All 4 books by Edward Tufte now in
paperback editions, $100 for all 4
Visual Display of Quantitative Information
Envisioning Information
Visual Explanations
Beautiful Evidence
Paper/printing = original clothbound books.
Only available through ET's Graphics Press:
catalog + shopping cart
Edward Tufte e-books
Immediate download to any computer
connected to the internet:
La Representación Visual de Información
Cuantitativa, (200 páginas) $12
Visual and Statistical Thinking, $2
The Cognitive Style of Powerpoint, $2
Seeing Around + Feynman Diagrams, $2
Data Analysis for Politics and Policy, $2
catalog + shopping cart
Edward Tufte one-day course,
Presenting Data and Information
San Jose, May 5, 6, 6
San Francisco, May 7, 8, 9
Bethesda, June 3
Arlington, June 5, 6
Portland, August 4
Seattle, August 5, 6
Sidenotes or footnotes or what?

Does anyone know how to make MS Word create sidenotes? I & my L.A. are quite stumped by this.

-- John Schedler (email)

Response to Sidenotes v footnotes

There's an Ask E.T. thread on this at:

http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0000QS&topic_id=1 .

Good Luck.

-- Steve Sprague (email)

Response to Sidenotes v footnotes

If a million monkeys type for a million years at a million keyboards. . . .

This is what page layout programs--Adobe InDesign and Quark XPress-- are for.

Or perhaps there are some mid-level page layout programs--InDesign Lite or QuarkXPress Lite?

-- Edward Tufte

Response to Sidenotes v footnotes

On this issue more generally, take a look at the layout of Richard Feynman, Robert Leighton, and Matthew Sands, The Feynman Lectures on Physics.

One page of that book is reproduced (rather small) in "The Cognitive Style of PowerPoint." Here it is somewhat larger:

The problem fundamentally arises from the 8.5 inch page width. That is too wide for a single column of type; thus 2 columns of text are often used in books with an 8.5 inch page width. In general, a line of text should not be more than 2 or 3 alphabets long, unless there is spacious leading. (An unsatisfactory solution is to make the type real big to fill the horizontal width.)

So what should be done with that extra space on the right of the page when a single column of text is used? The design of my books and of Feynman's book puts small images, image credit lines, numbered notes, and references out in a narrow righthand column. For my books, the righthand margin column also often serves to help present big images that run the width of the entire page. The design of Feynman's pages are probably a better model than my books, which are so intense, idiosyncratic, and personal--and are therefore not a good workaday model for document design. Long ago I once proposed to a client that computer manuals (with 8.5 by 11 inch paper imprisoned in 3-ring binders) follow the design of Feynman's physics textbook and stop wasting all that paper in the margin.

Of course, the big-column/little-column format should not be done mechanically; there needs to be worthwhile material that naturally belongs in the margin. If the marginal materials are simply references, then the standard footnotes (at the bottom of each page, not ganged together at the end of the document) are fine. Maybe the empty space in the margin can be left for marginal notes written by the readers, if any.

Many, although not all, of the legal documents I have seen (at The Smoking Gun, for example) have too few words per page with over-generous leading and even double-spaced lines. Given frequent repetition and boilerplate, legal documents sometimes appear bulked up and puffy, all that paper for not that much substance per page. This leads to the appearance that productivity is being measured by the page. The chief of a large group producing computer manuals once told me that they "produced 46,000 pages of manuals" that year. "What a shame," I replied, thinking that one-third that many would probably be better for the readers although much harder to write.

-- Edward Tufte

Response to Sidenotes v footnotes

" ... Aren't there middle-level page layout programs? QuarkXPress Lite? InDesign Lite? ... "

Take a look at Adobe's Pagemaker, which you might call InDesign Lite. It's marketed as a business document (brochure, newsletter, flyer) application but it has some serious layout and professional publishing features and handles type really well. All for a base price a couple of hundred less than InDesign, about $500.

-- Steve Sprague (email)

Response to Sidenotes v footnotes

On the issue of double-spacing in legal documents: A big part of the problem is that double-spacing is often required by court rules. To make matters worse, most of the federal courts of appeals require that if you use a proportionately-spaced font, you have to set it at 14 points. Talk about too few words on the page.

-- Neal Goldfarb (email)

Response to Sidenotes v footnotes

Double-spaced lines give the reader room for word-by-word notes (much more so than wide margins). Some scientific journals require submissions to be double spaced for exactly that reason. Maybe the rules are designed to let people add plenty of their own comments to documents?

-- Martin Wattenberg

Response to Sidenotes v footnotes

Max DePree's wonderful book Leadership Is An Art is double-spaced, specifically, as he says in his preface "In some sense, every reader "finishes" every book according to his or her experiences and needs and beliefs and potential. That is the way you can really own a book. Buying books is easy; owning them is not. There is space for you to finish and own this book...Intent and involved readers often write in the margins and between the lines. (You may end up doing a lot of writing and reading between the lines in this book!) Good readers take possession of what they are learning by underlining and commenting and questioning. In this manner they "finish" what they read." It is an interesting and compelling reason for leaving room in books for the owners to use.

