Getting ready to print

8 June 2010

At NIAS Press, we call it ‘flicking through’. I’m sure there is a proper name for the process – final-final proof? – but essentially this is a last-minute checking of the cover(s) and body pages to ensure there are no silly mistakes.

This task is usually carried out by press staff (often including the production editor). The author is almost never involved.

A few of the classic errors picked up at this late stage are:

  • Details on the cover (e.g. subtitle) do not match those inside the book.
  • The wrong publication date is stated on the copyright page (normally because publication of the book is delayed).
  • Chapter titles in the table of contents do not match those in the text and/or in the running heads (names may be different or it could simply be that the capitalization differs).
  • Page numbers stated in the table of contents and in the lists of tables, maps, figures, etc. do not match the actual pagination in the text.
  • Caption text in the lists of tables, maps, figures, etc. do not match those in the text (as with chapter titles above) but note that this may be deliberate (e.g. because only abbreviated details are stated in the prelims whereas on the actual page the full details are given, including source and acknowledgement of permission to reproduce the material).
  • A chapter title, heading name or caption that should be listed in the prelims is missing there.
  • Chapter titles in the running heads are out of sync with the chapter pagination (e.g. the last two pages for Chapter 2 use Chapter 3’s running heads).

No matter how hard the production editor tries, some blemishes can slip through. This can be embarrassing but it is rare that the error is serious. (That said, it can be; I remember one publisher having to withdraw and reprint an annotated edition of the Koran due to a single spelling mistake being found a few days after publication.) However, once this last-minute check is done and any mistakes corrected, the book files will be uploaded to the printer.

Finally, your book is (almost) out of your publisher’s hands. All that remains is to create the final print files and send them off to the printer, something done very quickly and simply these days via the internet (often via the medium of a FTP server).

Now our attention can shift to the printer. The question is, which printer?

(Post #3 of the Printing section of a lengthy series on the book production process, the first post of which is here.)


Final proofs

5 March 2010

Once the text is completely stable, and any illustrations have been sized and inserted, the typesetter paginates the book and produces a second (and hopefully final) set of proofs for checking. In urgent circumstances, it is not unknown for authors to receive only a single set of paginated proof pages, but two proof stages are more common. At the same time, you should get a proof of the finalized cover.

Page-proof checklist

However, the second (paginated) proofs are not yet another opportunity to check your text; by now, as we have seen, your production editor will be pretty intolerant of ‘unnecessary’ changes. While you should keep an eye open for any errors in the text not previously spotted, at this stage the things that you should be focusing on are quite different. Now, you should check that:

  • All corrections marked on the first proofs have been correctly implemented.
  • All figures, illustrations, captions, tables, etc. are placed where they belong (or in close proximity to this).
  • Any illustrations look as they should (in terms of quality, size, colour, orientation, etc.).
  • All footnotes (if used) are placed on the correct page.
  • Chapter titles in the table of contents match those in the text, not only in wording but also in upper or lower case. (The same applies for figure captions, etc.)
  • Chapter titles in the running heads match (or are reasonable short forms of) the real chapter titles.
  • Page numbers stated in the table of contents, list of figures, etc. are correct.
  • Any ‘hard’ cross-references (like ‘see overleaf’, ‘see page 43’, etc.) are correct.
  • Pagination of the book is consecutive (with numbering of the preliminary pages as a separate series using roman numbering).

Of course, any corrections and other changes to these proofs should be marked up as discussed in my earlier post.

Proofing the cover

I have already discussed finalization of the cover in quite some detail elsewhere. Suffice to say here that what you should be seeing now is not some cover concept or even a well-developed draft but the final version. As such, you will need to be proofing not only for errors but also omissions (a promised photo credit in tiny type on the back cover, for instance).

Given the central importance of the cover, it is wise to take special care on the cover proof.

Avoiding the errata slip

Publishers hate having to insert an errata slip in a book, not least because it is time-consuming, expensive and unnecessary (and because it is an open admission that the proofing of that book was inadequate – not a good look).

As such, because this is likely to be your last chance to check the entire book (see below), be rigorous with your proofing. This means, too, that you should look at the whole book – both inside pages and cover – making sure that everything is right. (This is especially relevant because the cover and inside pages are usually produced by different people.) Otherwise you may end up disappointed when holding your wonderful new book in your hands and discovering a silly mistake.

This is what happened to one of our authors a few years back. The title page she delivered for editing had an old subtitle. There was nothing wrong with this (hence it survived the editing and typesetting unscathed); it was simply the wrong subtitle. The new subtitle appeared on the cover of her book and in all sorts of marketing material. It was also the one that was registered in various bibliographic databases.

