Archive for category Design

Beyond The Core Knowledge of Cartography

Posted by G.P. in Design, Workflow on October 26, 2016

 

It’s difficult to teach cartography as an expert because there are fundamental concepts that one can easily forget to impart on the students, though this can be ameliorated somewhat by following a comprehensive guide to core knowledge areas such as The GIS&T Body of Knowledge*. Even so, there are certainly some concepts that nobody anticipates struggling with, teacher or student, that are unique to each project undertaken. To tackle these project-specific struggles the cartographer must possess a certain amount of patience, research prowess, and a passion for discovering new solutions through a willingness to test and revise.

It’s these project-specific issues that I haven’t written much about on this blog, instead focusing on fundamentals like typefaces, colors, and layout. Thinking about this hole in my writing, I decided to take notes while creating my most recent print map. Hopefully these notes can be instructive in not just detailing specific solutions to specific problems, but also in providing a more general picture of the types of things that come up in a typical cartography project.

These notes include ideas on skills employed, issues encountered, and paths not taken. Here is the map, in reduced resolution, because it is not considered a fully finished product just yet. The final version will be somewhere around 8-10″ on a side. The notes follow.

 

LewClark

 

PROJECTION USA contiguous Albers equal area conic, an easy choice for the display of the contiguous U.S. and distances should be fairly accurate for the middle latitudes, which are inclusive of this data. This is important since I’m running some distance calculations and reporting on them in the layout, though of course I can always conduct the calculations separately from the display projection by simply starting a new GIS project with a different projection. In this case both needs are met with the one projection. So here I’m thinking, “does the projection look right? Does it calculate right (i.e., not distort the things I want to measure)?”

 

TRAIL DATA The trail line is a subset of linework from a university teaching database and I have not thoroughly explored its validity though comparisons by-eye to other maps of the journey show at least a superficial agreement. My need to trust this line is a problem. What if it’s wrong? How many times to cartographers blindly use data without complete knowledge of it’s correctness? In this case I’m helped by the fact that my partner in this project is a historian who can do some verification for me. In any case, verification of data should be a part of every cartographer’s workflow.

 

LINE RESOLUTION AND DISTANCE CALCULATIONS Another potential issue with the line trail data that crossed my mind was the issue where perhaps the line-digitizer made the line too “squiggly,” perhaps due to an unsteady hand (red line in the zoomed-in example below). If this were the case, then the distance calculations could show the Lewis & Clark crew going much greater distances than they really did. A worry, indeed, and next on my list of things to take a close look at. (Of course, conversely, distance calculations coming from a simplified line would under-report the true distance traveled.)

 

One section of the trail. The non-dashed black line is the real data, the dashed black line illustrates how distance would be too-low if calculated from a simplified line. The red line indicates distance would be too great if calculated from a line with inaccurate and artificially high numbers of vertices.

One section of the trail. The non-dashed black line is the real data, the dashed black line illustrates how distance would be too-low if calculated from a simplified line. The red line indicates distance would be too great if calculated from a line with inaccurate and artificially high numbers of vertices.

 

TERRAIN DISTANCE On a related note, the terrain distance can be much different from the straight-line distance for each segment. The straight-line distance over a ravine is shorter than the traveler’s distance, since they presumably hiked down and then back up the ravine on the other side. In the elevation profile, the true terrain distance is reported. This may account for the fact that my total mileage is greater than the total mileage reported elsewhere for this expedition.

 

LINE SPEED INSET: for this kind of visualization, sometimes we see the width of the line changing by speed but most often it’s depicted with a sequential color scheme instead. I’ve chosen to go with a pseduo-sequential color palette so that all the colors are clearly visible in the small space. The boldness of color needed for the size of the space is an interesting consideration that I haven’t written about before. In this inset, the expedition’s speed is calculated from a length field divided by a to/from date field converted to # of days per segment. This length field isn’t terrain-distance and thus could be a source of error. The inset still serves as a good at-a-glance comparison of the relative time it took the expedition to traverse each segment. The placement of the inset came from an inspiration map. Maintaining a rich library of inspiration maps is a must.

 

PAGE EFFECTS Who knew that you’d need to figure out how to create squiggly lines in Inkscape to evoke old parchment edges? Nobody, that’s one piece that can be learned on-the-fly. Also, get good at finding svg or raster textures to use as backgrounds, especially on these historic map styles. I’ve also seen lots of textures on contemporary map styles, usually in grided dot patterns. (Hint for the squiggly line page border: draw a rectangle and then use Filter>Filter Texture>Rough and Glossy and fine-tune the result with the Filter Editor as the Inkscape filters often default to startingly bold results.)

