As recently as 3 years ago, these were the dominant cartography-software workflows, as garnered from a survey of CartoTalk posts:

With the exception of a few of these products, many cartographers still use one or more of those paths on a daily basis.

Might the following compilation of the most-tagged terms on GIS Stack Exchange give us a glimpse into what the most common workflows of the future will be?* The number of references is listed on the right.

* Two caveats: (1) QGIS uses stackexchange as their primary forum for questions and answers (but that doesn’t preclude its applicability). (2) There are quite a few new tools out there that people are having success with that aren’t in the top tags.

There isn’t a lot of chat on this blog about specific software. Here are some reasons: (1) it would be a lot of work to compare and contrast all the software out there and it’s been done (wikipedia comparison) (2) the focus of the blog is more on design–it’s up to you to figure out how to get to that well-designed end-point, (3) the list of products keeps changing, and (4) everyone has their favorite workflow and they are all good.

If anyone can benefit from a quick, very rough, back of the napkin style analysis, then here’s an attempt at a two minute overview of various software options for making maps. Note that we’re talking about making maps, not GIS data creation, maintenance, or analysis. In the interest of expediency these are not complete sentences. It’ll be okay.

ArcGIS = Accurate! Not usually very fancy cartography output unless you try REALLY hard and even then there might be some difficulties (but I’ve done it).

Adobe Illustrator = Smooth, rich. Not accurate if you move things around. This software is difficult if you are used to ArcGIS. For example, you have to use a special button for selecting the artboard in order to move it, not the regular select buttons. But then, I’m biased.

MAPublisher = Get smooth, rich Illustrator outputs but with added ability to preserve accuracy via projections, edit features and tables while maintaining spatial integrity (i.e., edit your GIS data). For professional cartographers who do non-scientific style maps like tourist maps and the like, MAPublisher and Illustrator might just be the best bet.

Photoshop = You could export from ArcGIS to Photoshop to do some final styling, and folks who are very experienced with Photoshop (but maybe not with Illustrator) will go this route. Illustrator now has a lot of the Photoshop features so if you’re new, you might just head straight to Illustrator, which is better for print, typography and whatnot since it has vector capabilities. If everything you’ve got is in raster maybe Photoshop would be better for you. This doesn’t seem likely. But I’m not expert on Photoshop.

Global Mapper, Manifold, Freehand = Some people use these products to produce maps. Has Manifold had a recent release? As of a few months ago it hadn’t. Can’t say much about these software products except that some people seem to really like them. Worth checking out at any rate, if you are new or want a change, or need something cheaper.

Etc = QGIS is easy and free and has some okay carto-qualities. There are other options, this list is certainly not exhaustive and is entirely focused on print mapping rather than dynamic, digital mapping, for which there are a lot of open source tools out there. I’m keeping my eye on Kartograph too.

Add your likes and dislikes in the comments.

I’ve just been evaluating the MapSpeller extension to ArcMap, at the request of Edgetech America. An employee there had noticed my admonitions to map makers to do spell checking and asked if I’d evaluate their product.*

The key thing about MapSpeller is that it is a location-sensitive spell checker. For example, you might tell it that “Peterson” street is spelled incorrectly and should be “Petersen.” When it spell-checks that area again on the same or on another map that you’ve made, it’ll remember that that location label is “Petersen.” If there’s another street on the same map named “Peterson” it’ll be able to differentiate the two. This is a great thing for massive map labeling projects. Their term for this feature is locationary. It also takes into account what words are not likely to be placed on a map but are similar to words that would be used on a map. For example, it’ll mark “sight” as a potential misspelling for “site”.

Using the extension was easy and intuitive. The only major downside was that the install process was a bit cumbersome in that you had to unzip and save three text files to a certain directory on your machine. It would be nice if that were automated. One license for the standard version is $295. The key difference between the professional and the standard versions seems to be that the professional version has the capability to spell check attribute tables while the standard version does not (it just spell checks what is on screen). Check out their website for other licensing options for large firms and the like.

There are plenty of customization options just like in full-featured word processing software. You can chose to have it not check words in all-caps, camel case, and/or upper-case. There are three English dictionaries to choose from: UK, Canada, and US (additional language dictionaries are planned, according to their website). You can also change the distance at which the locationary determines if a word is spelled correctly or incorrectly.

The really great thing is that there’s a full-featured evaluation version that lasts for 3 months. So it is definitely worth checking out!

*I was not paid for this review.

