FE map with scale

Quote from: robinofloxley on July 29, 2021, 06:41:35 AM

Quote from: JackBlack on July 29, 2021, 04:21:44 AM

You didn't just open it.
You opened it, and then took a screenshot, producing an image file.

The screenshotting was in no way necessary

But it was done.
That means you have now converted it to a different format.
But even if you didn't, it still doesn't look like what is on OSM's website.

So again, did you make a new map, or merely convert an existing one?

Again, your way of looking at things is unnatural to me. I might see some nice wallpaper and take a photo in case I decide that's the one I want. I don't call that converting the wallpaper. You are quite at liberty to install some software and open the thing yourself and then the screenshot is entirely beside the point.

I don't care if it doesn't look the same, that's down to styling. Note that this is an "export". Typically export is not the same as a simple save. I can write a Word document formatted with a variety of fonts, colours etc. If I "export" this as text and try and open it again, all the styling is gone, but I still recognise it for what it is.

You simply asked me if I believed this thing, this OSM file, was a map. I opened it with some software which natively understands it and it looked very much like a map, so it passed my duck test sufficiently well, so I called it a map.

You can take a black and white photocopy of a map and I'll happily call it a map even though it's not the same as the original.

Quote from: robinofloxley on July 29, 2021, 06:41:35 AM

It may not look exactly like your original map

i.e. it was converted to something different.

If it was, I think it was the export process which did that. The opening didn't convert anything as far as I'm aware, what you see is what you get. Why don't you open your OSM export and read the documentation on their Web site and show me which part of the file tells you how to style it. It might be there for all I know. You are assuming that the export file is a complete representation of what's on screen in your browser when you perform the export, including all the styling information. I don't think it is, but you can easily prove me wrong.

So again, did you make a new map, or did you convert an existing one?

And again - neither.

Quote from: robinofloxley on July 29, 2021, 06:41:35 AM

Yes, it's another variation of my flexible duck test. It will just open and look (and indeed work if it has embedded calculations) as a spreadsheet, so it's a spreadsheet.

And did you know that what it actually is is a zip file with a collection of primarily xml files?

Again, so what. The PNG and GIF standards and formats are different, but to me they are both still image formats, so I'm happy to call them image files and when opened, images.

So long as the XLSX file opens in Excel and works as a spreadsheet, then it easily passes my duck test and it's a spreadsheet as far as I'm concerned. The binary XLS file isn't a more pure form of spreadsheet just because it's a binary format.

Quote from: robinofloxley on July 29, 2021, 06:41:35 AM

Go ahead then let's hear a properly formulated sound, deductive argument, which will inevitably make me reverse my position. A nicely formulated syllogism perhaps.

As you are the one claiming to have made a map, the burden is on you to do that.

I'm quite happy with trial by jury or some alternative where a human being or beings weigh up evidence presented and make a judgement. You are the one suggesting an alternative. I don't believe that your approach works in this case because it comes down to opinions. Opinions can be changed with argument and evidence. If you think your approach works (you suggested it), then let's see it. We've both made claims, we both need to demonstrate our claims. So far we've both failed to do so. I don't accept your claims, you don't accept mine. Stalemate.

Quote from: robinofloxley on July 29, 2021, 06:41:35 AM

I already explained that my gold standard is you just double click and it opens.

That means your gold standard fails for a .osm file

Yes

and declares that it is not a map.

No. I don't have a single yes/no test for this. I look at the evidence on a case by case basis and make my decision. My gold standard to me is a no-brainer. Pass that test and it is what it is. If I can't double click on it, but can open it with some software I already have, then this isn't quite as firm, but I'm still calling it. If I have to download some software, but otherwise everything checks out, then I'm less certain still, but I'm still calling it.

If someone sends me a .model file and tells me I need CATIA to open it, I'm not paying £50k for some software just to look at it, so I'll provisionally take their word for it that it is what they say it is, but if you ask me if I'm sure, then no, I'm not at all sure.

Right at the other end of the scale, someone sends me a file. The file type is generic (.TXT) or something I've never heard of and Google comes up empty. I can't find a single piece of software anywhere, paid or free that even claims to open this thing natively. Maybe the person who sent it claims it's a cat photo. I'm rejecting this one for sure, pending further evidence.

Quote from: robinofloxley on July 29, 2021, 06:41:35 AM

So to complete the task, I have to break it down, but I mustn't break it down. Yeah that makes sense.

Are you capable of responding honestly?

That is almost literally what you said. I have a complex task (your claim). Obviously that task needs to be completed. You said the way to do that is to break it down. OK, so I need to break it down. You then say I mustn't break it down. So which is it? How am I going to solve this so called complex task then if I can't use the very approach you suggested?

The aim is to solve this task, no matter whether the task is simple or complex. If I can do that by breaking it down into simple steps, then the job is done and I don't really care whether you call it simple or complex. Whatever it was, it was solved, simply.

You say this task is complex, you can't explain why it is complex, now you appear to be saying, it's simple to solve. OK, I'll run with that, it's an incredibly complex task which is incredibly simple to solve.

There is a very big distinction between completing a task and trying to find where complexity is.

Fine, I no longer care whether it's simple or complex, it's easily solved, correct?

So no, I'm dodging the question, you are continually either failing to understand what complexity is, or dishonestly pretending to not understand.

As usual you are just saying it's complex with no backup argument. But we're past that now, whether it is complex or not is irrelevant to me if it is easily solved.

And likewise, you are trying to pull the same dishonest BS as you first did when claiming map making is easy, by trying to ignore the hard part.

A part is something you get when you break something down remember? And I'm not allowed to break the task down because then I end up with simple parts and the complexity has gone, so there is no hard part in this case, there are only simple parts.

But no, why don't you explain how it is all simple, describing the entire process, starting from nothing and ending up with a map.

Well I just did, a three part process. What did I miss out? Follow the steps I gave you and you inevitably end up with a Web site containing a collection of geotagged locations. By your own definition, this is literally a map.

And with more dishonesty, you also seem to love ignoring significant parts of my post.

I'm not responding to everything you raise, no. Firstly these posts end up close to the limit as it is. Secondly I'm bored with all the tedious wrangling over terminology which at the end of the day boils down to your opinion and my opinion and neither of us are going to shift now are we. If I choose to respond at all it'll be on my terms.

Again, your way of looking at things is unnatural to me. I might see some nice wallpaper and take a photo in case I decide that's the one I want. I don't call that converting the wallpaper. You are quite at liberty to install some software and open the thing yourself and then the screenshot is entirely beside the point.

Conversion is typically discussed in terms of computers, where you convert from one format to another.
There are even online format converters, which convert from one format to another.
Even things like converting from PNG to GIF is converting.

Again, the here is if you made that map, or if you merely converted an existing one. It is clearly not the same, and not in the same format. So which is it?
This shouldn't be a difficult question.

It is clearly not the original map. So is this a conversion of the existing map, or is it a new map?

it passed my duck test sufficiently well

Only for you, with the right software. For me it failed. The best I could do is get it as a bunch of text. No map. So applying the same standard you used for GeoNames, it failed, unless you want to try suggesting my computer is defective because it can't open it to get anything that would be instinctively recognised as a map?

You are assuming that the export file is a complete representation of what's on screen in your browser

No, I'm assuming the export file is a list of nodes, and a list of relations and so on.
For this entire discussion the only time I have gotten into presentation (which would include styles), is to use this as an example that doesn't have that presentation to show your argument against GeoNames not being a map because of presentation would mean the .osm file isn't a map either.
And your defence of it being an export also applies to GeoNames, as that is an export, or dump, from their database.

Again, so what.

Because there is other software you can open it in to get a different result.
If I give you a spreadsheet saved as a .xlsx file, but have the extension changed to .zip, it opens as a zip file. Does that suddenly mean it isn't a worksheet?
No.

Again, what is important is the information present.
It doesn't matter what file extension it has, or in what way the information is presented; what is actually important is the information.
And the original images you produced had no additional information compared to the GeoNames .txt files

That is the point I am making. What matters is the information.
So if all you did was change the presentation of the information, that means all the information necessary to be a map exists in the geonames file.

The fact you can open it and present it in a different format has no bearing on if it is a map.

I'm quite happy with trial by jury or some alternative where a human being or beings weigh up evidence presented and make a judgement. You are the one suggesting an alternative. I don't believe that your approach works in this case because it comes down to opinions.

I am suggesting you provide a standard for what is required for something to be a map, which can then be applied to things which both sides agree are maps and both sides agree are not, and give that same determination.
So far what you have provided does not meet that.
The simple questions you asked regarding what a map is used fails, fails with your map.
Your argument about presentation and your simple fail with the .osm format.

