Talk:Yahoo! Aerial Imagery

From OpenStreetMap Wiki
Jump to: navigation, search

'Legalities' section

I added the legalities section and stated what must have been agreed (presumably) can someone confirm that, and elaborate at all? I guess yahoo must be restricting us, such that we have to create vector based maps as a derivative work of aerial images. Any other kind of derived work, and we would not be allowed to slap the open content license on it. ...but that's just a guess. -- Harry Wood 03:02, 26 January 2007 (UTC)

I think we need to actually know what the agreement is, as I'm not happy with the "presumably"s when we are dealing with something as important as licensing. Secretlondon 23:02, 10 March 2007 (UTC)

I just did a google for the terms of use and found this: terms of use. With all the different sources that they use just for the imagery I would think it would take thousands of not hundreds of thousands of dollars worth of legal services to hash out an agreement with OpenStreetMap. I'm skeptical that such an agreement exists. The use of Yahoo imagery and any data created with it should be thrown away. --Korea 11:42, 21 June 2007 (BST)

Yeah I don't think there is a written agreement (yet). That might be because it would be very complicated and require lots of lawyers as you say.
Poeple got a bit up-in-arms about the Yahoo! thing a while back (the JOSM plugin issue). Mikel briefly posted to the mailing list telling everyone to chill out. I pasted his comment into the legalities section there. So it seems there was a friendly verbal agreement made with Yahoo, which I guess Mikel knows more about.
I wrote the legalities section here, including the word "presumably". I agree with User:Secretlondon. It would be nice if somebody who knows more (Mikel?), would validate what I have written, and either change it, or just remove the word "presumably". But the people concerned have not paid much attention to this wiki page, or maybe they're staying quiet on the issue for a reason. I dont know.
Not ideal is it? But I don't think we need to be to defeatist. As Mikel said, the spirit of what we are doing is sound. -- Harry Wood 12:40, 21 June 2007 (BST)
There are at least four channels of communication within OSM (mailing lists, IRC, forum, wiki pages) - don't assume that anyone watches any given one of them. :) You could mail Mikel directly if you're particularly anxious. --Richard 13:02, 21 June 2007 (BST)
I'm not being a defeatist. I say go for it, but until you have the signed document in your filing cabinet you shouldn't touch the data. As the saying goes, I'm not a lawyer but... --Korea 13:21, 21 June 2007 (BST)

OK User:Mikel recently supplied some more information in mailing posts (which I've factored into the page). He also told me he would take a look at editing what's written here at some point, but said I could drop the word 'presumably'... so that's good. -- Harry Wood 16:05, 13 July 2007 (BST)

No agreement is necessary

No agreement is necessary. I'm certain that taking measurements from a photograph and then drawing a graphic based on those measurements is not a violation of copyright.--Korea 06:57, 20 June 2007 (BST)

Certain? How certain? Are you a lawyer? Is there legal precedent to back-up this statement?
"Taking a measurement" sounds reasonable enough. But we are directly tracing over the imagery. Not only that, but we are extracting the imagery from the yahoo mapping service and presenting it in our own tools to facillitate this tracing.
We are making the assumption that this kind of derivation definately requires an agreement. We kind of have to make this assumption otherwise we might as well jump straight in and use google aerial imagery without asking them. This would be great. Google has full UK coverage at higher resolution. It would enable us to map the whole of the UK without bothering with any GPS. Note that this would immediately rubbish the efforts of all OSM mappers who have been out and about with thier GPS units. But imagine if we spent all those man-hours tracing a map using google aerial imagery, only to be told we cannot release the maps with a CC license because we owe google royalties.
There's various legal grey areas. With this project we are always erring on the side of caution to ensure our data is free. I think it's important that all OSM mapping people understand this actually. It may seem like legal nit-picking, but in fact the whole OSM project is established on the premise that the only way to create a copyright free map, is to build it again from scratch. If we start deriving from sources we are not sure about, then we might as well not bother at all!
-- Harry Wood 10:21, 20 June 2007 (BST)
Ok, here be lawyers... it gets a bit scary.
First of all you need to remember that you cannot make absolute statements in this area without qualifying them as to jurisdiction. US copyright law is certainly different from UK which is probably different from other parts of the EU. How does this affect OSM? Well, there's your first problem.
Secondly, aerial imagery is not just "photographs". If you take a load of photos from a plane, they do not line up. They need to be "rectified" (straightened up and stitched together), which is a complex and painstaking process. To do this, you need some prior existing reference material. Sometimes this is GPS... but sometimes this is a copyrighted map. So you can see you're already into the whole "what is a creative work? what is a derived work?" issue.
Thirdly, there is a difference between copyright (what the law allows you to do) and contracts (what the person or website which 'gave' you the imagery allows you to do). Even if you could prove that there was no copyright infringement, the Google (or whoever) terms of use might well say "By accessing this site, you agree not to trace over it". So you're still stuffed.
Harry is right about caution being the watchword. Personally, I think that the upcoming months will show that a strong community outweighs any benefit of the fanciest aerial imagery in the world. But more of that anon. --Richard 13:03, 20 June 2007 (BST)


