[Imports] licence? | Re: Importing data from alltheplaces.xyz

[Minh Nguyen]

Thanks for taking a look, Rory. I think we cleared up the 
misunderstanding in OSMUS Slack. Kun was interested in adding businesses 
to this undermapped area, thought it was necessary to use 
alltheplaces.xyz as a primary source rather than a corroborating one, 
and unfortunately thought an "import" was the only way to go about it.

It's kind of ironic, because the U.S. has a cottage industry of 
businesses that scrape business listings exactly like this and charge 
big bucks to use it in search engines, knowing that copyright law 
doesn't have anything on them. But OSM has additional sensitivities that 
I'd hate to burden a well-meaning small-town America mapper with.

The good news is that -- to my pleasant surprise -- there's decent 
Mapillary and OpenStreetCam coverage in this corner of West Virginia. So 
Kun will be able to improve OSM's coverage of Moundsville without going 
down the rabbit hole of an import process.

Vào lúc 08:41 2020-08-22, Rory McCann đã viết:
> That site is “A growing set of web scrapers designed to output 
> consistent geodata about as many places of business in the world as 
> possible.”
> 
> I don't think it's CC0 licenced. It's a collection of other databases. 
> Have they gotten permission from all the business that they scrape that 
> they can licence this resultant data as CC0? I don't think so.
> 
> They could claim the data is CC0, or that the moon is made of cheese, 
> but that doesn't mean the data is CC0.
> 
> I don't think you can import this.
> 
> On 22.08.20 14:04, Kun Attila wrote:
>> Dear OSM community,
>>
>> I've been mapping Moundsville, WV nowadays (one of my friends live 
>> there so i chose that location). I'm ottwiz on OSM 
>> (https://www.openstreetmap.org/user/ottwiz) from Hungary. Sometimes i 
>> didn't know how to publish imports correctly  from 
>> https://www.alltheplaces.xyz/( it has CC-0 with waiver license), 
>> because i never imported anything. I already talked in US slack, and 
>> they saw no problems with importing these data, just they warned me 
>> they can't be 100% accurate, because the "spider" bot takes the data 
>> from websites that publish their coordinates on a map, and sometimes 
>> it can be like 100-200 (or more) yards off the place. Without further 
>> ado, I only imported those which are 100% accurate. (checked on other 
>> maps, that's why, but of course there is some that is on correct 
>> position on Google, but not in this database, so i omitted those ones, 
>> because taking stuff from Google is not complying with OSM rules). I 
>> also put a "Taco Bell" store, which address is 103 Lafayette Ave, 
>> Moundsville, WV 26041. It's between two shops (which have 101 and 105 
>> Lafayette Ave respectively), so it was easy to guess the location.
>>
>> It's a thing that I already finished my importing, just someone told 
>> me if i didn't go to the location, I shouldn't use "survey" for 
>> source. OK, i got it.
>> So, if nobody has any questions, i'd leave the already imported stuff 
>> alone(it's like 15 shops around Lafayette avenue, changesets which 
>> contain them: https://www.openstreetmap.org/changeset/89655873 ; 
>> https://www.openstreetmap.org/changeset/89742498); , and open a case 
>> in the catalogue and a small wiki page that i did this import on OSM. 
>> Sorry, just this is my first import, so i didn't know i had to take 
>> these steps.
>>
>> Ottwiz
>>
>>
>> _______________________________________________
>> Imports mailing list
>> Imports at openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/imports
> 
> _______________________________________________
> Imports mailing list
> Imports at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/imports


-- 
minh at nguyen.cincinnati.oh.us