-- Ann Stone (email)

Response to Sidenotes v footnotes

While I agree that making notes in a book is one way to `own' a book ( I think `finishing' a book for the author is a little presumptuous), I think that double spacing is a poor way of allowing the reader to make notes. Double line spacing is useful for copyeditors that need to correct grammar or typography but it is not useful for much else because it interferes with the readability of the text. Any typographer will tell you that correct linespacing improves the readability of text and aids reading of extended text. Notes between the lines of text interrupt the authors narrative and distract the reader from what the <author> has to say. Decent margins are more useful for notes and allow comparison between the author's and reader's messages. It's best to leave doublespacing where it belongs, in typewritten manuscripts submitted to publishers.

As to the main topic, How to make Word make sidenotes, I would agree with others that Word is the wrong tool for the job. A page layout package will do what you want with a minimum of fuss -- try InDesign or Quark or LaTeX (Convert your existing files in Word with Word2TeX)

-- John Walker (email)

Response to Sidenotes v footnotes

These are not new issues. Here's how they handled sidenotes in the 13th century, from a commentary on Euclid produced in Paris ca. 1266 now at Oxford's Corpus Christi College.

And here's a bit of superb page layout and typography in a text book, from a 13th century volume of Aristotelan texts produced in Paris ca. 1260 now in the Bibliotheque Nationale, Paris. The caption for this image reads "The manuscript was made with wide margins for glossing in the classroom."

Sure wish I could read Latin. Both found in Christopher De Hamel's "A History of Illuminated Manuscripts", Phaidon 1986, 1994.

-- Steve Sprague (email)

Response to Sidenotes v footnotes

Layout of David Foster Wallace fiction in Atlantic Monthly

I haven't found any posts on this site about the layout of David Foster Wallace's short story in the April 2005 Atlantic Monthly. Wallace likes to use a lot of footnotes in his fiction, and this story uses an interesting layout scheme of color-coded sidenotes rather than footnotes. On the regular website, you can't see the effect as clearly because the notes appear as pop-ups (http://www.theatlantic.com/doc/200504/wallace), but you can download a .pdf of the story if you are a subsciber.

I'm interested in hearing what you think.

-- John Nugent (email)

Response to Sidenotes v footnotes

Perhaps it is my conservative tastes, but I find the
technicolor tags get in the way of the text.

-- Tchad (email)

Response to Sidenotes v footnotes

I know this is a silly-sounding question, but how do you read footnotes, sidenotes, chapter notes at the end of a book?

I find that pausing my reading of the main text to persue the notes interrupts my reading rhythm. If I wait until I've finished a section or chapter, then go back to read the notes, I find that I've lost context and am not sure what I'm reading. So the question seems valid: How do you read sidenotes? Or: How do you productively read sidenotes?


-- Kent Karnofski (email)

Response to Sidenotes v footnotes

Put simply, this is an acquired skill. The labor-intensive answer is to practice: read a lot of texts that have footnotes or endnotes. If you do a lot of this, it will become clear to what extent you need to dive into the notes. This will vary between books (some texts rely on notes far more than others; some notes are rich, others dry) and it will depend on what you're trying to accomplish in your reading.

It took me a perhaps a year to be able to read through the citations, references, and footnotes in scientific papers when I first started reading the literature as an undergraduate. Now I read this format and it is transparent, just as a ledger is transparent to an accountant, an aircraft's instrument cluster is transparent to a pilot, or the Daily Racing Form is to a horse racing devotee.

One thing worth noting: notes can be used to illuminate the text, or to bury things that the author would prefer not to deal with. Examination of notes can tell you a great deal about the author's agenda and approach.

-- Alex Merz (email)

Response to Sidenotes v footnotes

I find that the context switch of moving between the text and the side or footnotes is easy if you expect to only read the notes you'll find interesting.

Much like reading modern hypertext, you follow the information that kindles some interest in you and skim or ignore the rest. A reference to a book with the originating material may not get a glance unless you recognize you have that volume on your shelf or make a mental note to pick up a copy, but something like a commentary on an inside joke may momentarily distract you from the flow.

I think the most important reason to consider when creating sidenotes is whether or not it's needed in the first place. Running commentary on another's work would seem to be the most oft-needed use, but in most cases for textual sidenotes, if the writer says what he or she means, does it need further explanation? As literature and as art, there lies the possibility that it's a stylistic choice.

As I read Steinbeck's The Log from the Sea of Cortez, a book where it would be very natural to throw in sketches of coves or lagoons or unique specimen, I'm fascinated that its only graphic addition is a map of the region—and that may be the fault of the publisher. Footnotes are used (I believe) exclusively to translate common names to the taxonomic, and as a reader I find myself not missing glossy images of the various creatures taken that would be so prevalent in any modern text.

In any text I would judge its sidenotes based on whether they enrich the body or detract from it. If they detract from the text, it's not worth the context switch and the flow is unduly interrupted. If they enrich the text, the sidenotes should act as part of the flow, like a lesser channel in an estuary that still ultimately leads to the sea.

