judging pixels by their, uh, cover.

[kaigou]

A few days back a friend was contemplating the cover design for her next story. With my bookseller-cap on, I naturally had feedback but then held back (partially, at least, because I wasn't up to sitting still long enough to type much, haven't been for awhile). But seeing the alternate covers got me thinking, or at least just scrutinizing cover art in ebooks a bit more closely.

A day or so later I came across this note in a publisher's submissions guidelines/faq (edited for more generic version after seeing roughly the same sentiment at three different publishers):

We do not allow authors to provide cover art... We have a staff of artists who produce all our cover art... we're proud of the quality of our cover art.

Which I find amusing in some respects, because some of those companies are the same ones with the covers for which my only reaction is, "my goodness, that's unfortunate."

Granted, the majority of publishers (not all, but the majority) have at least moved away from computer-generated Poser covers, long may those burn in whatever hell you please. But a fair number make mistakes that to me seem pretty basic: low contrast, funky-ass font faces, static poses, and some of the most uninspiring and clueless compositions (or lack thereof). Yes. Right. Goodness, that's unfortunate.

Which is especially amusing, considering that I've had my share of friends who've commented they'd never buy a romance, no matter how good, because the by-the-numbers cover art just screams lurid!bosom!heaving!crap! Yet here I am in the privacy of my own home contemplating various works and thinking, I would never buy that book, I mean, the cover is just so freaking pathetic that I can't even imagine clicking on it.

(That's particularly sad considering that one of the writers I really like, James Buchanan, has somehow been saddled with some of the most atrociously bad cover art I've seen in a long time. It's like some five-year old got loose with a stack of magazine clippings scanned at low-quality, and thought to make up a cover by cutting and pasting in Photoshop. It's that bad, and yet Buchanan's novels are sharp and good and fierce. Why doesn't someone of Buchanan's calibre rate a cover artist like Anne Cain, anyway?)

Just around the same time, I came across mention of an online series of workshops (now closed, sadly) that were hosted/run by several of the big names in ebook coverart. There were at least links to the artists' works, and I was struck by how gorgeous many of them were -- and yet something indefinably different from standard, off-the-shelf, book cover art.

(No, I'm not going to post a litany of the Worst Covers Evah, because it's not the author's fault, but it's their name on the book that may end up associated with the snark. It's their publisher's fault for not having the decency to hire better freaking artistic directors -- not artists, because some of the art can be decent but without proper layout/design, it's defeated by bad font, bad type, bad contrast, bad composition. Bad publishers! So, no litany of snark.)

Frex, here are four covers of paperback books, chosen at random from the compilation I posted awhile back (so they remain in my gallery and thus easily snagged again, lazy me).



Now, here are three of my favorites from Anne Cain and graphicfantastic, and resized to be approximately the same as the Amazon-snagged images above:



It wasn't until I realized that one of those books in the second group was designed for print that it dawned on me: the ones for print have a lot of itty-bitty words cluttering up the cover. Mention of the author's other bestsellers, one blurb, maybe even a teaser (or is that a second blurb?), the title, the author's name, and some other stuff I can't make out.

Even so, in the second group, that one designed-for-print has some tiny letters but it's minimal. Definitely not quite cluttery, not nearly as much as the covers in the first group. I mean, this makes sense; normally, when I hold a paperback and look at teh cover, it's a fair bit larger than the 200px high version you're seeing on your screen right now (or that I see when browsing ebooks).

That means there's a bit more leeway, possibly. I mean, if the first group of covers were among a selection of ebooks, I might take a second look, based on the cover design tempting me, but it's a fair bet I'd ignore them as I do the print-versions sometimes scattered in among ebook versions (a habit I honestly can't stand from some publishers, because then it looks like they're selling two or more versions of so-and-so's latest -- I'm there for ebooks, show me the freaking ebooks, and set the print versions elsewhere and stop confusing me! ... okay, snark dialing down again).

Of course, a second look at the first group of images might also be purely for the sake of laughing with my inner twelve-year old. Because the first one, in that top group? What pops out isn't the red-on-blue (a nice contrast for print) but the white in the center, making the title look like OOD LINE and ahahahah what a stupid name, my inner tween shrieks. (And then I look again, realize, and move to the next book now that I feel stupid myself -- ohh, so that's what it says, okay, FINE.)

