24 replies to this topic

[ArcMapper]

I was criticized for using the term 'pretty' maps. I definitely see the issue with this, as there is a lot of information, accuracy, science, etc that actually goes into a map. I use the term loosely, mainly to refer to what the final product should look like. I'll try to stay away from using the phrase though if it's really that offensive.

[David Medeiros]

I was criticized for using the term 'pretty' maps. I definitely see the issue with this, as there is a lot of information, accuracy, science, etc that actually goes into a map. I use the term loosely, mainly to refer to what the final product should look like. I'll try to stay away from using the phrase though if it's really that offensive.

I mentioned it in the "cost of living" thread. It's not so much about being offensive, though it is tinged with a bit of disrespect or disregard shall we say for the history and tradition of cartography as a profession. One that not only pre dates GIS by a long shot but has shaped its very foundation. It's a phrase that dismisses any notion of depth, data integrity or analytical power behind a "simple" map as opposed to one that's "dynamic". This is a long a windy road I'm stepping on here so I'll bring it back from the US and THEM arguments this always foments.

My real objective in bringing it to your attention was to say that aside form possibly hurting our feelings it actually hurts your work to think of cartography in such simple terms, even if what you are saying is that you'd like your final map to "look really nice". If you keep cartographic best practices bottled up in the notion of "pretty maps" than you won't be able to make the connection between GIS based spatial displays, maps and information design. Best practices and design matter more than you think in terms of communicating information. In simple terms how we say it can be almost as important as what we say. I think that "design matters" ethic is what cartography has to offer GIS but that I see GIS practitioners shy away from because they often have this baseless notion that cartography is an art concerned primarily with looks and not data.

[Mapgeek]

I think that I would have to disagree with using the term "pretty" to describe a map as being disrepectful. (Then again, that may be that I am more of an analyst than a cartographer. "Us v Them" apparently.)

I actually just gave a presenation titled "How to make a pretty map out of ugly data". I think it summarized my experience making the map quite well. I was trying to make a regional map of land use plans from 28 cities, towns and villages plus add the layers that were important to the county. That was a lot of data. That was a lot of different ideas on how things should be done trying to be expressed in one map. The process of combining the data was ugly. It was hair-pulling, agonizingly difficult. It took over 2 years (a lot of time was spent waiting for the data, most of which showe dup in the last couple of months.) In the end, I had a map that was pleasing to look at, aka it was pretty. The presentation talked about how I obtained the data, generalized it and integrated it. I had 11 categories of land use (some of which were parcel based and some of which were natural feature based), 5 overlay layers (none of which could be a transparency), plus the requisite base layer data. When my boss (a layman) asked me how the map was going the only way I could explain it to him that he could understand was, "It's hard to make a pretty map out of ugly data." He got that. He understood it then. I should also mention that at the last session on the last day and right before lunch, I had a full house. So, it got someone's attention.

It's kind of like when someone who doesn't know anything about GIS asks me what I do for a living. If I say "I am a GIS Manager", I get a blank stare. If I say "I make maps" (even though I cringe a little on the inside) they understand a little bit about what I do. Depending on the individual, their interest and the time alloted, I may expound on it and talk about the research, data acquisition and myriad of other more involved things that I do. Other than that, I drop it.

[Hans van der Maarel]

I'd say it's easy to make a pretty map out of bad data, but it's probably not going to be a good map.

[l.jegou]

IMHO, trying to make a pretty map is a righteous ambition, but only if it's after succeeding to make a real map, a map that achieve its objectives.

But trying to make a pretty map is the sign that you are trying to adapt to the public of the map, and that you are especially careful about the perception of the map, by refining its esthetic qualities.

The question seems to be more visible in the field of "Infographics", ie. illustration graphics for the news, with several competitions and showcases, for example here : http://infographicsnews.blogspot.com

Is this illustration a map, a pretty map, a real map ?

It's a very important discussion to me, as it's a large chunk of my thesis dissertation, and i'm curious of the various opinions of fellows cartographers

(I refer the scholars among you to the book "Rethinking maps", Edited by Martin Dodge, Rob Kitchin, Chris Perkins, one of the latest reflection on the concepts of cartography, in english).

[DaveB]

I think some people bristle at the term "pretty maps" because they've experienced that as a devaluing, monetarily and otherwise, of the their work, the cartographic process and the product.