But with the standard I have provided, the only points of disagreement are when you disagree with it being a map.

No. I don't have a single yes/no test for this. I look at the evidence on a case by case basis and make my decision.

You seemed to be appealing to your duck test quite a lot to try claiming that GeoNames isn't a map.
But when something you want to be a map fails, you then do everything you can to try to prop it up.

Again, you have already made software which effectively allows you to open GeoNames and see it as what you accept as a map.

And we are still yet to see what is the case for more common GIS software, with you providing 2 contradictory claims about that.

Right at the other end of the scale, someone sends me a file. The file type is generic (.TXT) or something I've never heard of and Google comes up empty. I can't find a single piece of software anywhere, paid or free that even claims to open this thing natively. Maybe the person who sent it claims it's a cat photo. I'm rejecting this one for sure, pending further evidence.

So if I give you a .osm file, renamed to .xml, with no information at all that it is an export from OSM, you will reject it being a map, and then I can use it to produce a map which I can claim is me making a map, you will happily accept that?

That is almost literally what you said.

It is blatant misrepresentation based upon quote mining.

I said COMPLETING complex tasks involves breaking it down into smaller, simpler tasks to complete.
And in response you trying to say something isn't complex and asking for what part makes it complex, I point that that shows you do not understand complexity, as you cannot take apart a complex thing to try to find what part makes it complex.

Rather than honestly present that as focusing on 2 different aspects, one being completion and the other being determining if it is complex or not, you instead pretend I am talking about the very same thing and contradicting myself.

Fine, I no longer care whether it's simple or complex, it's easily solved, correct?

So you no longer take issue with people saying it is a complex task?

As for easily solved, that then comes down to how you vet these people to make sure they aren't just "in on the conspiracy", and how you are determining latitude and longitude in the first place.

there is no hard part in this case, there are only simple parts.

A simple part can be hard. Simple does not mean easy. Even you emphasised that.

Well I just did, a three part process. What did I miss out?

The steps that make up those parts. Especially as you emphasised making your own software as you couldn't trust the software of others.

I'm not responding to everything you raise, no.

Yes, you seem to be avoiding anything that show you contradicting yourself and which would undermine your position.

Even things like converting from PNG to GIF is converting.

Sure, if I can effortlessly open a PNG or GIF with the same software, then I'm calling what I see an image, I don't care what the underlying format was. I might have a reason to convert from one format to another, happy to call that conversion. I can open an XLS and save it as XLSX and call that conversion. I don't view opening a file and looking at it as conversion. I open spreadsheets, images, maps, documents. If all I do is that and I don't save them again, doesn't count as conversion. Again, we're arguing meaning of a word and at this point I don't really care. You have your definitions, many of them I find unnatural, I have mine. We'll never agree.

Again, the here is if you made that map, or if you merely converted an existing one. It is clearly not the same, and not in the same format. So which is it?
This shouldn't be a difficult question.

My beautifully formatted Word document in a lovely cursive font isn't going to look the same saved as plain text. I reopen the saved file, won't look like the original. It was the export to text process which was responsible for difference. The saved file cannot be made to match the original, because information was lost during the export process. Same I believe with the OSM export.

You are right, did I make a new map, or did I convert an existing one is not at all a difficult question. I did neither. Simple answer to a simple question. You might not like the answer, not my problem.

Quote from: robinofloxley on July 30, 2021, 02:10:35 AM

You are assuming that the export file is a complete representation of what's on screen in your browser

No, I'm assuming the export file is a list of nodes, and a list of relations and so on.

Quite so and GIS opened and rendered exactly what was in the file as presented. The export missed out crucial information which would have told GIS how to display a road, how wide the line should be, what colour to use etc. etc. The export is not what you saw on screen. If you throw away information that's your problem.

Quote from: robinofloxley on July 30, 2021, 02:10:35 AM

Again, so what.

Because there is other software you can open it in to get a different result.
If I give you a spreadsheet saved as a .xlsx file, but have the extension changed to .zip, it opens as a zip file. Does that suddenly mean it isn't a worksheet?
No.

Yes. It's now useless junk unless you convince me otherwise. You can claim a zip is anything you like. I don't believe you. Why should I? Tell me how I can view it as a spreadsheet and I'll revise my opinion.

I send you an encrypted file. You cannot open it without the decryption key. I claim it is a cat photo. Do you accept my claim?

Again, what is important is the information present.

Not if you can't do anything with it. Send me a corrupted spreadsheet. Useless and why should I even believe it is a spreadsheet?

Quote from: robinofloxley on July 30, 2021, 02:10:35 AM

I'm quite happy with trial by jury or some alternative where a human being or beings weigh up evidence presented and make a judgement. You are the one suggesting an alternative. I don't believe that your approach works in this case because it comes down to opinions.

I am suggesting you provide a standard for what is required for something to be a map, which can then be applied to things which both sides agree are maps and both sides agree are not, and give that same determination.

I'll start with this definition from Wiktionary:

    "A visual representation of an area, whether real or imaginary, showing the relative positions of places and other features."

Google "world map", look at images, countless examples of things which fit this definition.

To me, a map has to be visual. I can't offhand think of anything I would call a map (in the sense I'm using the word, there are other uses obviously, such as a map of the human genome, mathematical map functions etc.), which isn't visual. I guess a braille map might, but even then as a sighted person, it's visually a map.

We've covered a number of things which can be processed, altered and interpreted in specific ways to produce something which would then present itself visually as a map. The difference here is I would view those things as ingredients, you don't. A difference of opinion.

If GeoNames choose to call their product a spatial database, then I'm not going to argue with them. If they had called it a map, we wouldn't be having this conversation. If dictionaries and thesauri consistently equated map and spatial database, we wouldn't be having this conversation.

Again, you have already made software which effectively allows you to open GeoNames and see it as what you accept as a map.

I could easily write software and turn it into a series of notes and call it avant-garde atonal music. A uses B and produces C. It doesn't follow that B = C.

Quote from: robinofloxley on July 30, 2021, 02:10:35 AM

That is almost literally what you said.

It is blatant misrepresentation based upon quote mining.

You can believe whatever you like. It sounded like a contradiction to me. That's fine, you have clarified it. I think I get the point now.

Quote from: robinofloxley on July 30, 2021, 02:10:35 AM

Fine, I no longer care whether it's simple or complex, it's easily solved, correct?

So you no longer take issue with people saying it is a complex task?

I don't accept that you've demonstrated or explained the complexity, but as I've said, I no longer care.

As for easily solved, that then comes down to how you vet these people to make sure they aren't just "in on the conspiracy", and how you are determining latitude and longitude in the first place.

No vetting. If anyone doubts a contribution, they are welcome to submit their own findings. If someone "in on the conspiracy" submits some deliberately falsified results, anyone is free to call that out and check it. Similarly any dishonest FEer doing the same can be called out. Frankly these will be much easier to spot because they will have to claim something e.g. The Eiffel Tower is not where the books tell us it is.

Quote from: robinofloxley on July 30, 2021, 02:10:35 AM

Well I just did, a three part process. What did I miss out?

The steps that make up those parts. Especially as you emphasised making your own software as you couldn't trust the software of others.

You insist that a collection of geotagged items is a map, so just the locations themselves are enough, they do not have to be turned into a visual representation to count (according to you). That makes the job a whole lot easier.

Quote from: robinofloxley on July 30, 2021, 02:10:35 AM

I'm not responding to everything you raise, no.

Yes, you seem to be avoiding anything that show you contradicting yourself and which would undermine your position.

I'm avoiding the repetitive cycle where you claim I'm contradicting myself and I say I'm not and keeping within 8192 chars (just).

I don't view opening a file and looking at it as conversion.

You did more than simply opening a file. You produced an image from it.
Regardless of if you needed to do that or not, you have done that.

Again, it really is a simple question (and no, saying you did neither does not answer it). It is quite clearly not the same. So is it a new map, or is it merely a conversion an existing one? Or is it just a change/update of an existing one?

This ties into your claims about GeoNames, did you make a new map, or did you merely convert an existing one? Or would you prefer that to be neither as well?

I open spreadsheets, images, maps, documents. If all I do is that and I don't save them again, doesn't count as conversion.

And that should produce an identical file, with only the metadata changed noting it was accessed and saved more recently.
If you open it one format and save it in another, that is now a different file, stored in a different way, making it a conversion.

Again, we're arguing meaning of a word

Yes, we are arguing the meaning of words, to argue if what you did constitutes making a map, and thus map making is trivial, or if it just counts as converting an existing map (or something else).
All you did was taking an existing file, open and save it in a different format (with the loss of a lot of information).
As far as I am concerned, that is no more making a map than taking a screenshot of Google Maps, opening a .osm file and then saving it as an image.