 

PROFILE GRAPH Exporting a profile graph is possible using a QGIS plugin called the Profile Tool. You can also create profile graphs with ArcMap’s Elevation Profile Add In. The tables of profile data that are produced for each segment of a given line can be exported to spreadsheet software to produce a nice looking graph (the graphs that are automatically output from these aren’t really suitable for a well designed map layout.) Using spreadsheet software (in my case, Excel) I made a graph, switched the progression to East–>West since that’s the direction they were headed, and further styled the graph in Inkscape. This was a process that probably took half a day or more from researching tools to implementation to styling. I even used a pen tool set to smoothing 25% to trace over the excel-derived profile line in order to create a more pleasing line (since my original profile had a point for every mile along the 4,000 mile route and created quite a jagged looking profile line.) Switching a profile graph’s direction to suit the data was a mental hurdle, overcome only by thinking, thinking, and more thinking about the data and the story I was trying to tell. This is something we’re always talking about: know your data, think about your story.

 

TYPEFACES The typefaces are chosen to complement the historic look of the map. Gabriola is used since there’s limited text and its decorative nature seems to suit the overall style. There are Myriad other fonts I probably could have used, and likely a better one will crop up as I revise. (The Myriad reference, ahem ahem, is a little type joke for the type nerds.)

 

X AXIS LABELS The x axis labels are problematic and will need to be redone. They depict distances at the segment end points, which is useful, but too jarring in the context of an x axis, which traditionally shows a smooth progression of numbers. An experiment gone awry. Sometimes, oftentimes, our experiments don’t work out. Note the axis lines and labels are in a medium-dark gray. I want them to remain firmly in the background as supporting information.

 

PROFILE LINE COLOR I considered changing the color of the profile line depending on the altitude. Green at low altitudes, progressing to brown, and finally white at the highest altitudes. A hypsometric tint for a profile chart! This seemed a bit much though, given the complexity of the rest of the page.

 

COLORS AND PRINTING The saturation of the hillshade will likely need to be adjusted once I see this printed on the final medium. It looks good on my screen but vast experience with this shows me that it can likely look too dark or too light depending on the press and the paper. Web designers have it easier in this respect. They’ll go on about how every monitor is different and that phones produce glare outside. But that’s nothing to needing a perfect print out on one media type and getting it wrong. There’s no excuses when it comes to printing a map in a known format (in this case, the map will be in a book so I will need to do a lot of exacting color proofing when the proofs come in.)

 

COASTLINE The overly stylistic coastline with drop-shadow was chosen specifically to comport with the historic style, though you’ll notice I didn’t interpret “historic” so literally so as to include such unfounded ornamentation as author cartouches or sea dragons. One must have boundaries. (Joke: boundaries, coastline, ahem.)

 

LET GO OF YOUR STYLING DARLINGS I am so hung-up on water=bright blue that when I began the map I of course made the water a very bold solid blue. It took several days of mulling over before I realized that was all wrong for this map. Pouring over the file of inspiration maps that I’ve been keeping for this project, it took only a few minutes to come up with an alternative to the blue, but only once I knew what I was looking for. This seems to be a frequent problem for cartographers who have a wide repertoire of map styles. They can intermingle to the great detriment of the final map unless time is taken to think things over. Knee-jerk styling such as my dark blue ocean (a modern map color for oceans) is only helpful if you keep to the same style-genre (e.g., historic, modern) all the time.

 

This wraps up my notes on the making of this map. Hopefully there will be a few bits you can use in your next project.

 

*The GIS&T Body of Knowledge is a great reference document when interviewing for a new GIS position. Use it as a check-list, checking off areas you have a firm understanding of and noting what subjects you need to study more.

 

 

No Comments

Swiss Topo Style, and risk in cartography

Posted by G.P. in Design on September 1, 2016

I haven’t written a post in a while so it seemed the best way to kick my butt into doing one was to lighten up and do a free-form post. Sometimes the biggest barrier to writing is the idea that it has to be perfect. The onerous task of revising and getting all the links and pictures just right takes so much time that it can seem daunting. So to heck with all that.

Earlier this week I posted something on twitter which, as usual with my twitter postings, probably made very little sense to those who read it. It went something like, “Everyone likes maps with hillshading, contour lines, and hachures. All on the same map.” And what I really meant was

I just read a tweet where someone said that these swiss topo style online maps that made the rounds recently should be formalized as best practice styling for online maps. In other words, a lot of people really like those maps. And I do too. But it made me wonder why everyone likes them. As someone who makes maps and who might aspire to actually making a popular map someday, it’d be good to note what everyone likes, correct?

