Minh Nguyen's Comments
Changeset | When | Comment |
---|---|---|
67664425 | almost 6 years ago | Fixed in changeset 73561163. |
67566162 | almost 6 years ago | Hi, I reverted this edit because it looked like you were practicing editing. It’s OK to practice editing, but please avoid saving in that case, because your test edits end up affecting the live map that everyone sees. If you’re interested in mapping schools, there are plenty of schools in the area that could use more detail. :-) |
67566162 | almost 6 years ago | Reverted in changeset 73560487.
|
67664405 | almost 6 years ago | Hi, this is nowhere near the old CooKoo’s Coffee Shoppe on Whetsel. If the garden does exist, can you place it in a more accurate spot and also move “Broccoli” to a more suitable tag, like “crop”? Thanks! |
71163894 | almost 6 years ago | By the way, the OSIP 6in background imagery layer is almost as up-to-date as DigitalGlobe/Maxar imagery but much higher-resolution and better aligned. |
73451357 | almost 6 years ago | Queen of Peace has a new church building; the one tagged here is actually the rectory and parish office. Fixed in changeset 73452007. |
72725668 | almost 6 years ago | Thanks for cleaning up GNIS nodes, but please be careful when combining them with buildings. Most GNIS entries since 2009 have very accurate locations, but older entries’ locations are only approximate. In this case, the engine house is next door at 1654 Marlowe. You can tell from OSIP imagery because it has a façade wide enough to accommodate two fire trucks. (Incidentally, a separate GNIS entry for the relocated fire station also indicates 1654 as the previous address.) In a previous case, I noticed that you had untagged a village hall as being possibly nonexistent, but actually it was across the street. If in doubt, please leave the POI alone and skip to the next Maproulette task. Thanks! |
71218483 | almost 6 years ago | These changes largely seem OK to me, but for the future, please note that highway tags in OSM aren’t intended to correspond one-for-one with HFCS. HFCS can be a starting point, but it would be better to look at its criteria and evaluate these roads on your own, perhaps taking those characteristics into consideration. Whenever you reclassify roads based on an external source, please double-check the roads in any available aerial or street-level imagery to make sure you aren’t propagating errors from that source. Also, please ensure good connectivity within each highway classification. When possible, avoid making a secondary road terminate at a tertiary road, as with State Line Road in Bright. Thank you for the care you’re putting into this aspect of the map. If you have any questions, please feel free to reach out to me directly or stop by the #indiana channel in OSMUS Slack. |
67095365 | almost 6 years ago | Hi, these ways lead into a parking garage. If you were trying to resolve a validator warning about inaccessible ways, the proper fix would have been to add an amenity=parking_entrance tag to the node that connects the ways to the building. Removing the oneway tags is incorrect, because traffic is only allowed in one direction on this entrance and exit. A lot of garage entrances in this region are missing amenity=parking_entrance tags, so I appreciate your help in adding those tags. |
70968470 | almost 6 years ago | Thanks for mapping out this preserve! By the way, the OSIP 6in imagery layer is much better aligned and clearer for tracing than Bing. |
68926660 | about 6 years ago | This changeset has been reverted in changeset 72842955. Please do not replace a circular way with a single node tagged highway=turning_loop. The highway=turning_loop tag exists for mappers who don’t have the time to draw a loop, but if someone has taken the time to draw a loop, that detail should be preserved. If you could communicate this nuance to your colleagues, that would be much appreciated. In practical terms, a large loop like this should never be represented by a single node, because the actual roadway is too far from the way. A turn-by-turn navigation application is likely to consider the user to be off-route due to their distance from the way. |
72470438 | about 6 years ago | Thanks for taking care of reverting those changes. I’m curious about your effort to form continuous primary routes between county seats. Were your changes motivated by cartographic concerns perhaps? Or does ODOT have a program to directly connect county seats with primary roads? I know that was the case back in the days of the intercounty road system, but a lot of those roads are now fragmented by newer developments and realignments, especially in Southwest Ohio. Plus it seems like ODOT is more focused on limited-access bypasses and such. |
68699921 | about 6 years ago | Thanks for these changes. By the way, when you’re mapping in Ohio, please consider using the OSIP 6in layer wherever it’s available. It’s more up-to-date than Mapbox imagery and at least as clear and well-aligned. |
72470438 | about 6 years ago | Changeset 72541941 reverts part of this changeset from Cuba to Leesburg. The bypass around New Vienna that you formed from township roads is nothing like what I’d expect of primary roads: they’re narrow, don’t have a double yellow line running down the middle, and at some points the pavement is so worn is looks almost like gravel. I retagged those roads as unclassified. In case it helps, we have some Ohio-specific examples of each road classification that you can use as a guide: |
72470438 | about 6 years ago | Hi TZLNCTV, this road should not be highway=primary in its entirety. Parts of SR 28 and SR 73 certainly could be, based on connectivity, but for instance the stretch of SR 350 going through Fort Ancient isn’t even built to the usual state highway standards, and I suspect there are other problematic areas too. Even though on paper it would seem to be a decent connection between the two cities, parts can be windy and slow, and there are weight restrictions all along. What do you think about reverting SR 350 back to secondary and promoting SR 73 from New Vienna to Wilmington? |
72320147 | about 6 years ago | Thank you for the clarification; I’ll keep that in mind in the future. Changeset 72410107 switches the main name to Phở Hòa, including the diacritical marks. |
69732445 | about 6 years ago | Undeleted the Anderson Township boundary in changeset 72098204.
|
67664298 | about 6 years ago | Deleted the fictitious church in changeset 71912688.
|
70866594 | about 6 years ago | I agree with you in general, but in this case the insurance agency, bail bond agency, and driving school are all one office – a single desk in a single office, at that. The fact that there are two websites for the same business isn’t really verifiable on the ground, but I added that tag as an FYI. In the insurance and bail bond industries, it’s very common for a single business to have two or three separate websites for better SEO. On a practical level, most renderers give it the same treatment as any office (due to the office tag). Some renderers, like Mapbox Streets, elide the bail_bond_agent value, turning it into office=insurance. Nominatim calls it an “Insurance;bail bond agent”. I was uncertain about this case at first, but the business’s driving school isn’t as important as its insurance and bonding services, so I think it’s OK to leave the driving school tag where it is. |
68705150 | about 6 years ago | Hi, thanks for helping to keep this area up to date. Unfortunately, this changeset broke several turn restrictions and turn lane data along several roads. Where turn lane tags are used, please start ramps and turn channels at the end of the turn lane. Sometimes this may result in less smooth geometry at the highest zoom levels, but this is the only way to preserve data necessary for proper routing. |