Let's not beat around the bush here. While it's an excellent map and resource that has been used on this site for many years, the QBAM is wrong.
As far as I'm aware this was first pointed out by @Ashtagon (I'll link to their DA post because I honestly can't be bothered trawling through the forum for the earliest mention of it here).
To summarise, the QBAM isn't in any known projection. It's a sort of hybrid between Robinson and Kavrayskiy VII, which fits because for a long time those were the two projections it was suspected to be. But it doesn't end there. Most of Africa, Eurasia, Australia and the Pacific is located too far to the west by several pixels, and very annoyingly it isn't a consistent misplace, with some areas further out from where they should be than others, meaning you can't just select a whole landmass and move it a few pixels to solve the problem - everything has to be subtly warped for it to work.
Aside from everything being not quite the right shape and not quite in the right position, there were some particularly glaring issues that have been known for a while, particularly around the edges of the map and at the poles where distortion is greatest. Antarctica is completely wrong for example, and this has been known and been pointed out for years. I myself was amused and annoyed to notice several years ago that St. Lawrence Island (the only other significant landmass beside Antarctica cut in half by the edge of the map if you take 10 degrees east as the central meridian) has long been shown on the QBAM as two separate islands, one on each side of the divide.
So yes, the QBAM is wrong. The problem is that I and many others have spent years using it to make AH-maps, and a great trove of associated geographical and historical patches have been produced to complement it. And this is where the sunk cost fallacy kicks in - it was only noticed that the QBAM was wrong after all the associated maps and other material had been produced. Thus most people decided to either ignore the problem and keep using the flawed basemap, or else attempt the almost impossible sisyphean task of trying to patch the problems away.
For a long time I was in the "the problems can be patched if we work hard enough at it" camp, but over the last few years I've slowly been coming to the conclusion that the best thing to do is re-start from fresh. As I mentioned above, sunk cost fallacy is a thing, and eventually you should realise that it's best to start over rather than continue pouring time and effort into a flawed project.
I've had a bit of a long creative slump recently (the same reason my current TL is long dormant unfortunately, though I'm considering a reboot), but back in July I started work on this map, and even once I was done I wanted to continue. So back in October I got to work fixing a major problem of the old QBAM - Antarctica. First I found a good equirectangular basemap (the old 8K-BAM, in case you were interested), shrunk it down to QBAM size and reprojected it using G.projector, then got to work adding a new coastline. I also decided to experiment by doing this using paint.net rather than MS paint, and while it took a bit of getting used to, having separate layers really came in useful. Doing this really hammered home just how wrong the QBAM can be at times, but it also made me realise that I had a good almost-QBAM scale basemap in Robinson that just needed tidying up.
So that's exactly what I started to do.
I consciously wanted this map to be a spiritual successor to the old QBAM, just because I'm so used to using it and have grown fond of it over the years, hence why the style is so similar. The funny thing is that on a regional scale, the QBAM can still be pretty accurate. Most of the borders of the southern states of the US were cribbed from the current most up-to-date QBAM with only minor modifications for example, while Chesapeake Bay is similarly a nearly pixel-by-pixel reconstruction with a few tweaks and modifications.
But that just underlines the problem. On a regional scale things are fine - at a macro scale between regions things break down, with single pixel discrepancies rapidly adding up over larger areas.
And then there were the bigger problems. Just in the areas I've mapped so far there have been some glaring discrepancies. In the US, the Pacific northwest is seemingly scrunched up to the east of where it should be, an island in the Canadian Arctic the size of New York City was missed entirely (Brock Island, in case you were curious), while the states of Mexico were so hilariously wrong in the original QBAM that they had to be redone from the ground up (the northern states were too big, and the southern ones squished and misplaced as a result).
On a related note, I'm also using this as an opportunity to perform a comprehensive geography update to the old basemap, as there are several areas I've never really been happy with.
As I posted on the map thread, progress is coming along slowly but steadily. I'm currently aiming to get the Contiguous United States, Mexico, Central America and the Caribbean done before the end of the year, with Canada and Alaska completed as soon as possible after that. Once North America is done, I'll move on to South America, then either Europe or Africa.
But first some announcements that will probably annoy a few of people.