Kind of like Cindy Brewer noting in her book that studies show that everyone likes blue and not a lot of people like yellow. We cartographers should pay attention to those things if we want our maps to be received in a credible light. The pure artists in the crowd will scoff at this of course. That guy making the huge (Michael Heizer) desert art installation says that art is only worth it when you’re taking a risk. Or maybe that art is only art when it’s risky.

I don’t know how to reconcile those two thoughts. As a cartographer do we want to be risky or do we want to use styles that we already know people will resonate with? Do you have to always go risky or always be conformal (little projection joke there)?

Someone read the tweet and is going to see if she can find a student to do a study on the swiss topo style to see if people do indeed like it better than other styles and if so, why. There’s some speculation that people might like the swiss topo style because the contour lines make them feel smart.

STFTW?

(Swiss Topo For The Win?) What do you think?

 So there we have it: a blog post written without a single edit session. Read at your own peril. Okay okay except the one revision where I added an actual title to the post.

__________________________________________________________________

 

 

No Comments

Raleigh Convention Center Coffee Shop Map

Posted by G.P. in Design on May 4, 2016

I’ve had a few requests for the pdf of the Raleigh convention center coffee shop map, otherwise known as the COFFEE4GNA map, that I created during yesterday’s QGIS demo talk at FOSS4GNA 2016 in Raleigh, North Carolina. Here it is!

Raleigh Convention Center, Coffee Nearby (PDF download)

coffeemapsnapshot

——————–

 

No Comments

The Making of a Map

Posted by G.P. in Design on April 4, 2016

I made this map a few months ago. At full size it’s 8.5″ by 11″ and meant as a handout. There’s a slightly different, but mostly matching, digital companion map at the bottom of the Hood Canal Coordinating Council’s Where We Work page that I made in conjunction with this. A discussion concerning different elements of the map follows.

WhereWeWork_HCCC

HIGHLIGHT MAIN LOCATION A typical task in this kind of cartography work is to highlight the main area. This is usually done in ArcMap with a masking layer or in QGIS with an inverse polygon symbolization. Either way, you are essentially assigning a semi-transparent white symbol style to the area that is outside the map focus. In this case we have 5 overlapping polygons that represent the focus of the map so the solution was to merge all 5 polygons together so that the masking layer consisted of everything not in an HCCC watershed jurisdiction.

OVERLAPPING LINES There are several approaches one can take for the display of overlapping lines. One is to depict the lowest lines in larger width sizes and another is to offset the lines so that they are side-by-side. Neither of these solutions were appropriate for this map since there are too many overlaps. After much trial and error I settled on a transparency for the highest lines. When all lines overlap the net effect is for the overlapped area to appear brown, which is indicated in the legend. Note that the legend entries are stacked in the same order that the lines are stacked in the map.

LOGO Although I dislike putting logos on maps as a general rule, there is often no way around it when the client requires it. In this case the map looks just fine with the HCCC’s single-color logo placed at bottom-left, which is the location at the bottom of the visual hierarchy.

TITLE The title has a subtle drop-shadow created in Inkscape by simply copying the text, nudging the lowest text down and to the right, and copying the grayish land color for the drop-shadow color. The main title text color matches the client’s logo exactly.

COLORS The orange boundary is the same color as other orange features on the client’s website. All the boundary colors match the digital map on the client’s website.

LABELS River labels on the main fish streams are extremely important for this map. They were done by hand in Inkscape and in many cases the individual letters were hand-nudged at 200% so that they would conform as well as possible with the sinuosity of the streams. The tribe labels match the color of the tribe polygons. In the online version of this map linked to at the beginning of the blog post, the smallest tribe (in terms of area) isn’t visible at the scale of the other features. The solution for that problem was to depict the tribes as points at the lower zooms and dynamically change them back into their representative polygons at the higher zooms.

SCALE BAR AND NORTH ARROW The scale bar is a custom, simple, graphic to keep the emphasis on the map and the legend. “Miles” is not capitalized, which further de-emphasizes itself and is in keeping with some more modern practices. The north arrow is likewise fairly simple. Both match the client’s logo color. 