If you don't want to care about the exact definitions, then don't object to people saying you didn't make a map, and instead focus on the real issue, if what you did would satisfy what FEers want? That is making a map without needing trust what they would dismiss as a conspiracy. And that would mean not using existing sources of data, regardless of if you want to call it a map or a database.

Quote from: JackBlack on July 30, 2021, 05:24:03 AM

No, I'm assuming the export file is a list of nodes, and a list of relations and so on.

Quite so and GIS opened and rendered exactly what was in the file as presented.

Not quite. The GIS software left out information about the nodes, such as their name.
But it did just open the file and show what was there with a particular presentation.

And what did you do?
You effectively opened the GeoNames file, which is a list of nodes, and rendered what is in the file.

Again, if you aren't going to appeal to presentation, the only justification you have for saying the .osm file is a map, but the GeoNames file isn't, is the lack of connection information in the GeoNames file. But your map doesn't have that connection information either.

Yes. It's now useless junk unless you convince me otherwise.

And do you notice the problem?
You are the one presenting GeoNames, not me.
Again, if take a .osm file, and present it as a .xml file so it can't just be opened to see it as a map, and claim to have made a map from it, would you accept that, assuming you had no idea about the .osm format in the first place?

Do you not notice the problem you are causing with that standard?
You either have the burden on the person not wanting to meet it, or you have it on the person who initially has no idea what it is.

If I do that, I have no reason to try to show the file is a map, just like you have no reason to try showing GeoNames is.

I'll start with this definition from Wiktionary:
    "A visual representation of an area, whether real or imaginary, showing the relative positions of places and other features."

Which means the .osm format is not a map.
It is logical or symbolic representation, providing a list of nodes and how they are connected.

This means this fails right from the start, as you agree that the .osm format is a map, but it does not meet your definition.
It is only when it has been opened in a piece of software which renders it in a particular way that it becomes a map.
And again, if that is allowed, then what you did with GeoNames, to present that information in a visual way would mean GeoNames is included.

And in fact, GeoNames had a place to do just that, complete with more information by using Google Maps as a base.
https://www.geonames.org/maps/mountains.html
It would allow you to plot all the points to get a nice visual representation, allowing you to choose what points you want to have.
It also has a map, quite similar to what you did, but better and with actual processing:


How about the second definition on wikitionary:
"A graphical or logical representation of any structure or system, showing the positions of or relationships between its components."
Notice the key part here is that it includes logical representations which means the .osm format is included. But it also means that the GeoNames format is. So the only part it disagrees on are those we disagree if they are a map or not.

Google "world map", look at images, countless examples of things which fit this definition.

Amazingly enough, if your limit your search to only look for images, it isn't surprising that all you get are images. But notice how no .osm file appears that image search?

If GeoNames choose to call their product a spatial database, then I'm not going to argue with them. If they had called it a map

Again, the sample applies to a street directory. They call it a street directory, not a map.

we wouldn't be having this conversation. If dictionaries and thesauri consistently equated map and spatial database, we wouldn't be having this conversation.

Again, they are not equal. Just like medicines are drugs are not equal and you do not find dictionaries equating them.
They are 2 overlapping sets. Not all maps are spatial databases.
The other issue is that most dictionaries do not have a definition for "spatial database" that is because spatial is being used an adjective for the database.

But here is something from ArcGIS, considering you want to appeal to GIS so much:
https://pro.arcgis.com/en/pro-app/latest/help/data/geodatabases/overview/the-architecture-of-a-geodatabase.htm
https://pro.arcgis.com/en/pro-app/latest/help/data/geodatabases/overview/feature-class-basics.htm
Here they are describing how they use a geodatabase (which would just be another word for spatial database), and how it stores various information which the GIS software presents as what you would accept is a map.

I could easily write software and turn it into a series of notes and call it avant-garde atonal music. A uses B and produces C. It doesn't follow that B = C.

You sure love contradicting yourself don't you?
Excel USES the .xlsx file to display a spreadsheet. So by the same standard the .xlsx file is not a spreadsheet just because it can be opened in excel.
Again, I was using YOUR STANDARD to show that the GeoNames file is a map. But then as soon as your own standard shows you are wrong, you reject it.

No vetting.

So with the FEers claiming a global conspiracy your system allows those in on the conspiracy to just produce a massive collection of data based upon that conspiracy.

You insist that a collection of geotagged items is a map, so just the locations themselves are enough, they do not have to be turned into a visual representation to count (according to you). That makes the job a whole lot easier.

But that doesn't tell me how you do those steps, and those steps were for the volunteer.
How are you setting up this website?
How are you having people determine their latitude and longitude?

I'm avoiding the repetitive cycle

It is only repetitive because you contradict yourself. Again, you said GIS understand GeoNames, but then when that would support it being a map, you claimed it has no idea what it is.

Quote from: robinofloxley on July 30, 2021, 07:18:46 AM

I don't view opening a file and looking at it as conversion.

Again, it really is a simple question (and no, saying you did neither does not answer it). It is quite clearly not the same. So is it a new map, or is it merely a conversion an existing one? Or is it just a change/update of an existing one?

This is a false dichotomy. You ask A or B. The possible answers are A, B, both, neither, undecidable. You are falsely asserting that the first two are the only available options. They are not and I choose neither as my answer. I've repeated my answer over and over. You keep re-asking the same question. I'm not engaging with this question any more, you have my answer, I will not change it.

If you open it one format and save it in another, that is now a different file, stored in a different way, making it a conversion.

Exactly so, completely agree. You opened OSM, saved it in a different format, a conversion. You converted it. The saved file is not the map you saw on screen.

I opened the file, it doesn't look like the original. Of course it didn't, you converted it, it isn't the same thing. I did no conversion at all, just opened what I was given. It looks like a map, that was your question to me, is an OSM export a map. My answer is yes. Does it look like your original, no, so what, that was not the question you asked.

Quote

I'll start with this definition from Wiktionary:
    "A visual representation of an area, whether real or imaginary, showing the relative positions of places and other features."

Which means the .osm format is not a map.

I opened it, it looked like a map. it's a map. I open a photo, it looks like a photo, it's a photo.

And in fact, GeoNames had a place to do just that, complete with more information by using Google Maps as a base.

Of course, you start with a base map and layer some additional spatial information on top, that's a very common use (but not the only use) of a spatial database. If a spatial database was a map, starting with a base map would be redundant.

Congratulations, you've actually found someone else who has used my very same technique. Not sure that proves anything, but well done.

How about the second definition on wikitionary:
"A graphical or logical representation of any structure or system, showing the positions of or relationships between its components."

I'm not certain, but that usage sounds more like a map as used in graph theory:

"In graph theory, a map is a drawing of a graph on a surface without overlapping edges (an embedding)." (source WikiPedia)

In any case, you asked me for my understanding of map, I chose a definition which I thought encapsulated what I think of when I hear map. I'm sure there are lots of different definitions out there, none are entirely satisfactory, I chose one which fits my particular viewpoint.

Quote

If GeoNames choose to call their product a spatial database, then I'm not going to argue with them. If they had called it a map

Again, the sample applies to a street directory. They call it a street directory, not a map.

To be honest I'm not entirely sure what a street directory is, it's not a term I've ever used, but Collins says "another name for street map", so it's obviously not clear cut is it.

But here is something from ArcGIS, considering you want to appeal to GIS so much:

I'm not reading all that. Does it say anywhere that a spatial database is a map? Pull out that quote for me and I'll change my mind.

Quote

No vetting.

So with the FEers claiming a global conspiracy your system allows those in on the conspiracy to just produce a massive collection of data based upon that conspiracy.

All of which could be checked by anyone with an interest. Half the people I see spouting FE nonsense I think are trolls who don't believe a word of what they are saying, so I don't think there is any vetting process short of waterboarding which would sort that out.

How are you setting up this website?

Same way I would set up any other Web site, is this supposed to be a difficult thing?

How are you having people determine their latitude and longitude?

Quite happy to leave it up to them, so long as they state how they did it. If someone uses GPS and someone else objects, they can just go there and repeat the survey with a sextant or whatever. Many flat earthers seem quite happy with GPS, because it takes them to where they want to be. They argue it isn't space based, that's fine, so long as it works.

Are you guys actually arguing over if the data used to build a map is actually a map or not?

Is a tree a forest?

Wait... it's moved on to if a screenshot of a GIF is still a GIF?

I'm so confused.  Is this like asking which Vision is the real one?