I did some looking on the Internet and found quite a few comments concerning copyright that I knew were not correct. Even something I found written by a lawyer on a legal website had at least one error I could see. I wonder how many mistakes it had that I did not see. I remember a lawyer telling me that in most US states intellectual property law requires a separate bar exam because it is so different. Even if we follow the law and have a good argument then we still have to pay lawyers and present the argument in court, and this could happen even if we have an agreement with the source of the photos. It may be best to avoid the whole deal regardless of whether we think we have a good legal position. --Korea 11:06, 21 June 2007 (BST)

Added 'Using Yahoo! Aerial Imagery' section

I wrote some blurb in a section called 'Using Yahoo! Aerial Imagery'. Seems like some info we were missing. I wrote some statements as if they are hard-facts, when actually I'm not really sure what I'm talking about... as is the wiki way :-) Please feel free to modify this!

Do people agree that the availability of aerial images significantly changes the way we build maps? I was reading the proposed approach for the forthcoming London mapping party and came to these conclusions. Do we even need GPS tracks any more?? -- Harry Wood 03:02, 26 January 2007 (UTC)

It changes it only for those small parts of the world with hi-res imagery. --Richard 07:49, 26 January 2007 (UTC)
We need all the data. The more data the better. --Korea 06:59, 20 June 2007 (BST)
Well I'm not suggesting we throw away any data, but I am suggesting that we no longer need to bother with gathering GPS traces in places like London. In fact I've since proved this by doing a lot of mapping around London without using a GPS, by sketching over the yahoo aerial imagery in a two phase approach as described on the page here.
The planning information for the recent Richmond-on-Thames 'Fill the gap mapping party' seems to demonstrate that people haven't really cottoned on to this yet (still focusing on using GPS in London).
Correct me if I'm wrong, but isn't the aerial imagery easier and more accurate? OK I concede that using a GPS might provide a minor additional aid to mapping (see where you were when) and of course it is a fun gadget. But really Londoners... aerial imagery is faster smarter and cheaper :-) -- Harry Wood 10:39, 20 June 2007 (BST)
We need to traverse the entire length of each road to find features to mark on them. GPS trails might or might not be used for the actual paths (the imagery is nice, but still not good enough to replace trails entirely in some areas), but I'd like to see you try and position a postbox on a random junctionless stretch of road accurately without having a waypoint where you saw it. Morwen 12:03, 20 June 2007 (BST)
That's right. If you have a junctionless stretch of road, and you want to position things like mailboxes, which can't been seen from the imagery, that's a situation where GPS is handy. Now I haven't hit that problem once with my mapping of Central London because there aren't any junctionless stretches of road! I can pinpoint the location of a mailbox to within a few metres (more accurate than a GPS no?) just by recording the right place in relation to the features of the road layout.
I did experiment briefly with trying to sketch in some country roads outside the M25. Definately gets quite tricky out there. I couldn't tell for sure if it was a road or just a hedgeway at the edge of field. But that was phase 1 sketching anyway. Phase 2 surveying and positioning of postboxes on an junctionless stretch of road, would indeed be difficult. You could note down the position in relation to a hedge or something visible from the air... or just leave that for a GPSer to fill in. -- Harry Wood 16:11, 11 July 2007 (BST)

Copying Street Names

"We will be faced with a terrible temptation to look at existing maps, just to get the names of streets, but we can't do that! It's copying!"

