This is the website of Abulsme Noibatno Itramne (also known as Sam Minter). Comments here or emails to me at abulsme@abulsme.com are encouraged... or follow me on Twitter as @abulsme.

Categories

Calendar

September 2019
S M T W T F S
« Aug    
1234567
891011121314
15161718192021
22232425262728
2930  

Book: Envisioning Information

eicoverAuthor: Edward R. Tufte
Original Publication: 1990-05
Started: 2011-02-13
Finished: 2011-02-16
Format: Physical
126 pages / 4 days
31.5 pages/day

Envisioning Information is the second of the series of information design books from Edward Tufte. After the first book, I thought, yeah, he makes some good points, but he is a bit overhyped.

It has been a long time since I read the second book, but I don’t remember anything that changes that assessment. In addition though, it seemed that most of the good points were made in the first book and there was a lot of repetition in the second book.

More and different examples to be sure, but used to make the same general points over and over again. Some of the points Tufte makes (making sure what you present is clear and uncluttered by “chart junk” for instance) I think are great.

Others, such as where he pushes the idea of super high density information displays of information I think are not as well placed. SOMETIMES having those sorts of things that are jam packed with information that you can explore for hours is great. But MOST of the time, having something that as clearly as possible focuses in on the specific things you are looking at can be far more useful.

There is indeed a place for both of course.

Anyway, there are of course lots of nice examples of information design. Some Tufte calls out as wonderful, some he calls out as horrible. Sometimes I agreed with him, sometimes I did not.

And don’t get me wrong, a lot of this is fun to look at, and there is good discussion. I just felt I had probably gotten most of what I was going to get out of it in the first book. And of course, this isn’t the last of the series.

And now, the ranking of the last 10 books reviewed by reading speed (ones with kindle locations first, then ones with only page speeds):

  1. 695 – Shadow Puppets (F)
  2. 656 – Fatal System Error (NF)
  3. 647 – War of Gifts (F)
  4. 614 – Shadow of the Giant (F)
  5. 446 – First Meetings (F)
  6. 326 – All the President’s Men (NF)
  7. NA – [Physical 42.2 pages/day] Ender in Exile (F)
  8. NA – [Physical 31.5 pages/day] Envisioning Information (NF)
  9. NA – [Physical 25.8 pages/day] Nurtureshock (NF)
  10. NA – [Physical 9.08 pages/day] Agile Project Management with Scrum (NF)

And now the graphs:

% of the last 20 books I reviewed that are now available on Kindle:

kr201307220726

% of the last 20 books I read (including 27 I haven’t reviewed yet) that I actually read on Kindle:

rok201307220727

(I bought my Kindle when the first ratio hit 50%.  I will do these charts until each ratio gets to 90%.  We are close, but not quite there yet.)

The Visual Display of Quantitative Information: Second Edition

Author: Edward R. Tufte
Original Publication: 2001
Started: 2010 Oct 24
Finished : 2010 Oct 30
Format: Hardcover
197 pages / 7 days
28.1 pages / day

So, back to a physical book again.  This one was part of the spoils from going to Tufte’s Lecture back in June.  All the attendees got his four major books, of which this, The Visual Display of Quantitative Information is the first.  Since I went to the lecture for work, this is on my work non-fiction list.

Of course, I’d seen this book around and heard about it for many many years.  I remember talking to colleagues about this book way back when I first started doing web and internet related stuff in the 90’s.  That would have been the first edition of course.  What I read now is the second edition, but I gather the changes were more in terms of corrections and improvements to images and such, rather than fundamental changes to the content.

Tufte’s books been, and especially this one, have been held up essentially as bibles for good visual information design.  Having now seen his lecture and read the first of his books, my initial impression is that while he certainly makes a number of good points, especially when it comes to things that are routinely done badly and which should be avoided, he falls down a bit in terms of his positive suggestions for new sorts of presentations.  It seems that perhaps he is a little overhyped.  Of course, there is also the possibility that it only seems that way because many of his recommendations have become so widely regarded over the years that when you hear them now you think “Well, duh!  How could anyone think otherwise?” whereas when he first articulated his ideas they were a brand new perspective.  I think this is quite likely.

Of course, his major complaint is with information graphics that are full of what he calls “chart junk” or misrepresentations.  Basically decorative elements that add no information and often detract from understanding the data, or choices of graphical treatment that leave a casual viewer with an incorrect impression of the data.  He gives many examples of these sorts of bad usages.  The core message is to strip away the extraneous elements and concentrate on the data itself, and then be extremely careful to show that data in a way that highlights the important aspects of the data honestly.  This is excellent core advice.

He does get a little too sparse in some cases though, or recommends presentations that he himself says may stray into “puzzle” territory, where the person looking at the chart has to figure out what some of the new conventions mean.  I suppose some of this just has to do with being used to the standard conventions.  But in some cases I think it is just a case of taking good general principles a bit too far.  For instance, remove the frame and gridlines around many plots.  Now, certainly gridlines should be light enough to not interfere with looking at the data, but I like having them damn it, and the frame separates out the data from whatever is surrounding it.  Yes, I get they may not be completely necessary, but I actually think they help in many cases.

Oh, and of course Tufte introduced sparklines.  These can be useful in certain contexts, but I generally don’t like them.

There.  I got my curmudgeonly bit out.

In any case, this is a classic on these topics, and well worth reading.  Even if you don’t buy into every last bit of what Tufte is saying, the general principles are right on the money and worth considering whenever you are trying to present any quantitative data graphically.

Tufte Today!