The point of my explanation above was merely that when data is retrieved and parsed from specific URL “X”, it does not affect legality which specific HTTP client / programming language was used to do that fetching+parsing (e.g. Firefox+JS or ATP+Python). So, my point was that either both HTTP clients are legal to use in OSM, or both are illegal to use in OSM.
In other words, if one is allowed to copy/paste opening_hours data from https://lidl.hr from their Firefox browser to OSM, then it is also legal to copy/paste from output of that Lidl ATP python script to OSM. And conversely, if copy/pasting from Firefox browser is illegal, so is copy/pasting from ATP.
I also agree here too.
For that other necessary part for importing data to OSM, i.e. whether OSM mapper is allowed at all to copy those opening hours from websites to OSM, I defer to the decision in those LWG minutes. (It seems to me that it says it is fine to copy opening hours from https://lidl.hr to OSM, but if people disagree with that interpretation, it is completely different discussion than the one in last few of my post, and can be discussed of course)
Also, note that that whole multi-post explanation of mine only holds for that lidl.hr that I’ve checked (although if I had to wager I’d guess that other Lidl sites behave similarly). I agreed already there might be other ATP scrapers which do other things, and if those use different data sources than what the official public 1st-party website itself uses, then the whole equivalency of “1st party” obviously does not apply.
I would expect that it absolutely can affect legality, based on the text of some of the legal terms of service I’ve read over the last 40 or so years. I’m not a lawyer, but there have been plenty of times when I’ve thought “I don’t think I can agree to that”.
However, that is a discussion about the general case. In this specific case, has anyone tried asking the shop whether the data on their website is (a) free for OSM to use (with all the caveats that come with that) and (b) is free of anyone else’s IP?
Have anyone ever encountered any example of any such ruling (e.g. that is it illegal to view public website with Firefox but legal with Chrome, or any similar example of legal discrimination by HTTP client used to access public website and its components)? I find it very hard to believe (and while any such ToS would be null and void in my little part of EU anyway, I’d be very curious to hear of such case).
No, because that unanimous LWG decision stating (quote) “Copying the opening hours of a business from its own website is fine.” So that particular question is already solved as far as manually copy/pasting the data from 1st-party data from lidl.hr into OSM is concerned.
At least, AFAIK that LWG decision has not been disputed by anyone so far?
The sub-thread rather went into automation aspect of ATP, specifically whether fetching data from exactly the same URLs from which web browser visiting the website fetches it itself counts as “copying from website” (with me arguing that it does).
In any case for that specific shop lidl.hr (which I just used as example because I can read the language easily, and not because I particularly care about that specific shop), it would be much easier and faster to just copy/paste those dozen on so opening hours manually as already indicated as “fine” by LWG.
In short, the whole point was generic look at a thing. I’d say nobody cares about Lidl specifically that much to read this whole sub-thread just for them.
Can you explain what “illegal” means (and doesn’t mean) to you? I get the feeling that it’s rather more narrowly defined than would typically be interpreted around OSM.
To me, “illegal” is something which is prohibited by a law. IOW, if someone does a thing that is illegal, they are breaking some law which applies to them (which usually, but not necessarily, might have some negative consequences for said lawbreakers).
I don’t even know how to parse “what it doesn’t mean?” part of the question, though. It sounds like “count all things in the world which you think are not dodecahedrons” kind of a instruction; it would take me a while to compile such near-infinite list (but maybe it was intended as a kind of Turing test? )
But you got me curious - what does “illegal” mean to you?
I know and I agree. For example, if I went to https://maps.google.com to retrieve those opening hours for Croatian Lidl stores, that would be illegal to copy/paste to OSM (as they are 3rd party, and different rules apply than to 1st party; see the LWG decision for rationale).
However, if I copy Croatian Lidl opening hours from https://lidl.hr website, I would take LWG decision of “Copying the opening hours of a business from its own website is fine” to mean exactly what it says - that it is fine if I do it (for all the reasons mentioned in that decision for copying from 1st party website). Would you agree with that interpretation? Or do you think I may not copy/paste opening hours from https://lidl.hr website and enter them in OSM?
I was actually thinking about the difference between criminal and civil law in England and Wales. For example, if I trespass on someone else’s land, I haven’t immediately “broken a law”, but the landowner can initiate legal action to get me to move on. There’s nothing on the statute book that says that “You must not go on $patch_of_land, and if you do the punishment will be XYZ”, but there are potential legal consequences of doing so. “potential legal consquences” is not quite the same as “illegal”, but is still to be avoided.
OSM has traditionally avoided situations where it could be legally challenged by people with more money to pay lawyers than we have, even if, in a fair and balanced process OSM might actually be in the right; for the simple reason being that any legal cost could far outweigh other costs of runnng the project.