In the last of our three week series, we’re carrying on where we left off last week, continuing to look at the challenges involved in publishing images in your book, and how to manage them. We’re concluding by looking at issues around image quality, and the particular challenges that relate to selecting a cover image. If you haven’t already, I’d recommend reading the previous two weeks’ posts first, then here goes:
Image quality
From the legal kind of technical, we now move to a more practical kind of technical challenge presented by the presence of images in a manuscript. That is the challenge of making sure they actually look good in the final book, both for readers of the print editions and for those readers using e-readers, tablets, phones or their PCs to read the book.
Lines
Line diagrams should in theory be much simpler to deal with than halftones. The problem is that authors often construct them in what can be best described as “esoteric” ways. I’ve processed many a line diagram produced in Microsoft Word, or which is just a photograph of something in a textbook. The former is a pain to work with because it is a kit of parts that will not remain stable when we start manipulating the text. The latter is a problem because it tends to lack the clarity of a purpose-drawn diagram and instead is functionally an – – invariably – – murky halftone image.
The ideal way to provide line diagrams is as vector graphics. This is because then we can scale them to fit on the page as we adjudge necessary, with no need to worry about resolution. The whole point of vector graphics being that they store the information about the relative positions of the lines and text in the graphic. They can thus be made infinitely bigger or smaller without becoming pixelated, or grainy. If that’s not possible, then creating a jpeg or similar static file format from an image created in Word (or Powerpoint, or Excel) is the next best option.
Resolving resolution
Halftones — again that’s photographs and other illustrations that are not simple line diagrams — need to be provided with sufficient resolution to look good in the book. Print tends to be the standard here, as for various reasons images can often look better on screens than they do when printed. In fact, a good way to test whether an image on your screen will look good in your book is to print it yourself and see how it comes out. Professional printing presses will produce better results than your home or office printer, but if the image is pixelated or muddy because of low resolution then a better quality print will, if anything, exacerbate this. It’s like playing a poor quality tape recording over a high-end hi-fi setup instead of cheap headphones, it renders the imperfections more obvious.
To put this in numerical terms, aim for your images to be at least 300 dpi at the desired size. I stress that last part because not everybody realises that dpi stands for “dots per inch”, and is thus a ratio of pixels to size. In other words, if you double the size of an image, you halve the resolution, because there are half the number of pixels per inch. So its no good having a 300 dpi picture that’s only a couple of centimetres square, unless you only want it to be printed as a couple of centimetres square.
Technically good enough is not the only kind of good enough
And we shouldn’t lose sight of the fact that there are other ways for halftone images to be poor quality. It is perfectly possible, for example, to have a large, 900 dpi image, but one which is blurry, smudged, where somebody’s thumb is half way over the camera lens, the light is bad, or which is just poorly composed.
It’s a less common issue than low resolution, but it’s surprised me over the years how often some authors will give us inexplicably unattractive images to use in their books. Sometimes this is because they are very literally-minded and want to illustrate everything, regardless of whether they have visually appealing pictures with which to do it. Other times they have the pictures and so they want to use them, apparently because they have them. There certainly seems to be an expectation that we’re going to take these uninspiring images and somehow make them look better, in the way that film and television crews can make any of us look a bit more telegenic with competent application of make-up and lighting. Alas, this is generally not the case.
And what about colour?
The feasibility of internal colour images is one of the features that has changed the most since I began working in publishing. There was a time when the answer was a firm “no” unless you could find some funding to cover the cost of an 8-page signature in the middle of the book, something which “trade” publishers still sometimes do. For textbooks with very large print runs it was possible, but would mean printing somewhere cheap in bulk and then having books shipped half way around the world.
Print economics, at least for smaller print runs, have changed dramatically since then. It is now a relatively trivial marginal cost to print a book in full colour throughout rather than in greyscale, and because academic books tend to already use quite high quality paper, there is relatively little reason to ever do a colour signature in such books. There are downsides. Notably, short run print suppliers able to print in full colour are still a minority, so full colour can introduce other complications to the supply chain. We might have to print books further from where we need them than might otherwise be possible, and the quality may be more variable (or the variability may be more noticeable) than it would be otherwise.
