Viajero Perdido's Comments
Changeset | When | Comment |
---|---|---|
45232769 | over 8 years ago | Nicely done. |
48098763 | over 8 years ago | PS, I noticed these on one particular OSM-based map (Geocaching), where they rendered quite prominently, unusually so. And only this handful of neighbourhoods, which tells me that mapping them this way was uncommon, and possibly incorrect. Then again, it may have been the carving-out-of-the-county that's the unsual thing. I'm not going to touch that. |
48075242 | over 8 years ago | ...and de-tertiary'd CanVec roads, the usual. |
34522003 | over 8 years ago | PS, I haven't seen the different-names-different-seasons situation yet, but I'll take your word for it. I don't have a good answer for that. :) |
34522003 | over 8 years ago | Interesting; I hadn't thought of that angle. And clearly it must be rendering on *some* map (a piste map maybe), otherwise why would you be doing it? An idea: If satellite imagery shows the area to be forested, as in this example, wouldn't that suggest, at a minimum, a swath through the trees that would count as a path in summertime? My own opinion (for what it's worth): you could safely tag even a ski trail across a meadow as a path, given that (in this area at least) we have plenty of mountain routes mapped as paths but with no evidence of such on the ground. An obvious route up a ridge to a summit, for example. (For that OSM has trail_visibility, but it isn't used much it seems.) I'm just hoping to make your work available to more people... Cheers. |
34522003 | over 8 years ago | Hi. I see you've been adding ski trails, such as Crystal Link #3 in this changeset, but without any highway=xxxx tags. (In general, highway=path would be appropriate.) I think lack of this tag would prevent the trails from rendering on most, if not all, maps. This example, at least, doesn't render on any of the OSM maps that I've looked at. |
41628891 | over 8 years ago | I meant, the coords are hard to find. |
41628891 | over 8 years ago | Hi Dan. I would think, IF there's not already something mapped (eg already a trail through a particular valley or up to a particular pass), then importing a coarse one is probably okay. That is, if it's attributed to CanVec or whatever, so someone later can evaluate the validity, as you say. Our group walked from the hot springs to just shy of the pass, then returned where we came, and we had the valley almost entirely to ourselves on a long weekend. Nice valley. From where we turned around, later in JOSM I could see the trail up to the pass from imagery, so I added it all the way to the top. And it looks like you've completed it down the other side now, an improvement. BTW, I've added a few campsites elsewhere from JNP-provided coords (Brazeau area), but they're hard to find, and not all together on a single web page. So I only did Brazeau, and my email to JNP asking for coords to them ALL went unanswered. Cheers,
|
41628891 | over 8 years ago | Hi Dan. Two polite requests. 1. I had to look at your profile to identify the source of this changeset. It would appear to be CanVec. Could you please identify your source in future changesets? 2. I'd like to urge you to stop importing POIs and trails from CanVec, because the accuracy is poor in all my experience. We have armies of hikers fanning out across the planet now, and they're recording accurate trails and POIs with GPS units. Once an inaccurate trail is recorded in OSM, it's hard to know if it really exists where mapped without a site visit. Case in point: you re-added a campsite near Whitehorse Pass that I'd earlier deleted as being an inaccurate CanVec dropping. I'd already mapped the correct location after a site visit. Did you not notice it a few hundred metres away? A hiker relying on the CanVec location of the campsite may have an uncomfortable night... My throat always clenches up when I see someone importing CanVec in areas I've been working to improve. Please resist. I hope you understand. Thank you,
|
43273039 | over 8 years ago | That was fast - thank you! I'm pretty sure one of those fixes will take care of the glitch. Cheers,
|
43273039 | over 8 years ago | Hi fx99. I'm trying to debug a breakage in rendering of the North Saskatchewan River (by MapsForge in Locus with OpenAndroMaps), and think I have it narrowed down to one of two possible causes. 1) The river through Alberta and Saskatchewan is now part of an extremely large (>1000km!) relation, which may be too much for the library to handle. Or, 2) I've noticed one member, the south bank inside Edmonton, 450129499, is marked as inner where it was probably meant to be outer. I don't feel comfortable editing something of this magnitude. Would you consider breaking the multipolygon into several shorter ones? At a minimum, a break near the AB/SK border would make sense; Several providers (OAM, Locus) provide vector OSM maps in province-sized chunks, and having members extend far outside the province may be the problem. Thank you,
|
40800657 | almost 9 years ago | So with this changeset you deleted a hand-edited Blaeberry Road (by me, and which still exists beyond the washout), and also deleted hand-edited streams and glaciers (by Sundance, attributed to Mapbox, for example 379470433), to replace them with inaccurate CanVec versions?!? I've earlier flagged the missing road portions with OSM Notes. It's still there, and knowing where to aim would help a hiker; I personally took the wrong quad trail when I didn't know where to aim to find the road. Please, can you stop doing this? |
41854246 | almost 9 years ago | Forgot to add, source=survey for East Pit Lake trails, now overgrown. |
41660269 | almost 9 years ago | Fixed the multipolygon error - which no renderer that I'm aware of seemed to have a problem with - in a way that is technically correct, but may annoy a landowner and one-time mapper. |
41838695 | almost 9 years ago | "re-annoy", "sanctuary". I had better things to do than deal with this mess-up, which had no real-world purpose other than satisfy some arcane mapping rules. Mapping rules: satisfied.
|
41660254 | almost 9 years ago | Reverted. Clearly in error. This area can be well-managed locally. |
41660269 | almost 9 years ago | Never mind, reverted. It was clearly wrong, no relations were involved. I didn't want to see the messed-up version in the next offline download, hence I didn't feel like waiting. This area can be well-managed locally. |
41660269 | almost 9 years ago | You moved the boundaries. Why? I researched their locations carefully, and I actively maintain this sanctuary. What multipolygon was broken, and how? I don't recall any errors when I last touched this area. Please repair this, and save me the trouble of figuring out the reverter tool. Thank you,
|
37712677 | about 9 years ago | I'd hesitate to revert this. I was thinking more about "urging caution" for future work... |
37712677 | about 9 years ago | Hi alester. Many of these CanVec waterbodies don't seem to exist any more; see satellite view. A random example:
(The geocaching map allows an easy back/forth comparison.) Farther west, I'm personally hand-mapping water bodies (from satellite) in an attempt to discourage CanVec imports. I hope it works. |