No, it's research. Expressing the well-known fact that St. Paul St. is called that, by writing it on a map, does not give the mapmaker any exclusive right to the underlying fact. Neither does my having learned it from the map rather than from a street sign forbid my passing the factual information on. What is subject to copyright (in the US, in a nonlawyer's understanding) is the creative expression of these facts in the form of a drawing, which we would not be copying in any respect. (It might nevertheless violate the terms of use of online data providers, but printed works don't have terms of use.).--Tompermutt 15:11, 11 July 2007 (BST)

Well yes. Legally we are unlikely to get into trouble for copying one or two street names from a map. But as with many of these things, its a slippery slope which we can't afford to start down. If we start systematically copying street names from maps, then we have a problem. If while copying lots of street names, we copy spelling mistakes or names of roads which actually don't exist. Is that enough to give someone a case for breach-of-copyright? You are claiming not, but I wouldn't like to find out.
In any case, if we are staring at copyrighted maps for hours on end, there's a good chance we'll be unwittingly copying some of the street layouts too. Best to keep the whole thing clean and uncontaminated by copyrighted information. It seems silly, but the safest approach is to empty your mind of any google/ordnance survey maps while you're editing OSM. Resist the temptation to look at them! -- Harry Wood 16:11, 11 July 2007 (BST)
Partly true... Facts are explicitly excluded from copyrights, but the presentation of facts isn't. Alexanderpas 04:31, 28 August 2007 (BST)
Have a look at the Case law. Most of it is fairly straight forward. Copyright is likely to subsist in a compilation of street names in many jurisdictions, just as it does in telephone directories and railway timetables. Since OSM ultimately has the capacity to significantly reduce the revenue for mapping companies, the possibility of this going to court and large damages being assessed has to be a real consideration. --inas 10:50, 28 August 2007 (BST)
And, there's not just copyright to deal with, but database right. 11:04, 28 August 2007 (BST)
In EU jurisdictions, yes. --inas 00:35, 29 August 2007 (BST)

Do not the local municipalities have maps with their street names? I've taken my local municipality map (although offline, they don't have a webpage) and it shows every streets name and it's more accurate than google maps) Gnetter 14:12, 19 March 2008 (UTC)

In the UK the local councils "have" maps certainly. They also provide maps to the public for free in some situations. e.g. tourist maps on big boards in a town centre. ...however they are always based on Ordnance Survey maps. Often they explicitly say "Crown Copyright", so even though they are allowed to present the map to the public, they are doing so under license.
In other countries, different mapping agencies are involved, but generally speaking it's safe to say that somebody somewhere down the line went to quite a lot of effort to survey and produce a map. Are they making this intellectual property available for free? (free to copy)
We certainly can't assume so. Does it do any harm to copy a few street names? Maybe not, but it's a slippery slope leading to large scale copying, which is not the premise on which OpenStreetMap is built.
If you can determine that a map is released to the public domain (by the poeple who are allowed to make such a declaration), then that's great! Copy everything, not just the street names! Otherwise... stay away. Dont even look at the map. Build it from scratch. That's challenge we are setting ourselves. --- Harry Wood 17:08, 19 March 2008 (UTC)
Non sequitur. Going to a lot of effort to find out facts does not make the facts intellectual property. Somebody else went to even more effort to post and maintain the names of the streets on signs. Why do you think the names on the signs are free to copy, but the names on the maps aren't?--Tompermutt 20:32, 21 March 2008 (UTC)
The fact of the matter is street signs are not copyrightable, but people are allowed to copyright maps, and my point is that generally people do copyright maps (because it was hard work creating them) and so any maps held by local councils or other government organisations will generally not be free to copy. The same applies to anything "derived" from maps, with the tricky point being what exactly counts as a derivative work.
It's something we need to be wary of because government people might show willingness or even enthusiasm for putting data into OSM, but unless they are in a position to relinquish copyrights on the map (which would be most unusual) then this doesn't help us.
Interestingly in the UK recently some local council GIS people have cottoned on to OpenStreetMap and started contributing, seeing it as a way of helping themselves to break free of Ordnance Survey's strangle hold. But they can't just take all their wonderful data and give it to the project, because it's all derived from OS maps.
-- Harry Wood 14:38, 27 November 2008 (UTC)