I definitely believe there is value in making a map attractive. For one thing, people are more likely to want to spend time looking at a map that looks good. But I also believe it's similar to ideas like "elegance" in mathematics. If it looks good chances are it communicates better (but not always). I think it's hard to separate good aesthetic design from good functional design because I think designing (and building) things to make them function well also tends to make them more aesthetically pleasing as well. As I said, there are always exceptions. Then again, maybe that's just prejudice on my part. I prefer things that are more aesthetically pleasing.

Picking up on what Hans said - Ugly data does not necessarily equal bad data, I would say. Maybe the data was not organized well, maybe it was collected at too large of a scale for the map, etc. Whatever it is is may be the data mat need work before it's good for a particular map or maybe the cartographer needs to do some generalization/selection of features, etc., to make a good map with the data available. An example of bad data might be a road that doesn't exist or is in the wrong position. An example or ugly data might be an accurate road that is broken up into lots of segments.

[David Medeiros]

A lot of what bothers me about the term can be seen in Mapgeeks reply (not to pick on you). You seem to be relaying the message that you consider the collection, preparation, and basic layout of your data to be a GIS function in the work you do and the final step of making it look good (making it "pretty") to be cartography. I consider all of that to be cartography. Making a pretty map out of ugly data reinforces the perception that most GIS professionals I have worked with or studied under seem to have - that a good looking, aesthetically clear map is just dressed up usually for public consumption (presumably a public who couldn't possibly understand the "real" map data).

I think this attitude shows a real lack of knowledge of the role Geography and Cartography play in GIS.

And why would you cringe at the notion of describing your work as making maps? After all, it is. It's not a full description but underneath that act of "making a map" is everything that geography and GIS contain. Making a map is knowing the data, gathering the data, analyzing the data and communicating the data. There's a lot more to making the data communicate effectively and look pleasing than pretty color choices.

Even though bothers me as much as it does (bristle is a nice term DaveB) I'm also a little happy to see the GIS community (of which I am firmly a part) continue to make this mistake, it means there is always room for people like me to make their work better.

[Charles Syrett]

I tend to bristle over the expression "pretty maps", for two main reasons:

1. As Dave says, this expression is usually used by people in a deliberately derogatory manner, to belittle the efforts of some professionals to present data in a way that communicates well.

2. When I was trained in cartography (which was long before GIS existed), there was a caution against adding unnecessary aesthetic frills that may detract from the purpose of the map. In other words, the notion that cartographers make "pretty maps" displays an ignorance of what cartography is all about.

That being said! -- there actually are lots of situations in which it's entirely appropriate to deliberately add aesthetic "extras" to a map. If any part of the purpose of the map is served by adding extra attractiveness, then sound cartographic principles are being followed. Obvious examples include faux-antique style maps for books, and tourist maps that are designed to attract travelers.

Apart from such special cases, maps that are well designed will probably be attractive anyway. At least to other cartographers!

Charles Syrett
Map Graphics
http://www.mapgraphics.com

I think some people bristle at the term "pretty maps" because they've experienced that as a devaluing, monetarily and otherwise, of the their work, the cartographic process and the product.

I definitely believe there is value in making a map attractive. For one thing, people are more likely to want to spend time looking at a map that looks good. But I also believe it's similar to ideas like "elegance" in mathematics. If it looks good chances are it communicates better (but not always). I think it's hard to separate good aesthetic design from good functional design because I think designing (and building) things to make them function well also tends to make them more aesthetically pleasing as well. As I said, there are always exceptions. Then again, maybe that's just prejudice on my part. I prefer things that are more aesthetically pleasing.

Picking up on what Hans said - Ugly data does not necessarily equal bad data, I would say. Maybe the data was not organized well, maybe it was collected at too large of a scale for the map, etc. Whatever it is is may be the data mat need work before it's good for a particular map or maybe the cartographer needs to do some generalization/selection of features, etc., to make a good map with the data available. An example of bad data might be a road that doesn't exist or is in the wrong position. An example or ugly data might be an accurate road that is broken up into lots of segments.

[David Medeiros]

IMHO, trying to make a pretty map is a righteous ambition, but only if it's after succeeding to make a real map, a map that achieve its objectives.

This goes along with my post above quite well. What is the difference in your view between a pretty map and a real map?