My advice on colour has therefore changed a little less over the years than you might expect. These days I would say that if the book needs it, if it will add substantially to the reader’s experience, or if the topic of the book really demands it, then go ahead. The impact on cost will be fairly small and it will help to sell the book. But I would also say that if the book doesn’t really need it, it is best avoided. There are still just about enough reasons not to go full colour that it’s worth avoiding where it doesn’t add any real value, and it should not be used trivially as the default.
The most important image of all…
…is of course the cover image. There’s more to say on this than I have time for here, but there are a few points to make in light of the above. In essence all three issues that apply to images, apply all the more so to the cover, especially copyright and quality.
Volume
Ok, this one doesn’t apply all that much to covers. Except when it does. There is a habit some authors have, of responding to their publisher’s request for suggestions of cover images with a huge stock of uncurated images. That can be helpful if they are genuinely good images and you’d happily use any of them. It is less helpful if it’s more a question of “here’s everything I have, do something with it”.
Then there are the subset of authors who shower us with images and propose that they would like a collage on their cover. I’m not going to say this can’t work, because there are exceptions to every rule, but it is rarely a good idea. Picture in your mind the most memorable book covers, movie posters, or album covers you can think of. How many of them are composite collages of a range of images? Exactly.
Copyright
Here everything I’ve already said applies, but much more so. There are three reasons for this.
First, the odds of somebody spotting that you’ve used their image without the correct permission is drastically higher if that image is on the cover of your book. We want to be able to display your book cover with absolute confidence that we have exactly the rights we need to use the image on it. It is by an order of magnitude the most noticeable image in (on) the book, that is after all the entire point of a cover.
Second, we need to pay extra close attention to the specific terms of any user license that applies to the image. Particularly if you’ve acquired the image commercially, rather than through a private arrangement. Photo libraries will often charge higher royalties or fees for images used on a book cover. Publishers providing permission are also likely to want to know how prominent an image is going to be, and will be very unhappy if they provide permission with the expectation that an image will be used internally if it will, in fact, be on the cover.
Third, the cost — financially but also in terms of time and effort — of getting this wrong is also amplified. Completely overhauling the cover design and rejacketing a book, as well as trying to make sure the correct cover is showing in every catalogue and webpage is a massive headache and very costly. We might never be able to fully suppress the original cover online completely, and we could be haunted by it popping up at random for years to come, probably accompanied by irate communications from the rights-holder demanding to know why such and such a website is still showing the original cover.
Image Quality
I am sure I don’t need to reiterate why basic technical questions around resolution apply ten times as much to your cover image as to internal images. Whatever you’re putting on the cover is the visual representation of your whole book, so it should go without saying that it should have a decent resolution and not look grainy.
In fact, we should probably go a bit further, ideally your cover image has a substantially higher resolution than 300 dpi. We want to be able to put it on flyers, posters and other marketing materials where it’s good to have the option of making the cover larger than it is on your actual book. It follows that 300 dpi on the cover itself is, as they say, the floor and not the ceiling.
Similarly, because this image will represent your whole book, it needs to be eye-catching, crisp and clear (unless you’re making a very deliberate choice for it not to be). At the same time, when selecting an image to suggest to your publisher, try to avoid being literal for its own sake. This is often a problem for STEM authors, in my experience, who are sometimes prone to sending a photograph of a piece of machinery used in the book, or an important graph of one of the effects described, or even some formulae drawn up to look a bit arty. Can this kind of thing sometimes work? Sure, especially if you have a phenomenally skilled cover designer as I did in my early days working on Engineering books. Does it mostly end up looking pretty lame? Absolutely.
Of course it’s good for the cover to visually indicate some aspect of what the book is about, but it’s better to have a completely abstract but eye-catching design, than it is to have a literally accurate, but rather dull one.
Perhaps we’ll go into these kinds of questions in more depth in a future update, but I think that covers the basic considerations for choosing cover images.
This has been something of an epic, stretched out over three weeks as it was, I hope it’s been helpful. I was very pleased to see some commissioning editors from other publishers sharing and recommending this series. Next week we’re going to zoom out and take a look at how the publishing process runs from start to finish.
As always, do feel free to get in touch on proposals@tbarnpress.com if you have a book project you’d like to discuss with us.