Firstly, the minor one - I won't be posting updates to the map using any recognised colour scheme. I'm making this map as I would one of my AH maps, which means Colour It Whatever You Like is in full effect, though in practice this just means largely using the old X2 scheme with other colours pilfered from every other scheme under the sun (most notably SUCK and DCS) in a rather slapdash fashion depending on what I think works best or what I like the look of. Apologies in advance, but this shouldn't be a major problem.
What may annoy people who want this done quickly however is that I do not want this to be a collaborative project. I'll get it all done on my own, no matter how long it takes me.
This is because of a problem I noticed following the production of the MBAM - that different people have subtly different styles of making maps. Person A might be a little too generous adding small lakes for example, while person B might abhor single-pixel lakes and avoid putting them in. Put these two styles together in one map and the result is a subtle but noticeable dichotomy between the areas mapped by the different people. In short, I want one person to go over the whole map with the same eye and the same perspective so everything is consistent. Once the map is done I have no problem letting other people patch it and modify it to their heart's content, but I want the basemap at least to all be consistent.
Finally the name. Why haven't I named the new map after myself as others have done before? Honestly, I've been seriously considering but never quite going through with a name change for my account on this site for ages, and it'd be a bit stupid to name such a major project after an old account name if I eventually choose to go through with the change at some point in the near future.
Instead I eventually settled on R-QBAM (with the R standing for either Robinson or Revised, I haven't yet made up my mind on that one). Also, as I mentioned above, I want there to be a lot of continuity with the old QBAM, and naming the new map the R-QBAM seemed like a good way to do that.
OK, now all the introductory stuff is out of the way, let's start off with the first patch;
- Added the Mexican states of Sinaloa, Nuevo Leon, Tamaulipas, Tabasco and Chiapas.
- Purged several lakes probably too small to show up on this scale in Mexico. (I'll do a general purge of Mexican lakes once the whole country is done, but these seemed like the most egregious ones).
- Added Guatemala (without first level divisions - to be added later).
- Added El Salvador (without first level divisions - to be added later).
- Added a portion of the Pacific to encompass the Mexican Pacific islands, Clipperton Island and Cocos island.
As far as I'm aware this was first pointed out by @Ashtagon (I'll link to their DA post because I honestly can't be bothered trawling through the forum for the earliest mention of it here).
To summarise, the QBAM isn't in any known projection. It's a sort of hybrid between Robinson and Kavrayskiy VII, which fits because for a long time those were the two projections it was suspected to be. But it doesn't end there. Most of Africa, Eurasia, Australia and the Pacific is located too far to the west by several pixels, and very annoyingly it isn't a consistent misplace, with some areas further out from where they should be than others, meaning you can't just select a whole landmass and move it a few pixels to solve the problem - everything has to be subtly warped for it to work.
Aside from everything being not quite the right shape and not quite in the right position, there were some particularly glaring issues that have been known for a while, particularly around the edges of the map and at the poles where distortion is greatest. Antarctica is completely wrong for example, and this has been known and been pointed out for years. I myself was amused and annoyed to notice several years ago that St. Lawrence Island (the only other significant landmass beside Antarctica cut in half by the edge of the map if you take 10 degrees east as the central meridian) has long been shown on the QBAM as two separate islands, one on each side of the divide.
So yes, the QBAM is wrong. The problem is that I and many others have spent years using it to make AH-maps, and a great trove of associated geographical and historical patches have been produced to complement it. And this is where the sunk cost fallacy kicks in - it was only noticed that the QBAM was wrong after all the associated maps and other material had been produced. Thus most people decided to either ignore the problem and keep using the flawed basemap, or else attempt the almost impossible sisyphean task of trying to patch the problems away.
For a long time I was in the "the problems can be patched if we work hard enough at it" camp, but over the last few years I've slowly been coming to the conclusion that the best thing to do is re-start from fresh. As I mentioned above, sunk cost fallacy is a thing, and eventually you should realise that it's best to start over rather than continue pouring time and effort into a flawed project.