Patches of Terra Incognita

For now, the map is a thematic map of Places We Geeks Like to Go, which has its own charm. As it grows to be truly a map of, say, Baltimore, I am concerned about the implications of patches of Terra Incognita, where nobody with a GPS and a notebook will ever wander. (I'd bet the makers of existing maps haven't been there, either.) Sure, mailbox-level data is nicer, but I'd rather we filled these patches in from aerial photos and other sources than left them blank .).--Tompermutt 15:11, 11 July 2007 (BST)

Yes I agree, and I think this an argument in favour of a two-phase mapping approach. We can slap in lots of roads quickly based on the Yahoo aerial imagery (or based on GPS traces in fact) This would be a boost for our completeness. Trouble is the streets wouldn't have names, and in general the contributions would be perceived as detrimental to our overall accuracy unless we can think of a way of rendering 'sketched' roads, so that they do appear on our maps, but look different in some way, such that it is clear there is phase 2 surveying work to be done. On this Yahoo! Aerial Imagery page there's already some description of this. I see this as an unresolved rendering issue. Maybe I'll discuss this more as a separate post.
-- Harry Wood 16:52, 8 October 2007 (BST)

Sketching - Is just tracing useful?

In September there was a whole mailing list discussion "Is *just* tracing useful?". I would have emailed my thoughts on this, but I missed the discussion at the time, and I don't tend to post to the mailing list. But I think is a very important question for OSM. The question should be answered here on this wiki aticle. Once we've debated what the answer is. It's not obvious, and if you read all the posts (not just the first one!) you will notice some polarised opinions. Perhaps we could establish some things we all agree on, using the wiki. We could list advantages/disadvantages and proposals.

"tracing" is something I've referred to in many places as "sketching". I also like to use the word "surveying" to mean the process of actually going there, visiting the streets and recording information (with or without GPS).

I think its crucial that people understand and agree that...

  • Surveying is very necessary in order achieve a complete and accurate map.
  • Surveying is very time consuming exertion of effort. It is the "hard work" of creating a map.

For a more completed accurate map, we must go to the streets and gather street names and other important details. We will never have a map which can be respected or relied upon, based purely on aerial imagery. As well as names, it will be missing other crucial details e.g one-way restrictions, or bollards/barriers at the ends of roads (common in London residential areas). It will also be wrong in places the aerial imagery is difficult to interpret (Is that a leafy street, a footway, or just a row of trees behind houses?)

Some people seem to feel quite strongly that sketching is bad, and that Yahoo Aerial Imagery should only be used as a mapping aid for drawing in roads where you have already done the surveying work. I want to argue that it is useful to sketch in road layouts, and leave these sketches within the map data (without necessarily intending to survey those same streets yourself) but before I do that, let me stress that I firmly believe and understand that the act of surveying is hard work. The people who do this (myself included) have put in a lot of effort in these areas of the map. Anyone who does sketching is, by comparison, making a relatively minor contribution to a given area of the map. Sketchers need to treat surveyors with respect...

...or rather (since these can be same person) sketched road data needs to sit quietly and respectfully alongside the superior surveyed road data. And this is the key I think. We need to devise a data representation and a rendering approach for sketched roads, which will give a better understanding of the relative effort that has gone into the data. My proposal? Simply make untagged ways display rather like the old unwayed segments did: faint grey sketchy looking lines in tiles@home osmarender (and why not also Mapnik?)