Unfortunately, neither the author nor anyone at the press noticed the minor discrepancy in subtitle wording until after the book was published. She requested an errata slip. Fair enough, but I was not amused.

Making a ‘right-brain’ assessment

In addition, at this stage it would be a good idea to flip through the book looking at each double-page spread (easiest done in Acrobat or Adobe Reader) and analysing the layout in a more ‘right-brain’ fashion. Are the pages balanced and aesthetically pleasing? Do the page bottoms line up? Do you like what you see?

Of course, it may be dumb me suggesting that you do this ‘right-brain’ assessment now as there is no way that either the typesetter or your production editor will contemplate major design changes at such a late stage. The time for such feedback should have been at the time of the first proofs or even earlier in the design phase (if you were consulted, that is. The whole issue of what you can and cannot expect to change at the proofing stage is discussed in detail here.)

That said, it is your book, your child. If you don’t care about how it is dressed, who else will? And what will your readers think if, when they encounter your book, they are distracted by its appearance and maybe even fail to take it seriously?

A wee bit of assertiveness with your publisher doesn’t harm once in a while.

Finalizing the proofing process

As with the first proofs, your job is to indicate any changes required either on the proofs or in a separate document, returning these (or a message that there are no changes) to your production editor. Usually, there is great pressure for this to be done quite quickly. As before, make sure that you retain a copy of these proofs.

At this stage, however, and before you return the proofs to your publisher, you may have an extra task to complete: preparing an index. If there are minimal changes to these second proofs, then it is normal that they are used for the indexing (that is, if it is the author doing the indexing); this saves time. More about this in my next post, which starts a new section of the production process looking at indexing.

And, as far as you are concerned, that is (almost) the end of the proofing process. (‘Almost’ because you should get a chance to see the typeset index and maybe even the whole book again after any second-proof changes have been implemented.) From now on, you will take a back seat as far as the production process is concerned. More proofing will be done but this will be by (or at the behest of) your production editor; as noted earlier, it is unlikely you will be involved.

Time then (after the indexing) to move on with your life. Indeed, already by now, you may need to start refocusing your attention on the promotion of your book. But that is another story.

(Post #9 of the Proofing section of a lengthy series on the book production process, the first post of which is here.)


First proofs

1 March 2010

Normally, initial typesetting of your book will not take too long, especially if no attempt is made to finalize the pagination of the book (for instance, no illustrations yet placed in the document).

As we have seen earlier, the result of the initial typesetting is a set of first proofs delivered to the production editor. These proofs will be quickly checked then your production editor will send you a copy for proof-reading. At the same time, chances are that someone at the press or an outside professional proof-reader will check this first proof as well.

About the first proof

The proof received will almost certainly be typeset text output on ordinary pages (described and illustrated here) rather than the galleys of yesteryear. In all probability, these page proofs will not have been finally paginated (essentially because the illustrations are missing and some changes to the text are expected). In addition, as pointed out in Leena’s mail above, much else about the proof indicates that it is a preliminary version of your book. For example:

  • The running heads are incorrect or non-existent.
  • The text will likely be loose in places (any hyphenation or massaging of the character spacing yet to be done).
  • Text may not completely fill the page so that page bottoms in a two-page spread are not lined up.
  • Tables may straddle two pages.
  • Figures and illustrations may be missing (though, as illustrated in my previous post, space may have been reserved for these).
  • Any ‘hard’ cross-references (like ‘see overleaf’, ‘see page 43’, etc.) should be flagged as provisional.

On the other hand, you should expect something more than a picture of the typesetting in its raw, newly converted state. In particular:

  • The book/page design should be apparent in the page layout (e.g. trimmed page size and fonts are correct).
  • Paragraph (and character) styles should have been implemented in the text.

Your primary task

What your production editor will be expecting you to do in quite a short time is carefully to go through the proofs, marking up any changes required. (My next post describes marking up in greater detail while the subsequent post lists the common proofing marks used in printing and publishing.)

Mainly, you should look for problems in the text, examples being:

  • Typos.
  • Corruptions in the text (perhaps caused by the text conversion problems described a month ago and again more recently).
  • Missing text.
  • Incorrectly formatted text (e.g. missing italics).
  • Paragraph appearance is inappropriate (possibly due to incorrect assignment of paragraph style).

Of course, you may wish to request other changes that have nothing to do with errors in the text. This is a problematic issue that I shall discuss in a separate post later this week.

