OpenStreetMap

SomeoneElse has commented on the following diary entries

Post When Comment
BOUNTY: identify changesets where street names were inserted 1 day ago

@jinalfoflia - OsmCha, while useful, doesn't actual do what is requested here, does it - "identify changesets where street names were inserted"?

If it does, perhaps you could explain how?

Potential contribution from ICT perspective 7 days ago

Re "we've done the quick fix for the app" can you confirm what changes you have made?

There are a number of issues. One is that the only changeset tag is typically "created_by" and is set to the mapper name, which is useless.

Another is that often the geometry of objects is implausible. Apparently http://www.openstreetmap.org/way/433221060 is a >3 storey building with a shed on the roof, with a footprint that is less than a meter wide. This seems unlikely.

Still another is that all edits from the last day or so seem to have added the building type (typically "house") as either the name or the addr:housename or both.

Another is that some other tags seem to be unique within OSM. Where did you discuss the tagging scheme that you are using, and with whom? With imports there would be a step discussing the mapping of GIS attributes to OSM tags, and it seems only sensible to do the same here, especially where you are creating tags unique in OSM.

Potential contribution from ICT perspective 8 days ago

What'd be really helpful right now would be if you and the "great team who have got your back" could try and resolve some of the duplicates that are right now being added to the OSM database because of your buggy "geoMapTool" software.

It's easy to spot. All of the data has mistaggings. Much of it is duplicated - the same building uploaded in the same are over and over again.

It's not the fault of the users adding this data - they are presumably only doing what they were told to do, and are "trying to do the right thing" by adding valuable information in rural areas in Bangladesh. Unfortunately, the software that they have been given to use is letting them down.

Best Regards,

Andy

New edits not showing (yet again) 22 days ago

Yes - just to reiterate what has been said above - your edits are fine, and anyone using OSM data will see them. All the mobile apps that use OSM data will include them at the next upgrade cycle. The various third parties using data from e.g. Mapbox will see them; even the public transport infrastructure and the company printing maps that are displayed near supermarkets and railway stations local to me will see them too. They'll appear on the "standard" map at osm.org too, just a little slower for now.

New edits not showing (yet again) 23 days ago

It's a known problem - there's just been a discusson on IRC about it.

Dressed-up Notification Mails about 1 month ago

I have one question - how do I turn it off?

The mails themselves do include a "text/plain" part, which is good, and it is often possible in view that by default in individual email clients, but not always (and it's cumbersome to have to do it many times).

Clearly some people actually like to read HTML formatted text (though I can't for the life of me think why). However the fact that this is still being adjusted to make it usable on mobile (https://github.com/openstreetmap/openstreetmap-website/pull/1436) overlooks the fact that there's an excellent solution to "make a message readable everywhere" already - just make it available in plain text.

Was a profile option for "send always-legible notification emails" considered? How hard would it be to add?

Are you joking ?? about 1 month ago

... which for the curious is here in OSM.

Are you joking ?? about 1 month ago

It's perhaps worth adding that what you're seeing is "landuse=forest" in the sense of timber production - imported ways such as http://www.openstreetmap.org/way/64298415 . It doesn't correspond to "trees on the ground here and only here" (as the underlying imagery shows, there's a light covering of trees all over). There's a bit of background about the ways that people map forest/woodland/trees in OSM here - unfortunately it's quite complicated. See also here for part of the discussion in OSM's "standard" style about how to map forest/woodland/trees.

Pokemon Go trash about 2 months ago

Taking the points raised here from the top...

As it was recently discovered, Pokemon Go users have found that OSM data affects Pokemon "nests" location.

I'm not convinced that that has been discovered. It's certainly true that some fake stuff is being added (as you mention, some helpfully marked "not actual park", but also some water features such as https://www.openstreetmap.org/way/469266458 ), but I don't think that anyone's found the "smoking gun" yet showing that Pokemon game locations actually derive from OSM. We know that the game map claims to use OSM data in South Korea, but that's a different issue.

So the next question is, what to do about it? There are two bits to it - the new users, and the data.

In the case of the data, if (as here) it's obviously fake, then it makes sense to revert it so that those cars in Troutville are no longer parked in a pond. However quite a lot of the Pokemon Go new edits that I've seen are actually trying to map real-world features. They may not be making a very good job of it, but they're not in any sense "vandalism".

The other issue is the new users. In this case, the fake Troutville pond mapper has made exactly 4 edits. They likely have absolutely no idea what OSM is; they only know it as "source for in-game items" (and whether that's true or not is an entirely different issue). Personally I'd like to think that there's a non-zero chance that we can explain to them what OSM is and have them become a "real" OSM mapper. Ethan's response seems entirely reasonable to me in terms of trying to guide them on that path - it may not work; but if it doesn't we've lost nothing, and it doesn't stop us fixing the data,

Finally, with regard to how to fix the data I suspect that a JOSM revert-based approach is likely to work better than an overpass find-and-fix one, as I suspect that that will miss stuff that other people have edited in the meantime - JOSM revert also flag up where there have been subsequent edits (which might be valid) that need to be considered.

"Crisis of anarchy" about 2 months ago

I'm sorry, but if you think that the choice is between "a cool project or the best map" then you're talking through part of your anatomy not best adapted for the purpose :)

Let's take an example. According to Wikipedia that is a "is a place noted on a map". That curious form of words was used after various people tried to get the article deleted because the place does not exist. Perhaps it did, 150 years ago or so. Unfortately, the wikipedia admins, who thought they knew best, decide that because some left-over historic text appears on some maps of the area it deserves an article, and for that reason it lives on as a 21st-century "trap street" advertising the folly of wikipedia.