What's wrong with the current state of affairs? The renderers show nothing for untagged ways (sketched roads). This has given some sketching people the idea that maybe they should have a guess at a highway tag. These then get rendered, and the sketched roads are only distinguishable from superior surveyed road data by the fact that they dont have a name. In fact they start to look very complete (when they are not!) Very irritating for anyone who has put the effort into surveying. We're giving the wrong message, that sketched roads are just as good as surveyed roads. It's also bad for anyone who is trying to see which areas still need surveying, or find those one or two streets which still need a name. (Actually a nice trick in JOSM is to do a search for "name" to show up the surveyed roads in white)

But if we sketch in untagged ways, we can't easily see (on the rendered maps) areas where we have sketched in road layouts. For the people who hate sketched roads that's just fine, but for the rest of us, this is sub-optimal. You can see something in the Maplint layer, and some would argue that "That's what maplint is for. Showing incomplete bits of the map", but it hardly promotes awareness of the sketching approach, or awareness of the presence of sketched roads in a area you might be about to survey. Who goes to look at maplint for blank areas of map? In my opinion we will benefit greatly from having the information right there on the map, but it should only appear as respectful faint grey lines.

To be clear, I am suggesting a simple change: The renderers should show untagged ways with faint grey lines.

With that in place, we could encourage sketching, rather than having half the community whinging about it. I would argue that sketched roads will actually create a good impression to the end-user (potential newbie mapper). People might look for their street and think "Blank white map. Where is my street? This map is hopelessly incomplete". It would be better if they look at their street and think "That's my street, but it's just a sketchy line. That's a bit rubbish, but I could fill in the street name. Let's try editing!". Basically it's a boost for our map completeness.

This brings me onto my other more radical pro-sketching idea. I think we should aim to sketch everywhere, over all the yahoo imagery! We wont even need to do it ourselves. Richard F drew parallels between sketching and wiki gardening here. He wasn't being complimentary, but why not? The internet can draw crowds and crowds of "casual contributors" who will indeed develop some kind of weird addiction, and will sketch over the whole of greater london and every other yahoo covered city from Beirut to Bangkok! It will happen quickly if we only let people know that this is what we want to do, and tweak the tools to make it work. ...it won't be a finished map, it won't look like a finished map, but it will look like something in all those huge areas we haven't got to yet.

-- Harry Wood 16:52, 8 October 2007 (BST)

Since writing the above, I notice around London there's been an explosion of sketching using unnamed highway tagged ways (the approach I was not advocating)
Take this view of South London zoomed out. Looks pretty complete hey? Impressive stuff.
Now zoom in a notch North of New Cross for example. While zoomed in, you can see that the vast majority of these highway ways do not have names. I'm assuming this means they're sketched in from the aerial imagery.
This whole area represents a probably a few hours work (by TimSC?), rather than many weeks of surveying work (proper mapping), and the problem is that when you look at it while zoomed out, you can't tell the difference. No wonder people get annoyed about the sketching over aerial imagery.
But the problem is not the sketching, but the representation and rendering.
-- Harry Wood 12:15, 30 November 2007 (UTC)
I hope to have an unequivocally out-of-copyright London streetmap for OSM to use in the next few weeks, so you'll be able to fill in some of the names. --Richard 10:48, 3 December 2007 (UTC)
Regarding tagging of ways vs. tracing ways, we are balancing convenience of mappers against possible benefit to users. I think I will comment again on this after the South East London Mapping Party which is in a traced area with tagged highways. I did post my initial thoughts on Harry Wood's talk page. I guess we will see how the party goes!
On the topic of standardizing how mapping should be done, there is a surprising lack of current documentation. The standards page is marked as out of date and contains many unanswered questions. Obviously there is a page and a process for updating the map features but even this is optional. "You can use any tags you like." I like this free form approach to mapping but it can be a bit chaotic! I don't want to unnecessarily inconvenience other mappers. :) --TimSC 17:45, 23 January 2008 (UTC)