Are you guys actually arguing over if the data used to build a map is actually a map or not?

I think so, but in order to confirm that we probably need a debate about the meaning of "argument" and "data".

Is a tree a forest?

Please don't do this to me.

Wait... it's moved on to if a screenshot of a GIF is still a GIF?

I'm so confused.  Is this like asking which Vision is the real one?

I think I came here for a 5 minute argument and ended up with the full half hour.
 
" class="bbc_link" target="_blank" rel="noopener noreferrer">

This is a false dichotomy. You ask A or B. The possible answers are A, B, both, neither, undecidable. You are falsely asserting that the first two are the only available options. They are not and I choose neither as my answer.

If you want to state it is neither of those options, then state what it actually is.
Again, it quite clearly isn't the same, so nothing is not an option.

Quote from: JackBlack on July 30, 2021, 03:14:11 PM

If you open it one format and save it in another, that is now a different file, stored in a different way, making it a conversion.

Exactly so, completely agree.
...
I did no conversion at all

It was a .osm file, with an xml format.
You then provided it as a .png file, a fundamentally different format. That is converting it.

I opened it, it looked like a map.

You opened it in software which converted the logical structure into a visual one.
But the .osm file itself has none of that visual information. It is a logical representation.

That means that by that standard the .osm file is not a map.

Alternatively, if you are happy with a piece of software taking that logical data and presenting it in a visual format, then you did that with GeoNames, meaning GeoNames is a map.

Of course, you start with a base map and layer some additional spatial information on top, that's a very common use (but not the only use) of a spatial database. If a spatial database was a map, starting with a base map would be redundant.

No, it wouldn't. That is like saying, if you have a map, every other possible map is redundant.
You have a map showing countries? That's great, every other map is redundant, you don't need a map showing streets, or businesses, or parks, or anything like that, that would just be redundant.

Or you could accept that maps can have different information, and thus using a different map as a base layer for another map is not redundant.

Congratulations, you've actually found someone else who has used my very same technique. Not sure that proves anything, but well done.

It is a map, on GeoNames site, using the data of GeoNames. But no, they didn't use your technique. They actually processed the data rather than just plotting it.
If you notice, their map has different brightness values for the same region, based upon how many points are there.

I'm not certain, but that usage sounds more like a map as used in graph theory:

"In graph theory, a map is a drawing of a graph on a surface without overlapping edges (an embedding)."

It is a more general definition of a map.
But notice how the one you provided requires it to be a drawing, which is less inclusive than the definition I provided which allows a logical reprsentation, like the .osm file.

In any case, you asked me for my understanding of map

I specifically asked for one which doesn't disagree with things we both either agree are maps or agree aren't.
The definition you chose excluded the .osm file as it is not a visual format. It is a logical format.
The fact that you can open it in different software and get quite different representations shows it is not a visual format.
The fact it likes the styling information to determine how to display it shows that it is not a visual format.

As I have explained many times, if you want to appeal to presentation, of it being visual, then the .osm format is not a map as it is not a visual format, it is not an image.

I chose one which fits my particular viewpoint.

i.e. a viewpoint in which you can pretend making a map is easy?

To be honest I'm not entirely sure what a street directory is, it's not a term I've ever used, but Collins says "another name for street map", so it's obviously not clear cut is it.

Which was the very point I was making. Person calling some X, doesn't mean it isn't Y.
As you want to bring up Collins, how do they define a spatial database?
"Sorry, no results for “spatial database” in the English Dictionary."

I'm not reading all that. Does it say anywhere that a spatial database is a map? Pull out that quote for me and I'll change my mind.

So you want me to pull out a quote which would be made directly to you?
If you want that, you email them and see what response you get.

Or, you could accept what I already provided describing it.
Again, it stores information as a geodatabase.
This includes the information which it portrays on screen, visually, in a format you accept as a map.

So by that standard, that geodatabase IS A MAP.

But how about this, considering you want a quote, how about you try providing a direct quote from OSM stating that the .osm file is a map.
In fact, here is a quote from their wiki:
"The major tools in the OSM universe use an XML format following a XML schema definition that was first used by the API only. Basically it is a list of instances of our data primitives (nodes, ways, and relations)."
Notice how they describe it as a "list of instances", not as a map. -

If you look on their help forum, you have someone just as helpful as you:
"Strictly speaking, you haven't download a "map", just the raw data that something can create a map from. A .osm file is a text file, so if you open it in a text editor you'll see the raw data. Most programs that "do things with OSM data" can read .osm files, but which of those you'd use depend on whether you want to view the data, edit it, or create some sort of map from it and then view that." - https://help.openstreetmap.org/questions/30178/how-do-i-openview-the-map-which-is-in-osm-file-format-exported-area
That sure sounds like your argument against GeoNames.

And as you wanted to appeal to a tutorial, how about this:
https://learnosm.org/en/osm-data/file-formats/
"Like any type of data, there are various ways of storing geographic data on a computer. It can be saved in a database, which is a specialized system for storing and retrieving data, and in fact there are database systems specifically designed for storing geographic data."
"The .osm file format is specific to OpenStreetMap."
"OSM data, on the other hand is designed to be easily sent and received across the internet in a standard format. Hence, .osm files are coded in XML, and contain geographic data in a structured, ordered format. A simple .osm file would look like this if viewed in a text editor:"
"Acquiring data in .osm format is easy - in fact you do it every time that you download data in JOSM, but using these files for analysis and map design is not easy. Hence you are better off converting the data into another format, or getting it from a service that converts the data for you."

"Many types of information are stored in database systems, which provide a logical way of organizing and accessing data. Geographic data is no different, although databases designed for geodata are specialized to handle the complex functions that querying geographic data requires."

"OpenStreetMap data is often stored in a PostgreSQL database with PostGIS extensions. This type of database provides fast access to the data and can be used easily with Mapnik, a piece of software that creates the map tiles used in web slippy maps. There are several tools available for importing raw OSM data into a PostgreSQL database."

And based upon another page there, it seems like the .osm format really matches your .txt example with excel:
https://learnosm.org/en/osm-data/osm-in-qgis/
"You can either use QuickOSM to import it clicking on ‘OSM File’ in the left bar. Once you used QuickOSM OSM files should have been made known to QGIS and you can use the regular vector layer import:"
Notice how it clearly describes it as importing, not opening?

So again, there are really only 2 options:
1 - presentation matters, not the information stored. That means neither GeoNames, nor the .osm file are maps.
2 - Information matters, and thus either the connections are important which rules out your image and the GeoNames file, or there is enough information in the GeoNames file for it to be a map.

All of which could be checked by anyone with an interest.

At which point why bother with the process at all, if you are just going to have to check it?

Same way I would set up any other Web site, is this supposed to be a difficult thing?

It isn't a simple case of "set up website" it has parts to it.

Quite happy to leave it up to them

Which means you don't have the entire process.

Is a tree a forest?

Is a large, densely packed collection of trees a forest? Or does someone need to stick up a sign saying forest for it to be accepted as one?

Quote from: JJA on July 31, 2021, 04:53:38 AM

Are you guys actually arguing over if the data used to build a map is actually a map or not?

I think so, but in order to confirm that we probably need a debate about the meaning of "argument" and "data".

Quote from: JJA on July 31, 2021, 04:53:38 AM

Is a tree a forest?

Please don't do this to me.

Quote from: JJA on July 31, 2021, 04:53:38 AM

Wait... it's moved on to if a screenshot of a GIF is still a GIF?

I'm so confused.  Is this like asking which Vision is the real one?

I think I came here for a 5 minute argument and ended up with the full half hour.
 
" class="bbc_link" target="_blank" rel="noopener noreferrer">

I bet you wished you had never asked. Don't say I didn't tell you. You want a map go buy one.

Quote from: JJA on July 31, 2021, 04:53:38 AM

Is a tree a forest?

Is a large, densely packed collection of trees a forest? Or does someone need to stick up a sign saying forest for it to be accepted as one?

Is my local lumbermill a forest because it has a densely packed collection of trees?

Will they still make a noise if one falls off the logging truck?

Can you still see the forest if you can't see the trees?

I need to know.

Quote from: JackBlack on July 31, 2021, 06:39:27 AM

Quote from: JJA on July 31, 2021, 04:53:38 AM

Is a tree a forest?

Is a large, densely packed collection of trees a forest? Or does someone need to stick up a sign saying forest for it to be accepted as one?

Is my local lumbermill a forest because it has a densely packed collection of trees?

Will they still make a noise if one falls off the logging truck?

Can you still see the forest if you can't see the trees?

I need to know.

