Responses on Optimum Page Size Question

[GFHayhoe -at- AOL -dot- COM]

About a week ago, I posted a question from Jim Giordano regarding research or
theory related to page size. Herewith the responses that Jim has collected
and asked me to post.

^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^
^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^

You bring up an interesting question. Assuming by size you are referring to
dimensions (i.e., page size) rather than length, let me try to address some
of the issues you bring up:

1) Industry standard

As you wisely noted, there is none. There are two reasons for this. First,
the size of the manual may vary based on what it's used for -- does the book
need to fit in a pocket? Or can it be spread out all over a large desk? Or
something in between?

As far as other usability factors, there is no empirical research to show
that any of the various page sizes has any inherent usability benefits, as
long as the book stays within reasonable size dimensions (i.e., it fits on
most bookshelves). What affects usability is how you use the page, what you
put on it. In other words, as us good old boys know, it's not the size, but
how you use it.

If you're hard-core about trying to empiricize (is that a word?) things, you
have to look at readability studies, which suggest that margins should be
10-15% of the page width (i.e., with a 6-inch-wide page, you should have
0.6" to 0.9" wide left and right margins). And then there are the studies
that recommend that line length not be longer than 2-1/2 to 3 "alphabets,"
where an "alphabet" is the lower-case alphabet typed out once. This will
vary pretty significantly depending on the typeface and pointsize of your
font. So you could measure 3 alphabets, add 20-30% for margins, and get your
page width that way.
2) Paper availability

What is far more critical for most page size decisions is the size of the
paper the printer uses. You probably already know this, but most page size
decisions are constrained by the size of the paper the printer will be
using. As you know, pages are ganged up on a mechanical, then folded and
trimmed, so the page size should be optimized so that it is almost an exact
1/4, or 1/8, or 1/16 of the size of the sheets of paper the printer will
use.

War Story: at one point many, many years ago, a graphic artist here at
Microsoft called for a page size that was just exactly a hair too big in one
dimension, so it cost literally twice as much to produce a manual because
each sheet of paper could only hold 4 pages instead of 8.

Although there is a lot of standardization, still, the size of paper varies
based on the type of paper, the size of paper a particular printer can
handle varies depending on the kind of press(es) he has and the kind of
printing you're doing, even your binding choice (perfect-bound books can be
printed on a different kind of press which can handle a different kind of
paper.

War Story: in an effort to reduce costs, the first version of Word for
Windows was going to be printed perfect-bound instead of three-ring binder.

Then they chose case-bound instead of perfect-bound because it didn't cost
that much more and was more durable. I don't remember all the details now,
but somehow the designer didn't account for the fact that a perfect (or
case-) bound manual can be sent through a higher-speed press, which can use
a different kind of paper (which is also a different size), and ended up
calling for a type of paper that was way more expensive than need be, and
the pages were ganged up in an inefficient manner, so it ended up taking
longer and costing as much as if they had just stuck with three-ring
binders.

In other words, a major factor is gonna be negotiating with your printer(s).
Of course, most of 'em will tell you that they can accommodate whatever you
want, and there is a lot of truth in that, but you're better off with a
printer who will work with you by giving you options and advice.

3) Microsoft does it that way.

In some cases, this is the right reason. When it comes to manuals, it is
probably a bad idea. To explain why, I'll go back to an old story:

A husband was watching his new bride prepare dinner, and asks her why she
cut the ends off the ham before she put it in the oven. She tells him she
doesn't know; that's just the way her mother used to do it, so she does it,
too. When the man asks his wife's mother, she makes the same reply. Finally,
he calls his wife's grandmother and asks her. She says that she always cut
the ends off the ham because she never had a pan large enough to hold a
whole ham.

For many, many years, every group of manual writers at Microsoft had a
full-time graphic designer. This resulted in a lot of graphic designers
without a whole lot of justifiable work, so they spent their time
redesigning the page for every version of every product. The result was
that, even though we shipped in three-ring binders, we couldn't send just
new pages with an update, but had to send whole new binders.  and of course,
we ended up having to special-order binders because they used nonstandard
binder sizes.