TEXT All text except the river labels is Museo Sans. Including the “miles” label in the scale bar, the title, legend, town labels, county labels, and tribe labels. (The river labels are in Georgia since natural features are typically labeled in a serif font.) I chose Museo Sans because it’s friendly, highly legible, and fresh. I’m going to go ahead and say it again: even the scale bar text was changed to match the map font. Don’t settle for defaults! Make your typefaces cohesive. Okay, getting off my soapbox now…

Those are the main cartographic decisions that were made, each appearing to be an easy decision but all were considered very carefully with regard to the audience and the matching digital map. Possibly 20 different color combinations were considered, for example. In all, you might be surprised at the length of time it took to make this seemingly simple map. 😉

4 Comments

Too Much New Software, Too Many New Libraries, Even for the Nerds

Posted by G.P. in Cartography Profession, Design on July 28, 2015

minioncompute

I’ve been learning d3 and it’s been a fantastic learning experience. It’s a low-level javascript library for making interactive visualizations–including maps–inside SVGs for nifty client-side rendering effects of datasets that can be updated easily with new data. Incidentally, Scott Murray’s book Interactive Data Visualization for the Web is a fantastic first resource full of the most important bits to know and written in an accessible style.

So that’s been great, yes. But even the nerdiest among us have their limits to how much of their day, their week, or even their year, they can jam with brand-new material. And that leads me to my thesis, which is that when you are designing a new product or a new library or a new software, keep in mind that it isn’t just your computer laypersons who have to leap a mental hurdle to even begin working with your product, it’s also people who have just simply saturated their number of “new things” that they can handle that week or that month, even if they are supremely computer-savvy individuals who would really have very little trouble with your product.

In product design teams, the engineers and managers all have a tendency to think about the lowest common denominator when designing, and that’s certainly not a bad thing. But the computer scientists among us hit their limits from time to time too. So you’re not just in the business of making things simple for the newbies, your making your product easy to adopt by everyone.

My forays into d3 came at a time when I was granted a few hours of paid time to work on it, and that helped. It also came at a time when I was ready to really dive into something new. Don’t count on that being the case for all your users.

As usual I’ve reached the end of my little thesis statement with a feeling that it could be argued in the opposite direction as well (the downside of being analytical). So if you’d like to argue the opposite please go for it. And keep in mind the goal here: giving the best advice for teams building brand-new products as well as individuals who are building brand-new tools and libraries.

 


 

1 Comment

Recency Bias in Mapping, an Exploration

Posted by G.P. in Cartography Profession, Design on May 27, 2015

 

I’ve been thinking about recency bias lately. In finance, recency bias is the problem whereby people think that securities trends from the recent past will continue into the future. Instead of focusing on market fundamentals or buy-and-hold strategies, people tend to jump on bandwagons instead, which works until the cycle changes.

Recency bias also applies to many other aspects of human behavior, so how might it apply to map design? One way it applies is that, until the cartography landscape was disrupted around the late 2000s, people tended to make all their GIS maps very similarly. I say “GIS maps” because there were some people making well-designed maps with Illustrator and other design tools before that but the GIS people, for the most part, hadn’t caught on to those skills or those techniques.

Along came some new tools like TileMill (now MapBox), an influx of design-oriented mappers to the field, and new and interesting datasets to map, and suddenly the bar of map design was raised. I predicted this in the 1st edition of GIS Cartography in 2009. Those who were assuming that GIS map design would remain the same were creating increasingly obsolete visualizations.

 

Let’s show this by example.

Prior to the late-2000s disruption, we frequently encountered maps that appeared to be slapped together, often with a multitude of diagonal hashes making one feel rather dizzy like in the map shown below. These maps would contain some superfluous details to meet edge-cases (observe the grid-like layer in this map example) and would omit basemap information such that no underlying geographic context could be ascertained. Now, admittedly, these were not always the fault of the map maker. Indeed, the edge-case layers would have been included at the behest of some boss who was the only one who wanted to see that information. Similarly, the basemap information would be omitted due to a lack of availability of that kind of information.

 

aldc_spd_lrg

 

Thankfully we see many fewer of the types of maps shown above and many more of the maps shown below. In this map we have geographic context, gleaned from a nice and subtle baselayer of buildings and streets along with some very nicely done trees. There is some hashing but it comprises a small minority of the overall space. Interestingly, this map manages a significantly better visual experience with much more information. The richness of information makes it both more functional and more beautiful.

 

12546_KTCP_UD014-Framework-Generational-Plan-Overall-480x528-1402446397

 

What kind of recency bias in mapping do we see today? What kind of recency bias is affecting your map designs? Could an adherence to established practices be holding back your cartographic potential?

No Comments