPiGntG6fs As I mentioned before in my comment on Changeset: 171090518 you should proceed with caution when using the Japan GSI seamless layer, because the angle it is captured at changes frequently, making it more difficult to know when you're looking at a building's walls. I recommend cross referencing other sources to aid your interpretation. Idealy one taken from a different angle and with shadows cast in a different direction. I'm assuming we have permission to use Japan GSI standard Map when contributing to OSM. It looks really useful especially for toponomy, but please compare it with the imagery; to me the footprints do not always appear accurate. For example I see no evidence in imagery to map the footprint I've been discussing as anything other than a rectangle. https://pewu.github.io/osm-history/#/way/1430792915 is positioned a few meters away from being in alignment with features in its vicinity. Try to improve it. If you choose to do so, then please send me a link to the changeset. I hope this is helpful. Thank you for your contributions! |
|
| 172086850 | Here is a good example of footprints that likely represent multiple buildings. It appears that the way the Eastern two were mapped previously was more accurate. Please keep this feedback in mind when contributing in future. Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/
|
|
| 172086495 | You did very well improving footpritns and deleting likely non-buildings. Some of the footprints you mapped are mapped as well as the imagery allows so I did not modify them during validation. --- On a few occasions you mapped some complex footprints which I think generally envelope multiple adjacent buildings. Try to see if a simple shape accurately represents the feature you want to map and if not then try more complex options. Consider if there is any distortion/illusion making it appear more complex than it really is. See osm.wiki/Roof_modelling#Typical_errors_in_the_interpretation_of_roof_geometry_from_aerial_images and roof:shape=*#Roof_shape --- Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/
|
|
| 172082128 | The pentagonal footprint here is likely square and it just looks like it's not in the imagery becasue it was taken at an angle. I cross referenced mapbox and it looks square there. Please see osm.wiki/Roof_modelling#Typical_errors_in_the_interpretation_of_roof_geometry_from_aerial_images --- It is possible to delete tags from features without deleting the geometry. If tags are consistently applied; it's simple to modify or remove them using JOSM.
|
|
| 172081801 | You improved the accuracy of some footprints here. Generally try to modify features instead of deleting and re mapping. In this case there's not much loss because there was not much history. Did you check it before making the deletions?
|
|
| 172081166 | Buildings with silver/grey roofs tend to be square.
|
|
| 172080592 | It looks like you mapped some non-buildings here. Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/
|
|
| 172039624 | Hi, the footprint you mapped here represents a building and seems fairly well sized, but it should have been squared and contains an excess node. Use backspace to undo the placement of nodes. After tracing and tagging features which are likely square or round, please remember to square their corners (q), or circularise them (o), because it is almost impossible and time consuming to draw shapes so percisely by hand. Buildings with metal or pitched roofs tend to have square corners; round buildings are identifiable by the distinctive cresent shaped shadow they cast. Unless the building is clearly a different shape then it's best to assume that it should be rounded or its corners should be squared. In the iD Editor, you can right click for access to editing functions. Since roofs tend to overhang walls trace the initial shape slightly smaller to allow a buffer for any change in size that may occur. In JOSM use the [buildingstools plugin](osm.wiki/JOSM/Plugins/BuildingsTools). In ID you must draw the shape accurately enough else shapes will not completely square. This [video about squaring features in ID](https://youtu.be/Xs5wX592E1o) has more information and a demo. Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/ |
|
| 172014119 | The oversizing of footprints seems to have resulted in you being unable to map one of the footpritns here with its actual shape. Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/
|
|
| 172013207 | All footpritns you mapped represent buildings but are oversized. Please do not include the shadow a building casts in its footprint. You appropriately mapped the western footprint as T shaped. When mapping buildings, please trace the shape accurately. Accurate building footpritns aid population estimates and prevent issues like data overlaps. Zoom in so that you can see the outline of the building and mark the corners carefully. Exclude shadows and yards when tracing the footprint. Keep in mind that you are looking at the [roofs of buildings](roof:shape=*#Roof_shape), but mapping their footprints. Depending on how the scene is lit, pitched roofs may have light and dark sections that belong to one building. Generally pitched roofs overhang the walls of a building, so a footprint slightly smaller than the roof is accurate. You can scale selected features in ID with shift+(-/+), or JOSM with ctrl.+alt+Lclick & drag. Take care to make contributions that others can build upon. Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/ |
|
| 172083770 | Good job modifying the existing data instead of deleting it; you improved basically all footprints here. The deletions you made were valid. The long T shaped footprint envelopes 2 buildings. Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/
|
|
| 172063981 | Hi Stella, It looks like you're identifying buildings in imagery but not mapping them as accurately as they could be. It seems like you might be accidentally tagging them with "name=q" instead of squaring them with the shortcut that shares the same key. Please watch This [video about squaring features in ID](https://youtu.be/Xs5wX592E1o) I recommend saving more frequently say every 5-15 minutes you spend mapping. Please keep this feedback in mind when contributing in future. Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/ |
|
| 172048371 | Both footprints here represent buildings and are quite accurate. Good job mapping the Northern one obscured by vegetation, though it appears to be slightly too long. Did you cross-refference other imagery sources when mapping it? You didn't mention it in the task comment, nor the sources of the changeset. It looks like 3 of its corners are visible in Mapbox.