(Fetches popcorn)

Good luck

But how about this, considering you want a quote, how about you try providing a direct quote from OSM stating that the .osm file is a map.
In fact, here is a quote from their wiki:
"The major tools in the OSM universe use an XML format following a XML schema definition that was first used by the API only. Basically it is a list of instances of our data primitives (nodes, ways, and relations)."
Notice how they describe it as a "list of instances", not as a map. -

If you look on their help forum, you have someone just as helpful as you:
"Strictly speaking, you haven't download a "map", just the raw data that something can create a map from. A .osm file is a text file, so if you open it in a text editor you'll see the raw data. Most programs that "do things with OSM data" can read .osm files, but which of those you'd use depend on whether you want to view the data, edit it, or create some sort of map from it and then view that." - https://help.openstreetmap.org/questions/30178/how-do-i-openview-the-map-which-is-in-osm-file-format-exported-area
That sure sounds like your argument against GeoNames.

Excellent clarification. OK happy to change my mind, if they say it's not a map, I'll go with that, OSM export is not a map. I am not at all familiar with the format, I just went with first impressions and called it a map, I was clearly wrong, my mistake. Happy to accept that it isn't a map.

Ah, but you can make a map with it. Hmm. Is it possible then that something which contains geotagged features and can be used to make a map is not itself classified as a map? Apparently so.

Excellent clarification. OK happy to change my mind, if they say it's not a map, I'll go with that, OSM export is not a map. I am not at all familiar with the format, I just went with first impressions and called it a map, I was clearly wrong, my mistake. Happy to accept that it isn't a map.

Ah, but you can make a map with it. Hmm. Is it possible then that something which contains geotagged features and can be used to make a map is not itself classified as a map? Apparently so.

At least now you are being consistent.
And with that you have set the bar incredibly low.
Want to make a map? Well just download the .osm file from OpenStreeMap, and then import it into GIS software and save it as an image.

Or even easier, just use OSM to load up the area you want to map and take a screenshot. No need to involve any extra software.

Likewise, want to make a cake? Just go buy some premade icing and a pre-made cake base, stick them together and you've got yourself a cake that you made.

I think I will stick to dismissing that as nonsense, and focus on the information rather than the presentation.

Lost in the weeds again, I tried to read it but it wasn't about how to make a FE map. Let me simplify.

Map the following places:

London UK
Capetown South Africa
Tokyo Japan
Sydney Australia
Tehran Iran

First, you need the distances. Airline schedules, google maps, any database you want. If you have different numbers from different databases, that would be more interesting than the FE map discussion, and I am not talking about differences of a couple miles in a thousand because they measured from different points.

If you think all sources are wrong and we can only know by crawling with our own personal ruler, then go sit in a dark room and drool on yourself, because without trusting other people's information, we can't know much at all.

Now lay out those points on a flat piece of paper and a sphere. One will work, the other won't (Gauss's Remarkable Theorem). If the earth is flat, you can plot all these points with their published distances and there will be no problem. If the earth is round and you plot the distances on a flat surface, there will be no way to make the distances match.

So if someone wants to take FE in this debate, lay out these cities with scaled distances on a flat sheet of paper. You will be the greatest FE hero. Cartesian plane is nmuch easier than 3 space geometry, should be easy.

I think I have to lay out the distances on a flat sheet myself, but why spend the time when FE will say "you don't personally know those distances." or simply see things are not going their way and ignore.

Quote from: robinofloxley on July 31, 2021, 08:14:46 AM

Excellent clarification. OK happy to change my mind, if they say it's not a map, I'll go with that, OSM export is not a map. I am not at all familiar with the format, I just went with first impressions and called it a map, I was clearly wrong, my mistake. Happy to accept that it isn't a map.

Ah, but you can make a map with it. Hmm. Is it possible then that something which contains geotagged features and can be used to make a map is not itself classified as a map? Apparently so.

At least now you are being consistent.
And with that you have set the bar incredibly low.
Want to make a map? Well just download the .osm file from OpenStreeMap, and then import it into GIS software and save it as an image.

Or even easier, just use OSM to load up the area you want to map and take a screenshot. No need to involve any extra software.

Likewise, want to make a cake? Just go buy some premade icing and a pre-made cake base, stick them together and you've got yourself a cake that you made.

I think I will stick to dismissing that as nonsense, and focus on the information rather than the presentation.

So, you gave me some evidence to convince me that OSM export is not a map and I changed my mind. Do we now both agree then that OSM export is not a map? Curious to see if we have found some common ground at last.

Lost in the weeds again, I tried to read it but it wasn't about how to make a FE map. Let me simplify.

Map the following places:

London UK
Capetown South Africa
Tokyo Japan
Sydney Australia
Tehran Iran

First, you need the distances. Airline schedules, google maps, any database you want. If you have different numbers from different databases, that would be more interesting than the FE map discussion, and I am not talking about differences of a couple miles in a thousand because they measured from different points.

If you think all sources are wrong and we can only know by crawling with our own personal ruler, then go sit in a dark room and drool on yourself, because without trusting other people's information, we can't know much at all.

Now lay out those points on a flat piece of paper and a sphere. One will work, the other won't (Gauss's Remarkable Theorem). If the earth is flat, you can plot all these points with their published distances and there will be no problem. If the earth is round and you plot the distances on a flat surface, there will be no way to make the distances match.

So if someone wants to take FE in this debate, lay out these cities with scaled distances on a flat sheet of paper. You will be the greatest FE hero. Cartesian plane is nmuch easier than 3 space geometry, should be easy.

I think I have to lay out the distances on a flat sheet myself, but why spend the time when FE will say "you don't personally know those distances." or simply see things are not going their way and ignore.

Now my approach would be a little different. You've listed 5 places. 4 are reasonably accessible, one isn't. If possible, find someone reasonably local to each place, get them to measure where it is (latitude/longitude). Whatever tools you feel are acceptable, so probably either GPS or sextant+chronometer.

Check these locations against published sources. Hopefully they will agree. Now plot these on a globe and use a bit of spherical trigonometry to work out the relative distances. e.g call Sydney to Tokyo 1 unit and base everything on that.

Now you haven't measured any real distances, you didn't have to, but you can can then see if the relative distances make sense with other things such as flights.

If the FEers are right then there must be some mathematical mapping of these locations to a flat plane which exists and can be quantified. It can't be a very complex mapping, so they just need to work out what mapping to use to make everything fit.

So, you gave me some evidence to convince me that OSM export is not a map and I changed my mind. Do we now both agree then that OSM export is not a map? Curious to see if we have found some common ground at last.

Technically, I gave you evidence to show that something you had already accepted was a map was declared to not be a map by others. The purpose was not to show that OSM was not a map, but that your argument works against something that you had already stated was a map.

As far as I am concerned, the .osm file is still a map.

And for the purpose of the discussion on making a map (especially for FEers making a map when they claim conspiracy), taking an OSM file, opening it in GIS software and then saving the result is not making a map.

Quote from: JJA on July 31, 2021, 04:53:38 AM

Is a tree a forest?

Please don't do this to me.

Is cereal soup?

Quote from: robinofloxley on August 01, 2021, 01:55:05 AM

So, you gave me some evidence to convince me that OSM export is not a map and I changed my mind. Do we now both agree then that OSM export is not a map? Curious to see if we have found some common ground at last.

Technically, I gave you evidence to show that something you had already accepted was a map was declared to not be a map by others. The purpose was not to show that OSM was not a map, but that your argument works against something that you had already stated was a map.

As far as I am concerned, the .osm file is still a map.

And for the purpose of the discussion on making a map (especially for FEers making a map when they claim conspiracy), taking an OSM file, opening it in GIS software and then saving the result is not making a map.

So you don't actually believe your own evidence now? You went out of your way to track down evidence, which you don't accept, in order to post it on here, in support of a position which you aren't taking and actively reject.

OK, I'm lost for words.

Quote from: robinofloxley on July 31, 2021, 05:15:39 AM

Quote from: JJA on July 31, 2021, 04:53:38 AM

Is a tree a forest?

Please don't do this to me.

Is cereal soup?

Can I just say that continuing this particular debate about soup, cereal, trees and forests is starting to feel increasingly attractive and a much more sensible and productive use of my time.

Quote from: robinofloxley on July 31, 2021, 05:15:39 AM

Quote from: JJA on July 31, 2021, 04:53:38 AM

Is a tree a forest?

Please don't do this to me.

Is cereal soup?

That is another one I like, where are the lines drawn between soup, juice, stock, sauce, broth, preserve and jam (and so on, I'm sure I have missed some).

So you don't actually believe your own evidence now? You went out of your way to track down evidence, which you don't accept, in order to post it on here, in support of a position which you aren't taking and actively reject.