I just finished making a series of maps and was trying to note my processes while creating them. Here were my thoughts during the week of map-creation:

• Decide work-flow / software first. I didn’t have a lot of time to make these so I decided to stick with a full ArcMap production flow without using Illustrator. I’m not as good at Illustrator and work super-fast in ArcMap. I might have moved everything to Illustrator at the end to finish up some line work (simplify lines) but that would have taken more time than it was worth in this case. The lines looked decent enough anyway.
• Uh yeah – don’t forget that you shouldn’t put a north arrow on a map that’s in a Robinson projection because in Robinson, north changes depending on where you are. Use graticule lines to indicate north or nothing at all.
• Colors still take a long time to come up with. For one of the maps, which showed European countries in a 4-color scheme, figuring out what those four colors should be took several hours of trial and error. And I was even working with the colors for maps booklet to help me along. You just never know how it is going to look on the particulars of the geography you are working with until you try.
• Think about what your data is telling you and try to show it with your design so others can get that same message. Don’t underestimate the power of a good think-session.
• When making both a color and a grayscale version of the same map – as was the case here – create the color version first, then modify to grayscale.
• The least fancy but easiest way for me to create great labels (if there aren’t an unwieldy amount) is to auto-label then convert those labels to annotation and hand-modify from there.
• Don’t forget to zoom in on the layout to place things exactly right but also don’t forget to zoom back to a 1:1 scale periodically to make sure it looks good there too.
• For some reason ArcMap makes my 6-point Arial labels look impossible to read on-screen. But when I print out the map or convert to PDF and open in Acrobat they are legible again. Know what your final output is and design for it by doing some test-exports, even at the beginning of the project.
• If you have a list of specifications for the map, be like Santa Clause and check them twice.
• If there are examples of maps similar to what you are trying to make, it can save a lot of time to sift through them and determine what you’d like to emulate and what you want to avoid.
• You know you are well immersed when you consider at least 20 different shades of blue for the ocean background before settling on one, then changing your mind again.

I’ve written about the exigency of critiques in the map-design process in the past. Usually I am all for them. They provide feedback that is very important for the novice and even advanced cartographer that help to make the map better. I’ve even suggested that children and non-cartographers can be a source of good critique in that their untrained eye may uncover foibles or potential improvements that other cartographers could gloss-over.

However, yesterday, as I was plowing through yet another few pages of Mark Twain’s autobiography (a behemoth) I saw the other side of the equation. In one of the chapters, Twain describes how a wealthy man approached him to write an introduction to a new translation of some Joan of Arc papers. Because Twain had already written a book about Joan of Arc (Personal Recollections of Joan of Arc) he was a natural for the task.

Unfortunately, after Twain submitted the introduction to the wealthy man, who was a novice editor, the man edited it and sent it back. The way Twain describes the situation is much more humorous than I’m making it sound here, but suffice to say that Twain was not pleased and wound up writing a reply saying that the man needed to not publish the introduction whether edited or not and that, further, the man was to destroy any copies of it so that it would not wind up in print.

I’m not sure what Twain would have done if the proposed edits had actually been appropriate ones (they were not) but it seems as though he would have had a similar reaction regardless. Apparently one does not edit an author’s introduction…ever.

The main moral for me was that one should consider your own level of competence relative to that of the critics’ and act accordingly. In other words, do not be afraid to throw out critical remarks that you deem unhelpful or just plain stupid.

I’ve been thinking lately about how my workflow has changed since I started out in GIS 12 years ago. One obvious change is that I no longer take copious workflow notes in spiral bound notebooks and Word documents like I used to. It used to be very important that every step was recorded for every intermediate dataset created and analysis conducted so that going back to a previous iteration after going down a wrong path would be less painful.

In basic terms, this meant that if I was currently working on ForestConversionStats6_clip (!) and discovered that going back to the non-clipped, unresampled, data from yesterday was necessary, that this could be done by looking in the notes and finding the dataset labeled, say, ForestConversionStats2 and starting again from there. I’m not kidding, I had MANY notebooks with dataset names and processes.

I still take notes, usually just in Notepad and saved to the project directory that I’m working in, but they focus more on specific parameters or perhaps long selection strings that I can just copy/paste when needed. Over the last five years or so it’s been nice to finally feel like the steps to create a good analytical result are second-nature (not for every single possible thing that can be done, but for many of the common tasks) enough to not need to back-track or write everything down.

With specific regard to cartography, this type of professional growth manifests differently. The cartographic workflow is very often not written down and not revisited. In fact, it can be difficult to tell if the end result (the map) is even meeting the original needs. Or at least it is harder to discern than the end result of an analysis. Either an analysis works or it doesn’t. You might even run some error tests on that analysis to test if it works, if you are a good scientist. But with cartography there are two barriers: #1 you don’t know if the map is as good as it could be and #2 even if you realize it isn’t as good as it could be, you don’t feel like going back and starting from a previous step.

The seasoned cartography professional is much more adept at those two steps than the novice. The seasoned map maker, though this may come as a surprise, is more critical of the end result and more knowledgeable about how that product should look. They will do the equivalent of error-testing analysis: they will send it out for peer-review, whether formal or informal. Also, they will take the time to change things based on that self-critical assessment and that peer-review, even if it means completely changing one of the key first decisions that had been made.

For example, let’s say that a map needs to be made to display at a public meeting about a proposed nature area. The cartographer makes a key decision at the beginning of the project to zoom out quite a bit from the nature area so that a nearby city is shown. The cartographer reasons that the overall context of the nature area in relation to the city area is important. However, when the map is done it is apparent that the map is too busy and the nature area has wound up not being the central focus as it should be. The experienced cartographer will go back and start over again. The novice may just keep it as-is.

I suppose you could argue that a seasoned cartographer doesn’t make bad choices to begin with. It may be that design decisions like that become easier and better with experience. However, no matter how experienced we are, there is always a chance that we will make blunders. It’s if we fix them that matters.