In fact it gets worse - it even has a wikidata article, and that has been shoe-horned into a category of "hamlet" rather than "place that used to exist but doesn't any more" because whatever created it had no local knowledge. I say "whatever" rather than "whoever" there with reason, because the history suggests it's bot-created and almost-entirely bot-modified.

In short, wikipedia is an excellent example to OSM - of exactly what not to do.

"Crisis of anarchy" about 2 months ago

@salmin Yes, because obviously we just need to follow what wikipedia does :)

More seriously, wikipedia and OSM are very different. We've seen people coming from wikipedia recently and seeming shocked that they were asked about what they had been editing - in OSM it's not good enough to just copy from somewhere else. In many ways OSM is the anti-wikipedia - original knowledge is preferred.

Whilst communities online and offline can always learn from one another, I'd take advice from wikipedia with a very large pinch of salt.

Egleton Triangulation Pillar 2 months ago

The wiki pages often suggest other tags that I don't know about (I had no idea that "type=triangulation" was used, for example). Looking at the area in an OSM editor, I can see that the "NLS - OS 1:25k 1st Series 1937-61" map that's available as background does show it as being just above a 475' (145m) contour, so somewhere in that region would do I guess. Depending on the editor you're using you may need to zoom out to zoom level 16 to see this background.

Re mapping parties, the talk-gb mailing list is the place that most things will be mentioned. There may also be discussion in the talk-gb channel on IRC. There are regular active meetups (usually in pubs) in at least London, Birmingham, Notingham/Derby and Edinburgh. I think someone has recently been trying to organise something in Oxford too. Mapping meetups are rarer - I'd guess April/May for the next Midlands one.

Egleton Triangulation Pillar 2 months ago

Re the elevation, I probably wouldn't use the elevation from an eTrex 10 as I don't think that's got an altimeter in it and is just reporting the GPS altitude. It might be close to the real value, or it might not. Personally I'd also be careful with an altimeter-equipped value, based on changing weather conditions and how recently calibrated.

However, any trig point that's been there for 50 years or so is probably going to be on an out of copyright map somewhere - presumably you could use the elevation from that?

Re "Bing's idea of the absolute position of points on the earth's surface" - that's an excellent question. Imagery such as Bing's can be offset (though it's less of a problem than it once was) so it makes sense to check available compatible sources such as GPS traces as well. In the absense of any other information where you've put it is as good a place as any. The general way things work in OSM is that things get added and then refined later - don't get worried about "I might not have put it in exactly the right place" - someone can always move it later.

For info - if you've got any general questions you can always ask over at the help site: https://help.openstreetmap.org/questions/ .

Finally, welcome! You actually missed a mapping party at Scalford just up the road from you at the weekend :)

2016 Washington New Users Report 3 months ago

That's great - it's brilliant that you're seeing a difference in the "time to stick around mapping" over a significant number of users.

3 years of welcome messages, more than 3400 of them 3 months ago

A while ago I did have a look at the retention rate in the Polish community, where 75% were welcomed and 25% not (see https://lists.openstreetmap.org/pipermail/welcomewg/2013-March/000022.html and the links back from there). There wasn't an appreciable difference when I looked between 2 and 3 months after the message was sent (or not).

What I didn't look at that would have been interesting was contribution quality, and I've not gone back to look at longer term retention (anyone can do this - just look at new Polish mappers from that period whose UID was divisble by 4 or not).

Old-style multipolygon cleanup in Luxembourg 4 months ago

When are you going to do the pull request to P2 then (which IIRC still creates them)?

:)

Clean up the "fixme's" around you! 4 months ago

... and for the ones that need a physical survey, https://github.com/SomeoneElseOSM/Notes01 (shameless plug) will let you create a GPX file for a Garmin handheld for local surveying or both notes and fixmes.

mileage convertion when reading 2 points 4 months ago

Assuming you're talking about directions, use the website that's providing the directions (whichever one you're using), not the simplified interface from osm.org.

... or divide by 8 and multiply by 5 :)

Time to cleanup the wikipedia:xx tags? 4 months ago

For example, on the ground, you can't distinguish between administrative entity and settlement.

I think you can. To take an example that I used in this "talk" list post earlier today:

https://lists.openstreetmap.org/pipermail/talk/2016-November/077139.html

"Rossnowlagh" is a settlement. It's got a hotel and a cafe, and lots of other settlementy stuff. "Rossnowlagh Upper" and "Rossnowlagh Lower" don't have any of these - in fact when I was there there was nothing to indicate they exist at all on the ground.

Time to cleanup the wikipedia:xx tags? 4 months ago

Alas, the real world has inconsistency in it, and we have to deal with that. Wikipedia isn't "a database" it is "lots of different databases, with some coherence between them" OSM's wiki is notoriously a haven for people who want to tell the real world how to behave rather than describe it, so sometimes you have to take what's written there with a pinch of salt.

Sure, there will be examples where people have added multiple wikipedia keys by accident, but some may be deliberate, and the way to find out is to ask.

Blindly converting wikipedia entries to wikidata entries would have the same problem as adding wikidata without putting much thought into it (see https://www.openstreetmap.org/changeset/43749373 for a discussion of some of the issues that can occur there - a wikipedian with the best of intentions but no local knowedge missed things that would be obvious to a local).

Unfortunately, there have been so many "mechanical turk" additions of wikidata links to OSM recently by people unfamiliar with the things that they are linking that it's effectively devalued the ones that existed previously. That's sad (and is obviously a completely different issue to what you're talking about here), but it does mean that "wikidata is not the (only) anwer".