Did you even understand the point of showing it?

It was to deal with your idea of "Creator must claim it is X, or else it couldn't' possibly be X".

The best you get is someone on their help forum stating:
"Strictly speaking, you haven't download a "map", just the raw data that something can create a map from."

To me, that is no better than you claiming it isn't a map. But they even went to the extent of saying "strictly speaking" and even put "map" in quote marks. They are going for the technical point of view of what you downloaded is a text file, in the same way that if you download a .xlsx file you have actually downloaded a zip file which contains .xml files.

But again, this is ignoring the real point, the idea of what constitutes "making a map".
Is downloading a .osm file, importing it into GIS software, and then saving the image "making a map"?
In the same way that buying a premade cake base and premade icing and then sticking them together (poorly) is "making a cake"

Quote from: robinofloxley on August 01, 2021, 10:29:21 AM

So you don't actually believe your own evidence now? You went out of your way to track down evidence, which you don't accept, in order to post it on here, in support of a position which you aren't taking and actively reject.

Did you even understand the point of showing it?

It was to deal with your idea of "Creator must claim it is X, or else it couldn't' possibly be X".

Oh good, so to address a claim I don't make then.

I know you see things as black and white. I don't. Maybe this will help. Suppose I score 1..5 on the idea that something is (in my opinion) a map.

So a boiled egg scores a 1, page 3 of my world atlas is a 5.

OSM export originally scored a 4, so that's on balance, based on me knowing nothing about it, never seen it before, I'm leaning towards it being a map, but I'm not certain. My main reasons were a) GIS understood it, loaded it no questions asked and it then b) looked like a map (although not like the original, so yes that did raise a question mark), c) It came from something that self-identified as a map. A couple of minor black marks in my opinion are that OSM itself doesn't seem to have a way to reopen it and calls it an export.

You then provided me with some additional information, which strongly suggests that the creators and the people who actively use it, don't regard it as a map. So I've revised my score, based on that to maybe 3 or 2. I'm now between on the fence and no, not a map. I'll happily revise my opinion again if more evidence comes to light.

GeoNames, well that's a 2 for me. You can certainly make a map from it, but apart from that, all the evidence I see is that it isn't regarded by anyone as a map and to me this is much more clear cut than OSM. For starters, I'm much more familiar with GeoNames. I'm not saying categorically that GeoNames is not a map, just that I class it as a spatial database (as does pretty much everyone else) and I don't equate spatial database with map (and nor does anyone else I can find).

Is downloading a .osm file, importing it into GIS software, and then saving the image "making a map"?
In the same way that buying a premade cake base and premade icing and then sticking them together (poorly) is "making a cake"

Well if I grind some coffee beans and add water, have I made coffee? By your standard, I don't think so. I've processed the coffee beans and added something that wasn't there. In the case of GeoNames, I've processed the data, translated and scaled the values, turned them into pixels and added colour.

Oh good, so to address a claim I don't make then.

Which you did not directly state it like that, it was effectively a claim you were making.
You were continually appealing to how GeoNames doesn't call itself a map.

I know you see things as black and white. I don't.

No, I see things as lying along multiple independent axes, and trying to determine if something is a "map" just being giving it a single score doesn't work all that well.
But if you really want that, I will give them a score out of 10. (and something that is quite firmly not a map would get 0)
An online or otherwise interactive map which allows you to zoom in and out and easily get positions from a global scale to that of a small city scores a 9. Pretty much the only thing possibly better than something like that would be to give it vastly more map layers (for example, have the option to just show power lines or pipes or the like, and pick and choose just what layers to show) or have a direct interface to your brain, or map the entire universe like that.

Now, the .osm format is harder. This is because you are then limited to a single export. This export could be of just a tiny block, making it fairly useless, or it could be of the entire world, and if you were making your own, it could be just a single node. So it could score from a quite low 0 to a quite high 8, lacking presentation for humans (and also not directly a great format for computers as they would want to import that data into a proper database or object).

The GeoNames .txt file of the world is easier to judge. It lacks the nice interface, but includes points all over the world. But it lacks connection information between those points, and all the points are labelled. I would give it something like a 5.

Now your map, it takes that geonames file, plots it as points on an image and looses so much of the extra information. I would give it a 4 or 5. Sure, it is prettier and easier for a human, but you lose so much of the information, and pretty much made it useless for a computer to use as a map.

You then provided me with some additional information, which strongly suggests that the creators and the people who actively use it, don't regard it as a map.

No, I didn't.
I provided you with some additional information that they refer to it as something else. If you wanted to see if they regard it as a map, you would need to directly ask them that question.

And the closest you get to that is someone on the help forum (not the creators) clarify that in a strict sense it is not a "map". Based upon what the person wanted:
"how do I open/view the map I have downloaded."
What they wanted is likely to see it as a nice pretty picture.
(In a similar line of reasoning to technically you don't download a workbook. Instead you download a .xlsx file which is a zip file containing .xml files which tell

But that has no bearing on my score.
For my score, I don't care what the creators call it. I judge it based upon it, not what other people say about it.

If the creators of OSM came out and openly declared that OSM is not a map, and I mean the web interface as well as the .osm files, instead stating that what they have is a searchable geographic database that displays the results visually on your screen, and is in no way a map; would that then make you re-evaluate it and decide that even that web interface isn't a map?

If a creator of an Atlas comes out and declares that they do not have a collection of maps, instead they just have some pretty pictures/drawings of Earth, which they do not consider are maps; would that make you decide they aren't maps?

I don't equate spatial database with map (and nor does anyone else I can find).

I don't equate png file with map.
The spatial database is effectively the file format.
You store the map as a spatial database.

Well if I grind some coffee beans and add water, have I made coffee? By your standard, I don't think so. I've processed the coffee beans and added something that wasn't there. In the case of GeoNames, I've processed the data, translated and scaled the values, turned them into pixels and added colour.

It is almost like it isn't a simple black and white standard and it depends on how complex it is to make.
What steps have been done for you? What steps still need to be done/were done by you?
In the case of coffee, the coffee beans have been collected, removed from the fruit and roasted. You then have to grind it, prepare it in a particular which varies depending on what coffee you want, and then add water, again in a particular way.

Due to how simple the overall process is, it is not clear what side of the arbitrary line that would fall on.
Conversely with a cake, the wheat and sugar cane and so on has been harvested and processed in various ways. This produces flour, sugar, and so on. These are then combined in particular amounts and processed to produce cake batter, which varies depending on what type of cake you are preparing, which is then baked in an oven, to produce the cake base you would buy, and so on. Clearly there are a lot more steps, and buying a premade cake base has the majority of those steps done for you, the majority of the work done for you.

Likewise, with GeoNames, the majority of the work is done for you. People have surveyed the locations, determined the latitude and longitude and people have compiled all that on a website, which allows you to get a dump of all the points with latitude and longitude. All you then need to do is run a program which plots those latitude, longitude pairs on an image.

And I would say that falls far more into the category of using a premade cake base than making coffee from beans.

Quote from: robinofloxley on August 02, 2021, 10:01:27 AM

Oh good, so to address a claim I don't make then.

Which you did not directly state it like that, it was effectively a claim you were making.
You were continually appealing to how GeoNames doesn't call itself a map.

The part I am objecting to is this:

It was to deal with your idea of "Creator must claim it is X, or else it couldn't' possibly be X".

You are attributing a level of certainty which I don't hold. As explained, for me, it's a judgement call, an opinion. If I can reword this, I'd say:

If creator does not claim thing A is an X and on the contrary claims thing A is a Y and nobody seems to equate X with Y, then that shifts my thinking further in the direction of believing that A is a Y and A is not an X.

The GeoNames .txt file of the world is easier to judge. It lacks the nice interface, but includes points all over the world. But it lacks connection information between those points, and all the points are labelled. I would give it something like a 5.

So on a scale of zero to 10, 5 is the absolute mid-point, which suggests a 5 is you sitting on the fence, undecided. Now that has me confused. It sounds like you are saying you are very much undecided about GeoNames being a map or not. Is that correct?

Quote from: robinofloxley on August 02, 2021, 10:01:27 AM

You then provided me with some additional information, which strongly suggests that the creators and the people who actively use it, don't regard it as a map.

If the creators of OSM came out and openly declared that OSM is not a map, and I mean the web interface as well as the .osm files, instead stating that what they have is a searchable geographic database that displays the results visually on your screen, and is in no way a map; would that then make you re-evaluate it and decide that even that web interface isn't a map?

It would certainly influence my thinking, yes, particularly as I know next to nothing about this format and I am therefore partially relying at the moment on how others view it.