SUMMARY:

Base your page size decisions on these factors (not necessarily in order of
priority):
* How will the book(s) be used? Will they sit on a desk? Or. . . ?
* What kind of paper will you need (for example, how durable will it have to
be?)?
* How many colors will you be using (will affect which printing press you
use, and thus affect the paper)?
* What binding will you use? If three-ring binder, what size(s) are readily
and inexpensively available? If not three-ring, then what size paper does
the printer expect you to be using, and from this, what page size limits
does he recommend?
* What books have you seen that you like and that you think work well? Copy
them. Remember, good writers copy, great writers steal.

Let me know if you have other questions or if there is something I missed.


*********


From:   IN%"redish -at- quark -dot- umd -dot- edu"  "Edward F. (Joe) Redish"
To:     IN%"GIORDANOJ -at- delphi -dot- com"
CC:
Subj:   RE: Document Page Size

Received: by proton.UMD.EDU (5.57/Ultrix3.1) id AA16971; Mon,
 22 Jan 1996 15:42:01 -0500
Date: Mon, 22 Jan 1996 15:42:00 -0500 (EST)
From: "Edward F. (Joe) Redish" <redish -at- quark -dot- umd -dot- edu>
Subject: Re: Document Page Size
In-reply-to: <01I0B9D9AAZC9C1NPB -at- delphi -dot- com>
Sender: "Edward F. (Joe) Redish" <redish -at- quark -dot- umd -dot- edu>
To: GIORDANOJ -at- delphi -dot- com
Reply-to: "Edward F. (Joe) Redish" <redish -at- quark -dot- umd -dot- edu>
Message-id: <Pine -dot- 3 -dot- 89 -dot- 9601221546 -dot- A16964-0100000 -at- proton -dot- umd -dot- edu>


Hello Jim,

I don't have a definitive answer for you, but some
history and some thoughts:  A smaller size (say, 8 1/2
by 7) fits better on a user's desk than the
full size (8 1/2 by 11).  It also looks less intimidating.

You can chunk information nicely on an 8 1/2 by 7 inch
page.  Way back in the early 1980's, I had to
convince people to move out of the 8 1/2 by 11 size
to about 8 1/2 by 7.  Then, a few years later,
I had to convince some other people to move up
from 8 1/2 by 5.  That's too narrow a page.

With 8 1/2 by 7, you can get white space on the
left for headings, notes, etc.  and still have
a good short column width.  (By the way, I was
told years ago that the 8 1/2 by 5 inch page size
was not only because it was a regular sheet folded
over but because it matched the IBM 5 1/4" diskette
which it went with.

So, today, I would recommend 8 1/2 by 7.  Other nearby
sizes work too, but as you said this one doesn't require
trimming.  Microsoft does the volume to afford whatever
trim it needs.

A5 isn't a size that's common in the States.  Of course,
if most of your customers are in Britain or Europe, they'll
expect A5, so that would be a different story.

Usability arguments for 8 1/2 by 7 (or thereabouts):
  It fits well on the desk next to the computer.

You can get two uneven columns so you can outdent
    headings which makes them easier to find when scanning
    a page.
  You can get a good amount of text and a graphic on the
    page, but you don't overwhelm the user with too much at once.

I hope this helps,
Ginny Redish

********

From:   IN%"KentN -at- metrix-inc -dot- com"  "Kent Newton"
To:     IN%"TECHWR-L -at- listserv -dot- okstate -dot- edu"  "TECHWR-L"
CC:     IN%"Giordanoj -at- delphi -dot- com"  "'Giordanoj -at- Delphi -dot- com'"
Subj:   RE: Optimal Page Sizes


The size you choose depends on the type of binding you want to use, which
in turn is affected by your system's users, production costs, methods of
distribution, and frequency of updates.

For example, our users are generally IS and Management for field service
companies. They often use the manual in warehouses, shipping/receiving
docks, repair centers, and so on. They need manuals they can lay flat
when they refer to it.  Also, we update our manuals frequently (once or
twice a year).  To keep costs down, we chose to send updates to the
existing manuals instead of sending completely new manuals. Therefore, we
opted for a standard three-ring binder and a standard 8 1/2 by 11 sheet
size.  This is inexpensive and easy to produce and replace.  For European
distribution, we print the existing manuals on A4 paper size (with
corresponding A4 binders) without reformatting the documentation; this
minimally changes the size of the margins.