I've had a bit of a long creative slump recently (the same reason my current TL is long dormant unfortunately, though I'm considering a reboot), but back in July I started work on this map, and even once I was done I wanted to continue. So back in October I got to work fixing a major problem of the old QBAM - Antarctica. First I found a good equirectangular basemap (the old 8K-BAM, in case you were interested), shrunk it down to QBAM size and reprojected it using G.projector, then got to work adding a new coastline. I also decided to experiment by doing this using paint.net rather than MS paint, and while it took a bit of getting used to, having separate layers really came in useful. Doing this really hammered home just how wrong the QBAM can be at times, but it also made me realise that I had a good almost-QBAM scale basemap in Robinson that just needed tidying up.
So that's exactly what I started to do.
I consciously wanted this map to be a spiritual successor to the old QBAM, just because I'm so used to using it and have grown fond of it over the years, hence why the style is so similar. The funny thing is that on a regional scale, the QBAM can still be pretty accurate. Most of the borders of the southern states of the US were cribbed from the current most up-to-date QBAM with only minor modifications for example, while Chesapeake Bay is similarly a nearly pixel-by-pixel reconstruction with a few tweaks and modifications.
But that just underlines the problem. On a regional scale things are fine - at a macro scale between regions things break down, with single pixel discrepancies rapidly adding up over larger areas.
And then there were the bigger problems. Just in the areas I've mapped so far there have been some glaring discrepancies. In the US, the Pacific northwest is seemingly scrunched up to the east of where it should be, an island in the Canadian Arctic the size of New York City was missed entirely (Brock Island, in case you were curious), while the states of Mexico were so hilariously wrong in the original QBAM that they had to be redone from the ground up (the northern states were too big, and the southern ones squished and misplaced as a result).
On a related note, I'm also using this as an opportunity to perform a comprehensive geography update to the old basemap, as there are several areas I've never really been happy with.
As I posted on the map thread, progress is coming along slowly but steadily. I'm currently aiming to get the Contiguous United States, Mexico, Central America and the Caribbean done before the end of the year, with Canada and Alaska completed as soon as possible after that. Once North America is done, I'll move on to South America, then either Europe or Africa.
But first some announcements that will probably annoy a few of people.
Firstly, the minor one - I won't be posting updates to the map using any recognised colour scheme. I'm making this map as I would one of my AH maps, which means Colour It Whatever You Like is in full effect, though in practice this just means largely using the old X2 scheme with other colours pilfered from every other scheme under the sun (most notably SUCK and DCS) in a rather slapdash fashion depending on what I think works best or what I like the look of. Apologies in advance, but this shouldn't be a major problem.
What may annoy people who want this done quickly however is that I do not want this to be a collaborative project. I'll get it all done on my own, no matter how long it takes me.
This is because of a problem I noticed following the production of the MBAM - that different people have subtly different styles of making maps. Person A might be a little too generous adding small lakes for example, while person B might abhor single-pixel lakes and avoid putting them in. Put these two styles together in one map and the result is a subtle but noticeable dichotomy between the areas mapped by the different people. In short, I want one person to go over the whole map with the same eye and the same perspective so everything is consistent. Once the map is done I have no problem letting other people patch it and modify it to their heart's content, but I want the basemap at least to all be consistent.
Finally the name. Why haven't I named the new map after myself as others have done before? Honestly, I've been seriously considering but never quite going through with a name change for my account on this site for ages, and it'd be a bit stupid to name such a major project after an old account name if I eventually choose to go through with the change at some point in the near future.
Instead I eventually settled on R-QBAM (with the R standing for either Robinson or Revised, I haven't yet made up my mind on that one). Also, as I mentioned above, I want there to be a lot of continuity with the old QBAM, and naming the new map the R-QBAM seemed like a good way to do that.
OK, now all the introductory stuff is out of the way, let's start off with the first patch;
- Added the Mexican states of Sinaloa, Nuevo Leon, Tamaulipas, Tabasco and Chiapas.
- Purged several lakes probably too small to show up on this scale in Mexico. (I'll do a general purge of Mexican lakes once the whole country is done, but these seemed like the most egregious ones).
- Added Guatemala (without first level divisions - to be added later).
- Added El Salvador (without first level divisions - to be added later).
- Added a portion of the Pacific to encompass the Mexican Pacific islands, Clipperton Island and Cocos island.