If a creator of an Atlas comes out and declares that they do not have a collection of maps, instead they just have some pretty pictures/drawings of Earth, which they do not consider are maps; would that make you decide they aren't maps?

It would make me rethink somewhat, but I am perfectly familiar with an atlas and I'm sure there are many many definitions of atlas that include the word map in the description, so this would not be enough to reverse my opinion in this case.

Quote from: robinofloxley on August 02, 2021, 10:01:27 AM

I don't equate spatial database with map (and nor does anyone else I can find).

The spatial database is effectively the file format.
You store the map as a spatial database.

Again, this is you just claiming something, this is opinion, not fact. I don't trust your opinion and I don't agree with it and as far as I can see, you are a lone voice in the wilderness claiming this, so I feel justified in rejecting what you say.

Quote from: robinofloxley on August 02, 2021, 10:01:27 AM

Well if I grind some coffee beans and add water, have I made coffee? By your standard, I don't think so. I've processed the coffee beans and added something that wasn't there. In the case of GeoNames, I've processed the data, translated and scaled the values, turned them into pixels and added colour.

It is almost like it isn't a simple black and white standard and it depends on how complex it is to make.

Sorry, but making coffee is making coffee. Everyone understands what that means.

To answer the question "Is the earth flat?", you just need the distance from 4 widely separated points. If the earth is flat, you can construct a scaled representation on a flat sheet of paper and it will work. If the earth is not flat, there will be no way to lay them out so that the distances are all correct.

Does not matter what a map is. Databases don't matter. Small differences between airline schedule and google maps don't matter. If you have two databases with different distances between the same two points, please show us.

Early in my FE experience, I checked many, many sources for the distance between Sydney and LA because I flew that route and everything matched. They all matched with a few miles. GPS matches in LA and Sydney. The geometric equation to calculate the distance matches. The time/speed/distance of an airliner matches. FE says, those distances are wrong, because they have to be, or the earth is round. So if you want to solipsism it, you can say "nobody knows anything", but then why even discuss, we can't know anything (zetetic).

It is this simple: If you lay out the scaled distances between London, Capetown, and Tokyo on a flat sheet of paper. Now take the published distances between each of these and Tehran. There will not be a place where all three distances to Tehran match up at the same point. Do the same on a sphere and they will match up. It will be the same for any big triangle and a fourth point in the middle of the triangle. If you can do it, the earth is flat.

So you either have to say that all these databases, web sites, etc are wrong, or the earth is round. No need to define map, I have not even used that word in this post. The only question is, are gps, google maps, nav equipment, the entire consistent and daily used with success sources wrong? Because if they are right, the earth is round.

Complexity is the FE's friend. If you want to take the RE position, drive towards simplicity. Do not go down the rabbit hole.

You are attributing a level of certainty which I don't hold. As explained, for me, it's a judgement call, an opinion. If I can reword this, I'd say:

While that may be the case, it is irrelevant for my argument.
It doesn't matter if you are saying it with certainty that they must refer to it as a map for it to be considered a map, or if you are merely saying that them not referring to it as a map is strong evidence in favour of it not being a map.

The fact remains it was targeting that line of reasoning of yours.
I didn't provide it to show that the .osm file is not a map.
I provided it to show something that you accepted was a map can have the same argument you used against GeoNames applied to it.
i.e. your argument against GeoNames being a map works against other things you accepted was a map.

So on a scale of zero to 10, 5 is the absolute mid-point, which suggests a 5 is you sitting on the fence, undecided.

No, the decision point would be more between 0 and 1.
If it is 0 it is not a map.
If it was 1 it has attributes which indicate it is a map and it is a question of how good a map is it.
If something is currently at 10, but then new things come along which are vastly better such that by comparison this former 10 drops to a 4, that doesn't mean it isn't a map, it means there is a significantly better map.

But this is an issue with trying to resolve it to a single scale, and cap that scale at a particular number.
For example, lets say there are 5 independent criteria.
A candidate has a score of 10 for 4 of these, and 0 for one of them.
i.e. it is really good at being a map for most things, but there is one attribute it completely fails at.
What should its overall score be? 0 or 8?
What if instead of 0 for 1 it had 1 for it? Would that switch it from 0 to 8.something?

And capping at a particular number means there is a perfect way to do it, so 10 would be a perfect map. Things which aren't perfect, but are still undoubtedly a map, will have a score less than 10. All that means is that they aren't as good. It doesn't mean that there is less certainty about it being a map.

It would certainly influence my thinking, yes, particularly as I know next to nothing about this format and I am therefore partially relying at the moment on how others view it.

I wasn't referring to the .osm file, I was referring to OpenStreetMaps in general, and superficially to their web interface, e..g this:
https://www.openstreetmap.org/

Again, this is you just claiming something, this is opinion, not fact. I don't trust your opinion and I don't agree with it and as far as I can see, you are a lone voice in the wilderness claiming this, so I feel justified in rejecting what you say.

It is backed up by an understanding of how the vast majority of the internet operates. It is backed up by those links to ArcGIS which you didn't want to read.
If you want a map which is interactive and/or searchable by a computer you don't just store it as an image. That does not give you interactivity you want, and it makes search extremely slow.

The .osm file is a dump of part of that database corresponding to a region.
It is the same data used by their website which portrays it with a particular style.

Sorry, but making coffee is making coffee. Everyone understands what that means.

If only you bothered reading the rest of what I said rather than ignoring it because it directly addresses the issue, of how such a simple task such as making coffee can be considered you doing it, while for a more complicated task, even if you do more than what you would do when making coffee, due to the complexity of the overall task it doesn't count as you doing it.
So great job deflecting yet again.

I provided it to show something that you accepted was a map can have the same argument you used against GeoNames applied to it.
i.e. your argument against GeoNames being a map works against other things you accepted was a map.

Yes and that's why I changed my mind (about OSM) and would be willing to change my mind again if more evidence comes to light.

My original decision to call OSM export a map was a margin call. My switch is also a margin call. I would not be prepared to bet money one way or another. After all I still know next to nothing about it. Far less than I know about GeoNames. I certainly don't know if OSM export is a map. I suspect it is not.

Quote from: robinofloxley on August 03, 2021, 04:43:15 AM

So on a scale of zero to 10, 5 is the absolute mid-point, which suggests a 5 is you sitting on the fence, undecided.

No, the decision point would be more between 0 and 1.
If it is 0 it is not a map.
If it was 1 it has attributes which indicate it is a map and it is a question of how good a map is it.

And yet again, your way of looking at the world baffles me.

Using your scale then to answer a question about a film you'd seen, it looks like your answer would be:

0 - loved it.
1 - 10 hated it to different levels of hatred.

So 11 categories and not one of them equates to on the fence or indifferent or undecided.

And in this case we have 0 - not a map, 1 - a map. 2-10 how good a map it is. As far as map or not map, sounds rather binary to me.

Quote from: robinofloxley on August 03, 2021, 04:43:15 AM

It would certainly influence my thinking, yes, particularly as I know next to nothing about this format and I am therefore partially relying at the moment on how others view it.

I wasn't referring to the .osm file, I was referring to OpenStreetMaps in general, and superficially to their web interface, e..g this:
https://www.openstreetmap.org/

Well that's several different things and several different questions. What I see at the Web site is a map, that's a 5 for me on my scoring system. It does other things as well, it's a full application, but for sure, a map.

I'm not sure what you mean by "OpenStreetMaps in general". It's a big project. There is a Web API, that's not a map, that's an API, there is a Wiki, that's not a map either.

If you mean everything associated with the name OpenStreetMaps, then no not a map. Google is not a map either, it's a multinational corporation with employees, offices, etc. etc.

Quote from: robinofloxley on August 03, 2021, 04:43:15 AM

Again, this is you just claiming something, this is opinion, not fact. I don't trust your opinion and I don't agree with it and as far as I can see, you are a lone voice in the wilderness claiming this, so I feel justified in rejecting what you say.

It is backed up by an understanding of how the vast majority of the internet operates. It is backed up by those links to ArcGIS which you didn't want to read.
If you want a map which is interactive and/or searchable by a computer you don't just store it as an image. That does not give you interactivity you want, and it makes search extremely slow.

Don't you? My understanding of Google maps and OSM is that what you see at any zoom factor is made up of a set of image tiles stitched together. They wouldn't have to be geotagged, you just need to understand which tile fits next to which other tiles. It wouldn't make any sense at all to build a map on the fly at the point of request for someone direct from a database of tens of millions of point nodes, ways and tags and a whole load of styling information when they visit Google maps, it would be a very heavy load on the server. Millions of people use Google maps, that would be an insane amount of work.