Proofing with the other side of the brain

In addition – although the lack of illustrations, final formatting, etc. may force you to stretch your imagination somewhat – now is probably the last moment when you can comment on the page design and request changes. (Unfortunately, for most authors it is also the first time that they have seen how the publisher intends to layout their book – so this is a last-gasp thing I am foisting upon you. Indeed, it is probably too late for you to make a similar ‘right-brain’ assessment with the final proofs that I suggest in an upcoming post.)

While it is unlikely that you can get your publisher to completely change the page design presented to you, there is a chance that you can convince your production editor to make some changes. For instance, if you can demonstrate that the fonts used evoke a mood/approach at odds with (your treatment of) the subject, then it need not be too late for such a systematic change to be implemented.

Be smart, however, in what you request. You’ll have a greater chance of agreement to global changes that can be implemented relatively easily rather than to custom changes that must be made on a case-by-case basis.

And finally

There are different ways of noting your proofing changes/corrections; these I describe in my next post. If you choose to mark up the proofs received from your publisher, then make sure that you save a copy of what you send back. This will be your reference at the final-proof stage, ensuring it is quick to check that the changes you requested have been implemented.

(Post #5 of the Proofing section of a lengthy series on the book production process, the first post of which is here.)


Typesetting phases and outputs

12 February 2010

For the last two weeks, I have been blithely discussing what typesetting is and the issues relating to it but without actually describing what is produced. Time for a short overview before we move on a new thread of posts, on the proofing process.

A duet

Typesetting is not a single event in the production of the book pages; rather, it is a multi-phased process and one interwoven with that for the proofing of the book. In a sense, one could look at typesetting and proofing as a duet sung by a tenor and soprano.

Their ‘performance’ looks somewhat like this (but note that approaches do vary between presses):

  • Initial typesetting
  • Output of first proofs
  • First proofing
  • Completion of typesetting
  • Output of second (often the ‘final’) proofs
  • Second (or ‘final’) proofing and indexing
  • Output of print-ready copy
  • Final-copy check
  • Delivery of print files to the printer

Let’s look at the initial typesetting, its completion and briefly at how the proofs are output. More detailed descriptions of each of the proofing phases shall follow in my next thread of posts on proofing.

In addition, also required (and not mentioned above) is typesetting of the cover/jacket (more about that in my next post).

Initial typesetting

Practices differ between typesetters and another major factor is which typesetting software is used. However, likely steps are as follows:

  1. Finalization of the page design (as discussed here).
  2. Creation of typesetting documents meeting the design specification.
  3. Conversion of those input files not conforming to the page design/software requirements (e.g. image files changed from colour to monochrome and from JPEG to TIFF format).
  4. Marking up of text files with consistent paragraph styles that match those defined in the destination typesetting documents.
  5. Import and placing of text files in the typesetting documents.
  6. Assignment of paragraph and character styles to the text and any tables (or, if already done at step #4, then fine-tuning of styles).
  7. Import and placing of any image files in the typesetting documents (often this step is left until completion of typesetting).
  8. Generation of first proofs.

Proofing output

Not too many years ago, when typesetting was done on specialist machines, the initial proofs were output in galley form, i.e. as continuous text without any page breaks marked and printed out on what looked like giant-sized toilet paper. As the layout was finalized, the book would be paginated, subsequent proofs clearly showing the page breaks. These proofs, too, could be printed on long galleys or guillotined into their individual pages.

The output of today’s PC-based desktop publishing systems is utterly different, being based on the industry-standard PDF format (though other output formats are possible). Galleys are gone; everything is either printed on ordinary (A4 or US Letter) paper or output as PDFs. Moreover, there is little difference in the appearance of (say) the initial set of proofs and the final print files sent to the printer (not least, all proofs are paginated).

The shift to PDFs has been a revolutionary development for authors. This, however, is something I shall take up next week in a new section of posts detailing the proofing and indexing process. Proofing outputs are discussed in greater detail here.

Completion of typesetting

Just what is needed to be done to complete the typesetting process depends of course on if step 7 above was done during the initial typesetting or has been left until now. If the latter is true, then likely steps are as follows:

  1. Keying of any text changes from the first proofing.
  2. Import and placing of any image files in the typesetting documents.
  3. Re-evaluation of the likely extent of the book including space for the index (not yet prepared, of course).
  4. Possible adjustment of the page design (especially of the font size and line spacing) to meet the final extent set by the production editor.
  5. Pagination of the book (including subtle adjustments to line spacing, to the placement and size of tables and illustrations, etc. to save on – or add – a few lines here and there so that the target page count is indeed reached); the ideal is that each double-page spread has even page bottoms, its composition is evenly balanced and the overall effect is aesthetically pleasing.
  6. Finalization of any page-specific cross-references.
  7. Generation of second (often ‘final’) proofs.

