World map for a peculiar project

[StickGrinder]

Hi there everybody. I subscribed this site some months ago but since my progress have been very spaced out due to family deeds, and since I had to surrender using GIMP (too few time to learn a new instrument the right way), I had some hard time posting on the board. Now I started to work slowly but steadily on my personal project and I'd like to have authoritative feedback both on my "rookie" efforts and on my basic idea.

About the project

All started when I thought about building a custom setting to drive a fantasy campaign with my fellows role players. I'm kinda family guy, dad and all the rest, so my time is very short. But since I role play without pauses since when I was young, I've never been scared by long time efforts. What I needed was a map, I ended up here and I found Saderan tutorial (which map stunned me for its beauty). Since my other passion (and my job) is programming and I do it the free/open-source way, I thought I could join the two mindset and provide my map for free to everybody to build their campaign settings on.
But soon I realised that a full-fledged world is hampering when you're building your setting. I mean, I would never use another setting map "as-is" as I want to choose where to place people, how to call cities, and so on. On the other side, building the whole world is a long process. The solution I've found to this dilemma is the core idea of my project. As I learned here, a fantasy world could be built bottom-up using what we can call a "multilayer protocol". First geology, than life forms, then people migrations, civilisation, social layer etc. This is a recurring concept in computer-science and I'd like to apply it to some sort of free/open world-building project, providing the first layer of the stack.
What I'd like to end up is a founding geological map that will be freely available in many levels of detail, so that it could be redrawn to fit many styles. My plan is to release the landmasses schema when I'll be happy with it on a dedicated site, then publish more detailed close-ups of the whole landmass, and in the end start working out my Saderan-like version of the world. In the meantime the site will be open to contribution based on the same landmass, so that the world could be redrawn (either entirely or region by region) in many different styles to fit many different settings. The rule of thumb will be that only geological maps, without cities/rivers etc. will be considered contributions. A full-fledged map will be considered derivative work. Obviously nothing prevents people to post the geological base of a derivative work as a contribution, so that others could take advantage from it.

About the world itself

Given my goals, I'm trying to conceive a map that even if imaginary, could be considered likely. After some bad attempt I came up with this:

fwp_map.png

Here I have tried to draw taking continental drift into account, given that I imagined it shaped by two spiralling movements, one big on the right and one smaller on the upper left.
I also included poles landmasses. The north one is intended to be a true continent (as our south pole), while in the south I imagined icy islands slowly (on a geological time scale) drifting around the pole.
I think this shape could be a good kickstart for a lot of different justification, thus enriching settings based on the map, given that authors would like to take this into account.
I tried to figure out the scale of this all, overlapping a true photographic shot of our earth, but I ended up thinking that this will definitively be an "application-related" issue.

Don't bother with lakes, since they are there just because I didn't addressed them yet. I still don't know if getting rid of them for the very basic "shared landmasses" or fixing their position, as well as rivers and relieves to give consistency to the whole, allowing contributors to just decide on climate and its effect on plant distribution, deserts and such. Obviously climate could result in empty basins, but well... I think I've made my point!

Now what I'd like to ask you is... advices and critics! Both on my idea and on the map itself.
Sorry for the looooong post and thank you for any feedback I'll receive.

PS: if someone could come up with a good name for such a project, he/she will be paid in beers!
PS2: my English is far from good, so please be patient if I wrote something wrong or "odd"... just point it out and I'll try to avoid errors in the future.

[arsheesh]

This sounds like a very worthwhile project Stickgrinder. Just a heads up though, to my knowledge no one has yet been able to convert Tear's Saderan Tutorial to GIMP. The tutorial involves using certain features (i.e. layer styles, layer groups, brush styles) that GIMP simply lacks. While there are ways around some of these issues (e.g. there is a layer styles plugin for GIMP), I've not found any way to replicate anything remotely close to the brush presets (bevel & emboss, drop shadow etc) that are required to product features of the landscape in Tears tut. What I'm saying is that either you need PS for this map, or you might want to try a different style of map. Best of luck though.

Cheers,
-Arsheesh

[StickGrinder]

Hi Arsheesh and thanks for your reply. Actually after month of frustration trying to "port" (not just replicate) Saderan on GIMP, I fell back on PS. It seems there is no escape. My plan is to drop PS for all other derivative works and self-contributions that are not Saderan-styled, since I really need PS just to replicate that stunning effect.

About the idea, I'm glad to hear that it sounds good. I searched for a map provided with similar philosophy but found nothing, so I rolled my sleeves up. Now I hope this will take roots somehow, but we'll see.
Bye

[RobA]

Hoy StickGrinder!

About the only comment I have is that (assuming this is a map of the whole world, toes to tops) there is much distortion not accounted for in the polar regions. Here is your map reprojected:

polar.png

The rest of the continents look very nice!

-Rob A>

[StickGrinder]

Hi there RobA! You're right, I tried to switch to polar coordinates with PS but I was unsure this was the real issue. Is there a method to fix it in a decent way? How did you produced your test image above?
Sorry for question bombing Many thanks!!!

EDIT: mmmh, looking at the picture I think I got the point: I didnt take into account more spacing between poles profile and image border. In fact north pole continent should be wider; the same is for the sea area surrounded by isles in the south. I'll try to fix this!

[Hai-Etlik]

EDIT: mmmh, looking at the picture I think I got the point: I didnt take into account more spacing between poles profile and image border. In fact north pole continent should be wider; the same is for the sea area surrounded by isles in the south. I'll try to fix this!

It's a fundamental problem with the Equidistant Cylindrical projection you are using. Anything that would look right on a globe is going to look squashed/stretched progressively more and more toward to poles, and vis versa. That's one of the reasons the projection isn't used much except where its simplicity outweighs its problems.

All projections have distortion, but by choosing projections carefully you can minimize its effects on your map's intended purpose. If you want shapes generally preserved, a conformal projection what you want, like Mercator or Stereographic. Sinusoidal, Hammer, and Mollweide are equal area options that keep things the same size relative to one another. Equal Area Cylindrical projections like Gall-Peters are particularly ugly so you should probably avoid them. If you want to balance different kinds of distortion out, then a hybrid projection is best, like Winkel Tripel or Robinson, though they often have a rather "modern" look to them.

Switching to any of these would require some editing of your map to fix the distortion. Reprojecting would preserve the distortion (As you can see in RobA's polar azimuthal projections) while simply declaring the map to be in another projection wouldn't work as the map is the wrong shape (Mollweide for instance is elliptical) so you'd have to clip off or fill in bits and then line things back up along the edges.