To me there is no difference, a maps aesthetic beauty comes at least in part if not in whole from it's proper construction and the data itself. There are techniques to enhance it's clarity and readability, but not strictly its beauty. I see my role as a cartographer a lot closer to that of a geographer doing information design and communication. When I'm making a map it has a message, it has data driven content and it may display the results of some analytics I haver performed. When they are put together with eye towards clear communication and and understanding of "whats the message, who is the audience" etc. then you will probably have something that looks nice, is aesthetically pleasing.

As Edward Tufte puts it: "The commonality between science and art is in trying to see profoundly - to develop strategies for seeing and showing."

[natcase]

I think the problem is not with the explicit but the implicit definition of "pretty." It is a diminutive term, like "cute." Think of the difference between a "pretty" painting and a "beautiful" painting. Pretty pictures do not inspire, amaze, inform, or excite. Beauty has that potential, and more. Prettiness soothes and causes pleasure. Pretty pictures make one feel happy about the status quo, or long for the conflict-free status quo within the picture frame. Pretty is the visual equivalent of "nice."

"Attractive" has the same problem: it's about superficial pleasure: it's aim is to "attract" viewers, whether with real riches or fool's gold. Fool's gold being easier to manufacture, an "attractive" map is often a Potemkin village, signifying less than its attractiveness implies. I think this is where the sneer from data people about "pretty" or "attractive" comes from.

I'd suggest when someone suggests you make a map "pretty," rather than correct or argue with them, agree that your job includes making a map beautiful. If that's what they meant, there is no argument. If they honestly feel that looking good is dishonest (and I've been in that trap myself any number of times in my life), this comment will suss that out.

[Mike H]

This seems like a semantic argument, at best. I go out of my way to make all my maps pretty; it in no way undermines accuracy, data issues or attempts to dazzle the reader with graphics. I hope it simply makes the map more enjoyable for my audience, who is largely unaware and blissfully ignorant of accuracy and data issues: they should just trust my map, and enjoy reading it.

[natcase]

This seems like a semantic argument, at best. I go out of my way to make all my maps pretty; it in no way undermines accuracy, data issues or attempts to dazzle the reader with graphics. I hope it simply makes the map more enjoyable for my audience, who is largely unaware and blissfully ignorant of accuracy and data issues: they should just trust my map, and enjoy reading it.

Mike: semantic maybe, but I think the use of the term "pretty" (as opposed to "elegant" or "beautiful" or some other term meaning the work of the cartographer has some meat to it) is a problem. Not because "prettiness" is a bad thing (well, I guess I did more-or-less say that that in my last post), but because it implies that what we do in making a map work visually is superficial. It means the important part of the map is what it says, not how it says it. Which in turn means the cartography as communication is not of the same level of importance as raw map data. I think the deeper implication is then that human interaction around the map and its subject matter is less important than that the data itself.

Is that true? I think hard-core objectivists would say that that is in fact the case. That clear unequivocal statements of what is, are what maps are and ought to be about. Then the hard-core subjectivists will argue that there is no there there: it's all a construction... Then we get into shouting matches at NACIS and everyone goes home until next year.

And I maintain that the specific word "pretty" is part of that basically useless argument. Both sides have half the truth: there is a physical world out there, and we do construct systems to organize and describe it. That work of organizing often includes making it make sense as a picture in a conventional sense, so that it speaks in a language familiar to our audience, so we can be understood without affecting such a thick accent.

I think your maps are elegant, interesting, beautiful even, Mike. If they are "pretty", well, that's fine. But it's not why I like them.

[Esther Mandeno]

I go away for a little while, and you all start a most interesting topic...

Why are 'pretty' maps bad? I don't know, but I can tell you, as a GIS consultant, I have had several clients say something like the following (in that oh so condescending voice):

"I don't want pretty maps. I want good science."

I'm serious. This coming from Fire Planners/Managers/Scientists, Hydrogeologist, Engineers, etc.

They did not want a cartographer and an illustrator. They want a GIS scientist. So much so that when they hired me, they would say something like the above.

The first time I heard it (like, my first job as a consultant back in 2002), I thought: Yikes, buddy. That's a bit harsh. Do you mean to say that all those wonderfully pretty National Geographic maps don't have some serious science behind 'em?