I did the South East London Mapping Party and attended the Surrey pub meet and I was thinking about the yahoo tagging issue. My personal experience is that tagged ways are useful because I can plan my mapping as I go on my GPS. In discussion some of the points, for and against tagging, were reiterated. I am content to leave it with no standardization since it is not necessarily to do things one single way.
If there is a problem with the mapping work flow and tools, it should be addressed by modifying the tools or modifying the work flow. IMHO This has already been addressed by maplint, JOSM validator, modified Kosmos, etc. --TimSC 17:21, 30 January 2008 (UTC)
Well it's interesting that you mention maplint, because many would use maplint as an argument against your approach. Maplint allows you to see which areas if the map have untagged ways. Some people argue that this is sufficient to allow sketchers to see what has been sketched. They get annoyed when people put sketched roads onto the rendered maps, by setting a 'highway' tag, because they should be using maplint to see their progress. That's the opinion some people have.
My opinion is that sketchers should get to see their work on the rendered map, but they shouldn't need to set the highway tag to do so. The renderers should show untagged ways as sketchy grey lines. ...and sketchers should use untagged ways
Are you arguing that sketchers should use highway tags, and then these other people should use maplint to highlight unnamed streets? (or they could use that Kosmos rendering I came up with recently) To me that sounds like a reasonable argument, but I know some people (quite a few people who chipped in on this mailing list debate) would not agree, and certainly it feels to me, that this approach is disrespectful towards the superiour surveyed data. There is also the (relatively minor) point that some ways *should* be unnamed, and these will always show up as incomplete patches under this scheme i.e. The presence or lack of a name tag isn't a water-tight system for distinguishing sketched vs surveyed streets. If the my suggested rendering change was made, it would all become clear, and sketchers would no longer need to go around setting highway tags.
-- Harry Wood 11:37, 7 February 2008 (UTC)

I like the idea to draw sketched ways as a gray line. I was faced with the problem myself. Before I do a survey, I usually sketch the area as much as possible. So I can print that area and use it for the survey to write down the data. This keeps the map in proportions and allows me to plan my route. But to get a map I have to set a tag after sketching. Which will result in the problems you described. Since this produced several wrong tags in the first place I'm not very happy with this approach. To have a significant different look on the map (when without a tag) would help a lot, and finding streets that need to be surveyed would become obviously. I support this suggestion. Toralf 19:25, 24 January 2008 (UTC)

Thanks Toralf. You could offer your support to the change here too : Tiles@home/Dev/Appearance#untagged ways as faint grey lines
Glad to see someone agrees. I thought maybe I was going mad.
One day I will do as Richard suggests, and fire a message to the mailing list to try to get more support for the "untagged ways as faint grey lines" change.
Anyone else agree?
Harry Wood 11:57, 7 February 2008 (UTC)
First off, I wouldn't oppose rendering untagged ways as you suggest. I'm not particularly bothered by sketching (and have done so myself). I think that if you can tell by the imagery that a street is residential, it's okay to put the highway tag on it. But sketchers shouldn't have to invent a tag to get the street rendered. Alternatives to not tagging at all: highway=not_yet_classified, fixme=unsurveyed, ... Anyway, go ahead and write to the mailing list -- that's what it's for. Robx 19:31, 25 March 2008 (UTC)

Copyable or not copyable?

When I first saw the aerial imagery, I thought: "Are they crazy? People will trace it and lawyers will bite!". Then I googled and found out we actually have permission.

Still, we have permissions for Yahoo!'s own imagery, not other imagery that Yahoo! also provides, right? See, for instance, the city of Milan in Italy (where I live). On this page, it's marked as "not covered" - and indeed, if one looks at the tiny writing in the bottom right corner of the map, one sees "Copyright 2007 Navteq, Teleatlas".

I suspect that Navteq and Teleatlas would not allow us to trace on their imagery, would they? Yet, I think it's exceptionally hard for the casual user (assuming there is such thing on OSM) to realize that only some of the imagery they see on the map is free for tracing! Myself, I only put two and two together when I saw Milan listed as "not covered"... otherwise, I'd just have assumed that the copyright notice was there by mistake.

Unless there's something I've misunderstood, I think this is quite a serious issue. There should be a notice saying "Warning, if you see 'Copyright anyone-but-Yahoo' on the map, you cannot trace it", or, even better, the non-Yahoo! imagery should not be displayed altogether. Seriously, people won't just notice the small print!