|
|
| 172044137 | Hi, it looks like you did not account for the fact that building walls are visible in the imagery you're using to map footprints. When mapping with imagery captured at an angle, please position features where they are at ground level. You can map the outline of the roof and then reposition the building to the base of the walls. I recommend you watch this [video about mapping dense urban areas](https://youtu.be/JAPiGntG6fs) and read this [written guide](osm.wiki/Roof_modelling#Typical_errors_in_the_interpretation_of_roof_geometry_from_aerial_images). You created a node shared by a building and a highway; this should only occur in special situations. Hold alt to prevent this. Please do not connect the corners of buildings to other buildings or features such as highways or residential areas. In the iD Editor, hold down the `Alt` key to prevent your cursor from snapping to existing data and accidentally creating shared(grey) nodes. This [video about connected nodes](https://youtu.be/ltn1VOiq5_0) has more information and a guide. I hope this helps. |
|
| 171995229 | Hi, you mapped a few yards as buildings here. Note the direction in which shadows are cast and what is casting them. Using Bing as a supporting source in this area is very helpful; please read my project comments if you haven't already. You mapped a dupliacte footprint here, please modify features if you can tell what it represents rather than deleting and re-mapping. Do not apply layer tags to features to make overlap warnings go away, instead resolved the root cause of the problem. Please keep this feedback in mind when contributing in future. Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/ |
|
| 172042888 | Hi, the footprint you mapped here represents a building and you followed the correct procedure to map it, but it's slightly oversized. See how I mapped it in https://osmcha.org/changesets/172043726 When mapping buildings, please trace the shape accurately. Accurate building footpritns aid population estimates and prevent issues like data overlaps. Zoom in so that you can see the outline of the building and mark the corners carefully. Exclude shadows and yards when tracing the footprint. Keep in mind that you are looking at the [roofs of buildings](roof:shape=*#Roof_shape), but mapping their footprints. Depending on how the scene is lit, pitched roofs may have light and dark sections that belong to one building. Generally pitched roofs overhang the walls of a building, so a footprint slightly smaller than the roof is accurate. You can scale selected features in ID with shift+(-/+), or JOSM with ctrl.+alt+Lclick & drag. Take care to make contributions that others can build upon. Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/ |
|
| 172042855 | Hi, it looks like you mistook shadows cast by trees, and vegetation for buildings, though I must say that this is quite difficult to tell using only Bing (I used mapbox to aid my interpretation). In future note how the scene is lit and in which direction shadows are cast, consider consulting a secondary source if it helps. I deleted them in Changeset: 172043535 --- Please keep this feedback in mind when contributing in future. Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/
|
|
| 172042585 | Hi Enisa, All these footpritns represent buildings, but could be more accurate. In short I recommend mapping the footprints of buildings with pitched roofs slightly smaller than the roof. See how I mapped them in https://osmcha.org/changesets/172042952 --- When mapping buildings, please trace the shape accurately. Accurate building footpritns aid population estimates and prevent issues like data overlaps. Zoom in so that you can see the outline of the building and mark the corners carefully. Exclude shadows and yards when tracing the footprint. Keep in mind that you are looking at the [roofs of buildings](roof:shape=*#Roof_shape), but mapping their footprints. Depending on how the scene is lit, pitched roofs may have light and dark sections that belong to one building. Generally pitched roofs overhang the walls of a building, so a footprint slightly smaller than the roof is accurate. You can scale selected features in ID with shift+(-/+). Take care to make contributions others can build upon. --- Thank you for your contribution. If you want to experience the OSM community or to get timely feedback from other mappers; I recommend that you attend a mapathon. You can find events here https://osmcal.org/ |
|
| 172022406 | All footprints represent buildings; good job spotting them. Generally I recommend mapping footprints slightly smaller than the roof you see to account for overhang and any change in shape that occurs when squaring.
|