I never said that to them, of course. But it made me shy away from some specific business practices for awhile, even though I always "prettify" my maps. It's part of the process. I do a lot of fire behavior analysis and displaying that data is never easy (for experts and the public).

So, though Nat Case would like to say that the argument is useless, that despite whether it's 'pretty or not' we need to organize data into a presentable manner, there is a population of experts out there that DO think data is more important than presentation. Folks that pay me lots of cash to make sure of that.

Personally, I think presentation is key and it is an integral part of GIS and cartography, but just how to convince those that do not think that?

Am I belaboring the topic? Should we just ignore it and move on?

[David Medeiros]

I go away for a little while, and you all start a most interesting topic...

Why are 'pretty' maps bad? I don't know, but I can tell you, as a GIS consultant, I have had several clients say something like the following (in that oh so condescending voice):

"I don't want pretty maps. I want good science."

I'm serious. This coming from Fire Planners/Managers/Scientists, Hydrogeologist, Engineers, etc.

They did not want a cartographer and an illustrator. They want a GIS scientist. So much so that when they hired me, they would say something like the above.

The first time I heard it (like, my first job as a consultant back in 2002), I thought: Yikes, buddy. That's a bit harsh. Do you mean to say that all those wonderfully pretty National Geographic maps don't have some serious science behind 'em?

I never said that to them, of course. But it made me shy away from some specific business practices for awhile, even though I always "prettify" my maps. It's part of the process. I do a lot of fire behavior analysis and displaying that data is never easy (for experts and the public).

So, though Nat Case would like to say that the argument is useless, that despite whether it's 'pretty or not' we need to organize data into a presentable manner, there is a population of experts out there that DO think data is more important than presentation. Folks that pay me lots of cash to make sure of that.

Personally, I think presentation is key and it is an integral part of GIS and cartography, but just how to convince those that do not think that?

Am I belaboring the topic? Should we just ignore it and move on?

You certainly shouldn't ignore it and move on, I think this topic actually strikes at a fundamental issue within our industry, and here I'm talking about geo-spatial visualizations. Where cartography, GIS, geography and information design intersect. Whatever part of that spectrum you're on, if your map making is data intensive (as most maps are regardless of their connection to a database), the difference between the perception of pretty and factual or accurate is one of visual communication not uncessecary augmentation.

The separation your clients perceive between "pretty maps" and sciecne is exactlty why we need to be talking about this. The problem exists on several levels: One from users who don't understand there is a science to information display and that cartography is in part the application of that sicence to geo spatial data. On another level there is a turf war going on here between some GIS users and more traditional cartographers, even as those lines continue to blur. The stratification of GIS into NeoGeo and PlaeoGeo further exacerbates the distance from modern GIS and it's roots in geography.

My experience has been that when explaining that what makes a map look better actually makes it work better as well (through visual hierarchy, symbolization, generalization, design to message and audience), the recipient of that information gets it. It's obvious when you start talking about maps as a form of communication that clear and concise always does a better job of it than a big visual data dump. Working towards that goal of clear and concise is what takes a map from "ugly" to "pretty". It's science people!

I'll also mention here that I have often observed a clear difference in how mappers with a geography and cartography background view this issue from those with only a GIS background. The carto/geo camp is by far more often inclusive of GIS, it's science, uses and principles than the GIS group has been of cartography or geography. There is an active exclusion of cartographic principles in many GIS camps to the point of dismissal or derision of our work and that comes out in the often ubiquitous term "pretty maps" as used by many GIS professionals today to discount the importance of the visual aspect to their work.

[Laura Miles]

I recall being instructed once to make a map "more cartographic". Ignoring the fact that logically that does not make sense (I know she meant to ask me to make it more aesthetically pleasing), it really rubbed me the wrong way but I didn't quite know why. I now realize it was touching on this issue that I hadn't really been aware of. Unfortunately since many GIS technicians lack cartographic training, I guess there's this tendency to believe it is concerned only with aesthetic issues. Really it is revelatory of a gaping hole in GIS education!*

*I'm not trying to be derogatory towards those with a GIS education...in my GIS program we had one course on cartography, therefore everything else must be "GIS" whereas this one course which focused on text placement and the use of color was "cartography". Luckily for me I had been around a bit beforehand and knew this wasn't all there was to it...and I found Cartotalk!