LjL 21:19, 13 September 2007 (BST)

You have misunderstood (fortunately). :) We have permission for all the imagery that is delivered through the Yahoo API. Yahoo has the rights to sublicense derivation rights, and they have given us this permission. It has been cleared by their lawyers. --Richard 22:30, 13 September 2007 (BST)
Then I suppose Milan can be moved from "Not covered" to "Covered", since as far as I can see, the imagery for it is quite high resolution (but anyway, how come if I access the imagery via maps.yahoo.com I get "Copyright 2007 i-cubed", while on the OSM editor it says Navteq and Teleatlas?).
Still, perhaps the Legalities section should be expanded with pointers to whatever the Yahoo! lawyers said, or the relevant discussions between OSM members, or whatever... I've browsed through the legal-talk mailing list, and couldn't find anything very relevant.
The "On the usage of our Aerials API" statement that's reported on this page doesn't seem very clear to me: "the OSM's usage of the free API for overlaying GPS and tracings appears to be well within the terms of use" may be read to mean, simply, that they have no problem with OSM displaying their imagery with maps and GPS traces over it. But what OSM is doing is creating said maps based on their imagery, not just overlaying it...
So I think a couple further clarifications on the wiki page would not hurt.
LjL 13:15, 14 September 2007 (BST)
Yes Milan is covered. I've changed the page accordingly. The list on this page is manually edited, and is supposed to be the list of everywhere where Yahoo has aerial imagery coverage, but the places keep getting added (to the coverage), so it's not always properly up-to-date. Feel free to add any others to the list, where you spot them.
Also feel free to use the imagery via Potlatch or YWMS plugin.
The copyright messages on the images, are not entirely wrong, because Yahoo are not signing away all their rights to the images, they are just allowing us to derive vector based maps from them. For example we are probably not permitted to download lots of them and store them/distribute them separately. It's true it would be better if we had a clearer written statement on these points, but rest assured we are allowed to trace our vector-based maps over the images.
-- Harry Wood 13:44, 14 September 2007 (BST)
I see the reason the copyright message is there, I just wonder why it says "Navteq, Teleatlas" when the imagery is actually copyrighted by i-Cubed, and it says so on maps.yahoo.com. It only says Navteq and Teleatlas if you click on "Map" rather than "Satellite". I guess it's just some little technical issue with the applet. LjL 14:26, 14 September 2007 (BST)
Ah yeah that's a bug in the Yahoo! API aparently. Potlatch/Bugs#Bugs in Yahoo! API -- Harry Wood 14:31, 14 September 2007 (BST)

source:yahoo

Would it make sense to start using source:yahoo? --U9f4y1f941f9h9 04:46, 15 July 2010 (UTC)

Seems there is some discussion about it on Talk:Key:source..

Goodbye, Yahoo

Might be a good idea if you are so inclined to express your appreciation for the years of imagery made available to the Project by posting a comment at http://developer.yahoo.com/blogs/ydn/posts/2011/06/yahoo-maps-apis-service-closure-announcement-new-maps-offerings-coming-soon/ . I've not done so yet, but will likely do so in the near future. --Ceyockey 01:31, 25 September 2011 (BST)

Still works in Potlatch...

http://www.openstreetmap.org/edit?editor=potlatch&lat=39.1057&lon=-94.56938&zoom=18 I see Yahoo imagery here. --NE2 12:48, 14 October 2011 (BST)

I was going nuts, because I missed announcement. Otherwise I could map all the areas for my city, which exists only in Yahoo. But I can also see Yahoo images in Mapzen. Can't we try to recover data, and put WMS in a separate place or connect to Yahoo in another way? City I'm talking about, is a big, capital of a country. Without Yahoo, 80% is left uncovered. Is there anyone else interested in doing this?--Xelgen 13:46, 14 October 2011 (BST)
Huh? --NE2 09:38, 15 October 2011 (BST)

It's weird. I'm fairly sure I tested it and found the imagery to be unavailable, but now it does indeed appear to be available again! I guess we should update the page again. -- Harry Wood 11:36, 17 October 2011 (BST)