After the return of any changes resulting from the final proofing and delivery of the index, typesetting concludes as follows:

  1. Keying of any text changes from the final proofing.
  2. Typesetting of the index.
  3. Generation of print-ready copy for checking/approval by press staff.
  4. Delivery of print files to the printer.

Time then (almost) to move on to the proofing and indexing of your book but first let’s return to your cover and its finalization.

(Post #21 of the Design & Typesetting section of a lengthy series on the book production process, the first post of which is here.)


Unrealized expectations

9 February 2010

You hoped that your editor would agree to the use of colour inside your book. Nah! Too expensive. (We’ve talked about this already.) And as for the cover, none of your ideas were even solicited let alone listened to (this also discussed here). But some of the other things you are expecting with your book – or just assume will be implemented (their merits are so obvious) – may not happen either.

Oh dear, time for a few disappointments (and a dash of incredulity).

Footnotes or endnotes?

Some authors hate endnotes with a passion while others naively expect to see the method they used in their manuscript replicated in the typeset book. Expect the worst. More than likely, your footnotes will be converted to endnotes and placed at the end of each chapter.

From a publisher’s viewpoint, this is quite reasonable. Technically, with modern DTP software, footnotes are relatively easy to work with (as easy as they are with a word processor like MS Word). That doesn’t mean that publishers are willing to retain an author’s footnotes, however.

Perhaps the main reason is conservatism but in many cases the problem is a question of avoiding layout hassles. This is especially the case when not only are there a whole lot of notes but also there are tables and illustrations to be placed in the text. Juggling the placement of text, notes, tables and illustrations while preserving an aesthetic balance can be a real headache, as can be seen in this example.

As such, it may be that you get little say in whether your book is typeset with footnotes or endnotes. Mind you, if you are a typical author, perhaps you haven’t given your publisher any idea of your wishes (let alone any incentive to take your wishes into consideration) until it is too late. As I’ve pointed out earlier, authors often show great interest in how their cover looks like but they rarely seem to concern themselves with the page design.

Other layout expectations

It is likely that other expectations you may have about the layout are less important to you. Some of the more common features found in an author’s ms that often are not retained by typesetters are:

  • Bold text (many typesetters regard bold as a ‘shouting’ style and prefer the use of italics, seeing this as a more subtle and elegant type style).
  • Capitals (another ‘shouting’ style, best replaced by text in title case, small caps or italics).
  • Headings (changing the alignment and replacing attributes like bold and/or capitals with small caps, italics, etc.).
  • Font (this may be predetermined in a house style for the press or this series, the need to save space may mandate use of a more compact font like Minion, or perhaps it’s Monday and the typesetter feels like using Garamond; certainly, it’s unlikely that the default font in Word – Times New Roman – will be at the first that the typesetter chooses).
  • Type size (determined in part by the font selected and in part by whether the typesetter needs to save space or pad out the text a bit).
  • Placement of special text (as discussed earlier, even in the age of Open Type, many publishers are averse to the use of non-Latin script, diacritics, etc.; such text may end up segregated in a separately typeset glossary, though if so this should have been discussed with your before now).
  • Placement of tables and illustrations (even if, as I recommended here,  you have indicated that this table should go here and that illustration there, the complexity of the layout – or typesetter laziness – may mean something different happens).

Disconcerted in other ways

The layout is not the only source of nasty surprises for authors. Even though authors generally get to see what is written when their book is first announced, it is amazing how often they act surprised when seeing the first proofs after the typesetting – specifically after looking at the title page and copyright page of their book. Common causes of authors expressing shock and horror are the:

  • Format/edition (do not assume there will be a paperback edition of your book; some publishers are especially unlikely to publish in paperback).
  • Pricing (if your publisher is aiming to sell to the academic library market only, then the price may be horrendous; forget about all your colleagues buying your book).
  • Subject (your book may be wrongly classified – labelled as politics instead of anthropology – but, just as bad, the classification may be too simplistic. This is no surprise; given the way bookshops function, your book may only be displayed in the China section – if at all – even though it offers a major breakthrough in sociological theory as well and thus deserves to be shelved under Sociology as well).
  • Series (unless you have approached the press proposing that your study be published in a specific series, it may be simply plonked in that series which your editor thinks is most appropriate; in fact, it may not be so).
  • Publisher (yes, some authors even get this wrong, thinking for instance that – because their book will be distributed in North America by Such-and-such University Press – the imprint of that press will appear on their title page; not so).