-- Jack Johnson (email)

In scholarly work, the main purpose of footnotes or sidenotes is to acknowledge and credit sources, and to refer to other relevant work. This is a moral obligation in scholarship.

But footnotes should not be overdone or pretentious, as is often the case in PhD dissertations. Also don't build in little all-too-knowing commentaries into the references, a graduate student disease, as in "the seminal article by G. P. K. Sitzplotz, [reference]." Sometimes acknowledgments can be put into the text, although vague insider eponymic references (a "Downsian model," a "Derridaist reading") that as Gore Vidal once said "cause the reader's brow to furrow," should be avoided. (Note the runnng reference to Vidal in the text here.)

In general, using sidenotes or footnotes to conduct a continuous running commentary underneath the main text is probably not a good idea. If the material is important enough to discuss, then it may well belong in the main text.

That is the general theory. Personal style is another and more complicated matter, and what follows is my peculiar view, which probably has little relevance to other writers.

Quite often I use sidenotes as detailed annotation or as a sidebar extension or a reinforcing detail to support the main text and, when doing so, try to include small images along with words--especially where such material would break up the main text if included there (better to interrupt the flow with a sidenote). I believe that such material should be almost entirely content-driven, rather design-driven--although at times, when I see a big empty sidenote column, I see a tempting opportunity to fill the space.

Behind each published example in my books, there are often 5 or 10 similar competing examples in the files, and illustrated sidenotes offer an opportunity to show some of those examples. Over the years, I've deliberately sought to intensify the pages of my books (compared to the first book, The Visual Display of Quantitative Information, which has rather spacious and open pages). This is, in part, because of the increasing emphasis on high-resolution displays of information (beginning with the second book, Envisioning Information), something I seek to exemplify in the page layouts. Also Envisioning is a very intense book with very intense writing and illustrations where I worked from a theory of "bright white writing" (no redundancy, no first person, every word and every illustration counts). In the last year of writing a book, there is usually a big clean up and cutting, as entire spreads, examples, and sidenotes are deleted. Near the end of Beautiful Evidence, I deliberately brought in examples on open nearly sidenote-free spacious pages (the Verrocchio horse, the Cezanne painting, the Galileo annotated drawing) in order to relax the scale of the book, vary the texture of the pages, and create a better architectural rhythm of page densities. Of course these were also powerful examples, so the choices were also driven by content considerations as well as by thoughts about the overall architecture of the book.

I also seek to optimize each double-page layout, sacrificing global consistency (or a global theory about sidenote use) across all the layouts; this is sometimes called "breaking the grid."

Of course, when it takes many years to write a book, things also tend to accumulate on the pages.

-- EdwardTufte

I've looked in vain in several typesetting programmes for the ability to set "live" footnotes in 2 columns below single-column text, multiple "live" footnotes on a single line, or "live" sidenotes. Perhaps those who have explored sidenotes know of such programmes?

The desire for the first-mentioned layout has been smoldering for some time, and was reignited today in looking at a colleague's 18th-century volume of poetry with the copious short notes set in 2 columns.

-- Ken Blackwell (email)

If you are looking for an in-depth discussion about the use of sidenotes and how to make them beautiful Robert Bringhurst wrote about them in his book Elements of Typographic Style. In the second edition look on pages 68,161,176.

-- Father Peter Mussett (email)

I'm amazed that nobody has referenced the most successful scholarly use of sidenotes - the Talmud. See this explanation and you may also be interested in my "talmudic-style" test on the CS literature here.

-- victor yodaiken (email)

Sidenotes are best when the reader can tell, out of the corner of her eye, whether they are worth examining. One of the many wonderful things about Tufte's books is, citation sidenotes are short and it is clear by the typeface and format that they are citations--"look at me if you want more information". If you don't need to know, say, in which issue of USA Today some horrendous graphic appeared, just keep reading.

On the other hand, graphical sidenotes in Tufte's books demand attention, and in most cases Tufte refers to them in the main text itself, adjacent to the graphic.

Sidenotes are much better than footnotes because you never know what the footnote holds. The bottom of the page is not in the corner of your eye, so you have to leave the main text, jump to the bottom of the page, look at the footnote text (invariably in a different font, requiring more visual re-adjusting) read a few lines, get annoyed by the pedantics, then, flustered, fumble back to the main text to re-establish your place. Footnotes are one step better than endnotes, but only one step.

-- Eric Howell (email)

I'm curious about Dr. Tufte's choice of sidenotes always on the right-hand side of the page.

I recently have been having non-PP meetings with hand-out styled after the books. I have found a 2-page spread effective, with side-notes on the "outside" margin. The left hand page has the notes on the left hand edge, and right hand page has the notes in the right hand margin. This choice seems natural for a 2 page spread, but it's not the one that Dr. Tufte made.

BTW, my non-PP meetings have gone well so far.

-- Ted Kostek (email)

Threads relevant to typography, book design, and publishing: