Potential uses of font attributes in visualization

This is another post on the use of font-specific attributes and potential use in visualization.

Marking up Text: e.g. Text Skimming, Speed Reading, Parts of Speech

In some of my current research work, I am looking at using font-specific attributes to facilitate text skimming – i.e. the task where you don’t actually read all the text, you just skim it for key facts, particularly focusing on unusual words, etc. I had an example in a paper (specifically figure 2) at the Knowledge Mapping and Information Retrieval workshop earlier this year. Here’s another example, in this case, the first few opening paragraphs from Alice in Wonderland with successively heavier weights applied to successively less common words (based on English language word frequency usage):

Text formatted for skimming

A text formatted for skimming using visualization principles to make the least frequent words visually pop-out.


It is interesting to look at some other examples. In the domain of speed reading (where the objective is to read all the words), here’s a patent application for rapid serial presentation where the markup of the text for presentation may include bold, italics, underline, color, etc. or another patent using motion as an element in the presentation. However, we can go back a lot further than 2005 to find much older examples. Here’s an image from 1916 from a patent granted in 1923 using typographic attributes to markup parts of speech using all the techniques available to a printer in the 1920’s: italics, bold, underline, small caps, etc.:

In this particular patent, J.H. Sheffield proposes a process that encodes parts of speech with different font attributes, for example, independent verbs in all caps (“WAS DRAGGED”, “AND SHE SAID”), whereas dependent verbs are in small caps (e.g. “HAD DISCOVERED”, “HAD LAIN”), the object is bold (“remains”), etc.

More modern graphic designers are not bound by the strict limitations of metal type. Nick Feltron’s 2013 Annual Report pairs categoric labels with typographic lines, where the lines are almost, but not quite, typographic underlines:

Beyond Serif and Sans Serif

All the above examples vary type attributes within a single font family. What about varying font-family? Maps will often use serif and sans-serif to differentiate between features, e.g. water features labeled in serif italics while places may be labeled in sans serif italics:


Subset of map labeling approaches for different publications (from the book Thematic Maps: Their Design and Production).

Few maps go beyond serif/sans serif. Ordnance Survey maps sometimes use other fonts, for example, blackletter font to label historic sites, such as Tumuli and Fort:

Ernst Haeckel went beyond serif and sans-serif font-families to differentiate between categorical systems in his Pedigree of Mammals chart (on display at the British Library):

Haeckel's Pedigree of Mammals.

Haeckel’s Pedigree of Mammals.

And there are other examples of using different fonts for multilingual labels. E.g. See crier mongrel on

So, if one was trying to create a palette of different font families to identify different categories, one would want to create a set of fonts each with very different characteristics: non-typographers are unlikely to be able to differentiate between Helvetica and Univers (both highly popular sans serif fonts from the 1950s). To differentiate between serif and sans-serif, the sans serif should likely have pronounced serifs (such as a slab serif e.g. Rockwell) or a high-stress (such as Bodoni). Going beyond serif and sans-serif, one might also include: a blackletter, a simple cursive font, and a simple hand-letter font as being easily distinguishable. A fixed space font might be feasible to add to the mix and be differentiated from the others, although Courier (fixed line width and slab serifs) would need to be used in conjunction with a highly different serif, such as a high-stress serif. Here’s a possible palette of font-families for categoric encoding with similar weights:


One challenge is how many different fonts could/should be considered? Certainly there are a wide variety of decorative fonts that could be considered. Handbills from the 19th century and signs from the early 20th century mixed many fonts into a single composition, e.g. (click for original images):

Handbill-PalaceCircus Handbill-Notice Handbill-Swing

In many cases these specialized or decorative types were designed for headlines. One criteria for considering whether a particular font may be suitable is whether the font is suitable for large blocks of text – i.e. a font created for readability. Something to consider for further investigation.


About richardbrath

Richard is a long time visualization designer and researcher. Professionally, I am one of the partners of Uncharted Software Inc. I am also pursuing a part-time PhD in data visualization at LSBU. The opinions on this blog are related to my personal interests in data visualization, particularly around research interests related to my PhD work- this blog is about exploratory aspects of data visualization not proven principles.
This entry was posted in Font Visualization, Text Skimming, Text Visualization. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s