Makes much more sense just to have all the tiles stored and ready to go.

Searching is a completely different function. You don't need a map for that at all. That's what spatial databases are for. Traditional paper maps don't have search features.

I can tell you that according to my copy of GeoNames, Yorkshire has a population of 3,978,484, 208,367 of whom live in the city of York. You can't get that from a set of map tiles and it's the set of map tiles that are used to build the actual map.

And yet again, your way of looking at the world baffles me.
Using your scale then to answer a question about a film you'd seen, it looks like your answer would be:
0 - loved it.
1 - 10 hated it to different levels of hatred.

No, it wouldn't.
0 - apathetic.
10 - loved it.
-10 - hated it.

That is because the scale is no longer 1 sided.

What is the opposite of a map which would create that negative side of a scale?

So 11 categories and not one of them equates to on the fence or indifferent or undecided.

And in this case we have 0 - not a map, 1 - a map. 2-10 how good a map it is. As far as map or not map, sounds rather binary to me.

Like I said, that is a limitation of you wanting to try to limit it to a single variable.

Well that's several different things and several different questions. What I see at the Web site is a map, that's a 5 for me on my scoring system. It does other things as well, it's a full application, but for sure, a map.

And if they declare that isn't a map?

Don't you? My understanding of Google maps and OSM is that what you see at any zoom factor is made up of a set of image tiles stitched together.

It has several different layers, including icons they place on top of it, and routes along a road.
How do you think they make them?
And even with tiles, it isn't a single image.
How do you think they determine what image to load? A database.
How do you think they allow you to search for a location? A database.
How do you think they determine routes between locations? A database.

It wouldn't make any sense at all to build a map on the fly at the point of request for someone direct from a database of tens of millions of point nodes, ways and tags and a whole load of styling information when they visit Google maps, it would be a very heavy load on the server. Millions of people use Google maps, that would be an insane amount of work.

Not really.
For starters, they wouldn't load those millions and millions of nodes unless they were actually needed.
For example, as you zoom out, some streets vanish. If they do that, there is no need to send the nodes for them.
And, you can shift the processing to the user's device. You can provide them with the styling information (do you know what a CSS file is) and the nodes and connections and have them draw it locally.

And remember, places like Google don't just show you the streets, they also show other things, for example, they allow you to see the traffic, displaying an almost live map of the traffic. Do you know how insane it would be for them to try to generate images for all the traffic in the world constantly just in case someone wanted to see it?

And of course, you could even just try to look at what data is sent, and see quite clearly that it isn't just a bunch of images. You get other data as well.

And yes, it can show directions, do you think it generates them on the fly, from a collection of nodes, or do you think it does the absolutely insane task of precomputing every possible set of directions anyone could ever ask for?

Here is an example of one of the "images" I got when loading Google Maps:

And, just in case that doesn't display, here it is as a link via WayBack machine:
Link


So it seems Google is quite happy generating images on the fly. Almost as if it has a massive server farm that can handle generating it when the user requests it.

But if you really did want to store it as an image, note: AN IMAGE, as in a single image, not a collection of them stored in a database of some form (even if it is just a collection of folders and file names to be able to quickly get the one for a corresponding region) to know which one to show for what location, then the size would be insane.
Google Maps has resolution better than 1 m. So if you wanted a simple equirectangular projection, that means you would need an image that is roughly 800 trillion pixels, just to have each m as 1 pixel. That is colossal and would require so much bandwidth to send to everyone it is insanity.

As for getting populations, that would depend on what type of map you are looking at. If you are looking at one that only shows streets, you wouldn't expect to get a population out of it.

Quote from: robinofloxley on August 04, 2021, 02:31:46 AM

And yet again, your way of looking at the world baffles me.
Using your scale then to answer a question about a film you'd seen, it looks like your answer would be:
0 - loved it.
1 - 10 hated it to different levels of hatred.

No, it wouldn't.
0 - apathetic.
10 - loved it.
-10 - hated it.

That is because the scale is no longer 1 sided.

What is the opposite of a map which would create that negative side of a scale?

So you now have a scale which goes from -10 to +10. You invent a different scale for every eventuality? Interesting. Why am I not surprised.

My 1 to 5 scale is my degree of confidence that this thing is a map. Do I really have to explain this? Apparently.

1 - absolutely not a map, 2 - probably not a map, 3 - no opinion either way, 4 - probably a map, 5 - absolutely certain this is a map.

Example:

1 - boiled egg, 2 - OSM export, 3 - "Highways and Byways" - Paul Klee, 1929, 4 - London Underground Map, 5 - World atlas, page 3.

Quote from: robinofloxley on August 04, 2021, 02:31:46 AM

Well that's several different things and several different questions. What I see at the Web site is a map, that's a 5 for me on my scoring system. It does other things as well, it's a full application, but for sure, a map.

And if they declare that isn't a map?

I'll be open to a rethink.

Quote from: robinofloxley on August 04, 2021, 02:31:46 AM

Don't you? My understanding of Google maps and OSM is that what you see at any zoom factor is made up of a set of image tiles stitched together.

It has several different layers, including icons they place on top of it, and routes along a road.

Sure, layers. Layers put on top of a map.

How do you think they make them?

Well that's kind of the point, isn't it. You are now calling this process "making", having insisted all along that this kind of thing isn't to be considered making. I didn't make a map, Google/OSM didn't make a tile. As you are so fond of pointing out, you can't have it both ways.

And even with tiles, it isn't a single image.
How do you think they determine what image to load? A database.

I don't know. Neither do you. You are reaching again.

How do you think they allow you to search for a location? A database.
How do you think they determine routes between locations? A database.

Searching and routing are independent of maps as already explained. You can present search results in text format and routing the same way. You use a database and a whole load of mathematics to work out routing. You can present a route (optional) as a layer on top of an existing map or as a list of instructions ("turn left onto A123, follow this for 10 miles..."). Take all the tiles away and the route layer is just a multi-segment line with no context.

Quote from: robinofloxley on August 04, 2021, 02:31:46 AM

It wouldn't make any sense at all to build a map on the fly at the point of request for someone direct from a database of tens of millions of point nodes, ways and tags and a whole load of styling information when they visit Google maps, it would be a very heavy load on the server. Millions of people use Google maps, that would be an insane amount of work.

And of course, you could even just try to look at what data is sent, and see quite clearly that it isn't just a bunch of images. You get other data as well.

Ooh yes, do lets. Google is a bit more obscure in how it works, but that's understandable as it is a commercial product. OSM is, well, Open, so maybe, just maybe they are a bit more transparent. Why don't we look...



So we have a collection of 256x256 PNG tiles coming from tile.openstreetmap.org. I went to the trouble of downloading one of these tiles and no EXIF information at all, so no, they are not even geotagged.

So this map is clearly a set of 16 non-geotagged PNG tiles.

Yes, once you have your map, you can add some extra layers, but that's what they are, layers.



So this shows the same area you were looking at in Google maps. The OSM tile clearly does not have the route on it. In OSM, this is implemented as a completely separate and independent layer.

So it seems Google is quite happy generating images on the fly. Almost as if it has a massive server farm that can handle generating it when the user requests it.

No, it doesn't. Google is far too opaque to make that conclusion. Reaching again.

OSM clearly just sends you the tiles you need in PNG format, it doesn't build a custom image for you from a server farm, it doesn't need to.

But if you really did want to store it as an image, note: AN IMAGE, as in a single image, not a collection of them stored in a database of some form (even if it is just a collection of folders and file names to be able to quickly get the one for a corresponding region) to know which one to show for what location, then the size would be insane.
Google Maps has resolution better than 1 m. So if you wanted a simple equirectangular projection, that means you would need an image that is roughly 800 trillion pixels, just to have each m as 1 pixel. That is colossal and would require so much bandwidth to send to everyone it is insanity.

Completely missing the point. The application in the browser is zoomed in or out to some zoom level or other (OSM has 21 distinct zoom levels). It is also panned, so at any given time, it only needs a handful of tiles from the server's large collection. The bandwidth is trivial.

The storage requirements on the server are huge, yes. Bandwidth, no. I would be surprised if OSM pre-generates every possible tile, many of them I imagine would be hardly ever required (e.g. some random location in the pacific or Antarctic at maximum zoom level for example), so every now and again, it might be asked for some obscure location and there will be a pause while it generates a new tile. But the point remains that your map consists of an array of PNG tiles. Whether those tiles were created 10s ago or a year ago is neither here nor there, they had to be generated at some time or other. You have referred to that process as "making". I wholeheartedly agree.