Of necessity, each of the points above deserve far greater explanation than there was room for here. I’ll do this in several posts another day. Meantime, let’s explore perhaps the greatest problem that you will face with your book: letting go, handing over your ‘baby’ to strangers.

(Post #19 of the Design & Typesetting section of a lengthy series on the book production process, the first post of which is here.)


Placement of non-text elements

4 February 2010

Imagine this. A key passage in your book surveys the tourist icons of Paris. As your readers learn about the gargoyles of Notre Dame, on the opposite page they see the iron lacework of the Eiffel Tower. Turn the page and the story of the tower is told – illustrated with a picture of buses and tourists outside the Louvre.

These things happen. They shouldn’t.

Double vision

A key difference between modern typesetting programs and a word processor like Word is that the typesetter works with double-page spreads, the same view that readers have when moving through a physical book. (The single-page perspective of e-books is one of their major disadvantages, by the way, but this is rarely mentioned.) This double-page ‘workbench’ makes it easier for the typesetter to place tables, illustrations and other figures to their best advantage.

(Easier but not always easy. The layout of page after page of straight text is quite easy. But having to juggle the placement of text, footnotes, tables and illustrations all within a few pages – and ensure that the result is both aesthetically pleasing and meaningful – is decidedly not simple. This is one of the reasons why some publishers insist on endnotes – but more about that later this week.)

Insertion points need to be clear

Unfortunately, typesetters aren’t mind-readers. While some diligently read the text so they can lay it out in the best possible way (for this reason, often it is the typesetter who notices errors and omissions not discovered at editing), others do not have the time for such a hands-on approach. What you should count on, then, is that your typesetter won’t have the time to decipher from your text where precisely this table should go or where that illustration.

For this reason, you would be smart to indicate in the text where every non-text element should go. Something like ‘INSERT FIGURE 3.2 ABOUT HERE’ may not be pretty but it is necessary. (Indeed, your failure to specify an approximate insertion point may cause the typesetter to completely overlook that illustration and fail to place it.)

Colour sections

Placement of colour illustrations (as discussed in my previous post) is slightly more tricky. There is no problem if the book is completely in colour (simply indicate an insertion point as above). But if your colour illustrations are to be gathered into a colour section, then there is an issue. The point is that this colour section is a section; here, we are not talking about a scattering of colour illustrations placed wherever they are referred to. No, a colour section is a 16-page, full-colour signature that must be inserted between two monochrome signatures. Precisely where this will be inserted cannot easily be predicted in advance. All that may be possible is to indicate an insertion point as close as possible to this or that passage of text.

Certainly, however, you should not leave placement of any colour section to the whims of your typesetter. Be proactive, discuss things with your production editor and ensure that placement of this colour section is specified in the page design brief (discussed in greater detail here, while my argument for author activism is here).

One last thought

What is discussed in this post is placement of non-text elements, not if they are actually necessary. Pictures may say more than a thousand words but they also interrupt the narrative, as do tables, charts and graphs. In each case, ask yourself whether this interruption to the flow of text is necessary, appropriate and desirable.

(Post #16 of the Design & Typesetting section of a lengthy series on the book production process, the first post of which is here.)


How many charts and tables in a book?

28 September 2009

An author asks:

About the tables and charts in the book, do you have any rough rule of thumb on how many tables and charts would be sufficient or too much? In the dissertation, I had 8 charts and 14 tables and I am looking to cut those numbers considerably. Also, do you have any general advice on what kind of tables and charts to keep and what to discard?

Many years ago, we published a study of entrepreneurship in Vietnam that included over 200 tables. This was quite a nightmare to typeset and not exactly easy reading. I’d never do this again and doubt that many other publishers would be game to take on such a mammoth task, either.

Mind you, ultimately, it all depends on the field and subject. This was an intense economic analysis, a type of study in which masses of charts and tables are almost expected. In contrast, one would expect a political or historical study to have far fewer charts and tables. As such, ‘merely’ 8 charts and 14 tables may be OK for such a study. The question is, however, is that number OK? Accordingly, ask yourself

  • What is the purpose of this table or that chart?
  • Can its meaning be worked easily enough into the text instead?
  • Is a table really the most effective way of presenting your information, or is a chart/graph a better communication tool?

In line with my ongoing obsession with readability, I would argue that you only include those charts and tables that illuminate the text. In addition, avoid like the plague any charts or tables that are oversized and/or have excessive detail, also those that are only meaningful if reproduced in colour (more about these points in later posts).