I hope this gives you a few things to consider.


*******


From:   IN%"73444 -dot- 3235 -at- compuserve -dot- com"  "Roger Grice"
To:     IN%"GIORDANOJ -at- delphi -dot- com"  "INTERNET:GIORDANOJ -at- delphi -dot- com"
CC:
Subj:   Document Page Size


Good question, Jim. I wish I had an equally good answer.

There are the standard studies that looked at line length. "Two alphabets
wide"
is the generally agreed-upon maximum--that argues for a smaller-than-usual
page
size.

The only other studies I can recall were studies of specific documents for
specific products where the focus was on: where will the manual be placed.
That's often a hard thing to know. But if you think about an 8.5 x 11 manual
opened flat, that takes up a lot of desk space. The size you're proposing
would
take a little over half the width.  That's goodness.

Another consideration would be: what other documents might the user be using
at
the same time. If the documents are all roughly the same size, it might make
for
an easier job of placing and balancing the documents, and it might mean less
of
a mental and visual adjustment when going from one document to another.

Not a great answer, I realize--more food for thought. When considering
usability, there are more situational dependencies than universal absolutes.

Please let me know what you decide to do. You've hit upon an important area.

Regards,
Roger


*********


From:   IN%"dmps -at- euronet -dot- nl"  "David Somers"
To:     IN%"TECHWR-L -at- LISTSERV -dot- OKSTATE -dot- EDU"  "Multiple recipients of list
TECHWR-
L"
CC:     IN%"Giordanoj -at- delphi -dot- com"  "'Giordanoj -at- Delphi -dot- com'"
Subj:   RE: Optimal Page Sizes


My personal preference is for 7 1/2 by 9, and I've designed a number of =
books in this size.
If you really want to stick to an ISO standard size, you could always =
use B5 (7 x 9 7/8).
FWIW, regarding the other favorite A sizes, I find A4 just too big, and =
A5 just too small to be useful for books (others feel free to agree or =
disagree!)

And having decided on you corporate page size, don't forget to get a =
profesional designer in to come up with the ideal typographic =
specification for your text [and yes, I am available for such work ;-) ]

David Somers
dmps -at- euronet -dot- nl


*********



From:   IN%"sgallagher -at- expersoft -dot- com"
To:     IN%"Giordanoj -at- delphi -dot- com", IN%"TECHWR-L -at- listserv -dot- okstate -dot- edu"
 "Multiple
 recipients of list TECHWR-L"
CC:
Subj:   RE: Optimal Page Sizes

In my present and two former companies (so that's three), our pubs
have been 7" x 9". For economy, my two former companies used perfect
binding -- the current crew is going with wire-o so the book will lay
flat.  The seven-by-nine size (with minor variations) seems to be
fairly common.

For the most part, standard paper sizes are irrelevant because the printers
use roll paper or large sheets and trim the books to their final size. You'll
need to consult your printer to verify this. However, if you have quick
printing to do in-house, the 8 1/2 x 7 (half-legal) size is a good choice.
Just make sure you can get your hands on some booklet-making software!

I don't know of any formal usability studies on book size, but there are
some generalized industry-wide myths:

* Big books take up too much room on the desktop.

* Loose-leaf binders tend to come apart too easily.

* Reference books are more convenient if they can be made to lay flat.

Hope this helps.


-Sue Gallagher
sgallagher -at- expersoft -dot- com


*********


From:   IN%"ibbetson -at- idirect -dot- com"  "David Ibbetson"
To:     IN%"Giordanoj -at- delphi -dot- com", IN%"TECHWR-L -at- listserv -dot- okstate -dot- edu"
CC:
Subj:   RE: Optimal Page Sizes

I like Kent Newton's suggestion of swapping between A4 and American Quarto,
except that, provided the manual isn't too thick, I would prefer A5 and
folded American Quarto. As I've noted elsewhere I find large manuals with
matching large type a pain in the neck.

David (the idiot) Ibbetson

David Ibbetson                            Phone (416) 363-6692
ibbetson -at- idirect -dot- com                 Fax     (416) 363-4987
133 Wilton Street, #506
Toronto, Ontario
Canada   M5A 4A4