The second one? Red on dark gray to black is catchy in the 3D world, but not quite as clear, so it's harder to make out the title without staring intently for a second or two. The third looks like 'Embracc the Nigot' from the extra curlicues of the c and e, and the way the upper bar of the h in 'night' gets lost in the line of the woman's dress. The last one would work except I can't for the life of me make out the author's name, lost as it is with light color on light background and barely-there outlining.

Lesson: it's a lot easier to mess with contrasts and font-styles when the final version to be used as basis for judging is a version at least eight inches high by five inches wide (or greater). When it's on a computer screen that could be of any variation in color-setting, brightness, contrast, who knows what the user will see.

(There is a special place in my list of annoyances for those publishers who show the cover and note the author's name underneath -- but not the title. I can't always figure out the title when the contrast or composition or font-style is unclear. I don't always order immediately, but come back and ponder again, and it's awfully hard to find a book upon your return when the author's name was in itty-bitty link under the image, and the book cover was fuzzy or unclear in some way. "Well," I say to myself, "it looked interesting, but who knows where to find it, so never mind.")

I've also noticed that some publishers cleave closely to the kinds of styles you see in four specific genres: romance, science fiction, westerns, and epic fantasy. It's like a type of framework that must bracket the front image -- like the author's name and the title get their very own frames, and the artwork has to fit inside this imposed frame, as well.

Some e-publishers do that for specific lines/brands -- you see that style, you know, this is one that's based on X theme, or is Y length, or has Z content. The only one who seems to have nailed this style is Torquere, whose short-story classes (short, medium, novelette, novella, etc) have a specific repeated -- but very spare, simple -- graphic. Most book covers do not have a picture of a highball or a martini on them, so it's pretty obvious it's not a cover per se, so much as a stock indication of story length or theme.

The majority of other e-publishers, though, have series-covers that look remarkably like a 'regular' bookcover, and it's just re-used. I end up wondering if I'm looking at the same book I was five minutes ago, or maybe that was another book, who knows. The design is so close to regular book cover design (at least of epub ilk) that I realize I'm not thinking, 'oh, dat's a graphic image' but thinking, 'book cover!' -- and to see six 'book covers' that are all identical is to see the first and visually, if automatically, skim the rest. (This is complicated further when some publishers separate the epub and print versions and list them both on the same page -- great, even more duplicate covers for me to sort through.)

Another thing I've noticed is that tpub (print) books tend to lean towards full-body, or three-quarter (knees/mid-thigh and up). Ebook covers, perhaps because they have to make more impact in a much smaller format, tend to be busts -- either shoulders and up, or the same amount of space but dedicated to a torso. That's if the cover isn't of a face really close up, which I only recently realized is not that common in print books, judging from a long swath through Amazon. It can be used to dramatic effect, but then again, a face taking up the cover when you're looking at it in eight inches high can be a bit more overpowering than in 200px, after all.

I wonder if this respect for the medium (or lack thereof) is why I find some ebook covers repellent on a level I couldn't quite pin down, before: the worst of the worst seem to consistently frame and place cover-characters on the same frame of reference as print books: full or three-quarters. I end up squinting at these blurry covers that don't really make an impression because there's too much in the tiny frame.

In some ways, epub is leading the pack, and I can't say there aren't artists who are really at the forefront -- Anne Cain and GraphicFantastic being absolutely two that just can't seem to produce a bad cover even if they're paid good money to do so. But the majority of epubs seem to be continuing to look backwards when it comes to cover designs -- looking to compositional lessons of tpubs, that is -- instead of realizing that their medium is so utterly different even to the means and method of letting covers do the selling.

It's a different medium, a different format, and the art direction needs to reflect that. That means higher contrast between title and background, and less of these funky fonts that may look good when the image is 400px or greater in height, but gets scrunched so much at selling-version of 200px that it's nearly pointless. Hell, in that case, I say just hide the title/author, put that below, and use the image as selling point but not necessarily as the identical format of the image I'll see when I open the PDF. (That is, text added in once I'm viewing in larger format.)

However, all that is beside the point for those epubs whose cover artists -- no matter how dedicated the staff may be -- still suck. Some of them seriously make me wince, and all I can think is: oh, you poor poor author, stuck with that cover... my goodness, how unfortunate.

Comments

[autopope.livejournal.com]

We do not allow authors to provide cover art...

That's probably a good thing. Not many authors have graphic design training, and only a tiny subset of those have actual cover design experience.

I'd like to throw a handful of other insights in the pot:

1. For the major publishers (who insist on DRM) ebook sales are currently so lame at present that there's no budget available for a cover redesign. I'm talking about sales of ebook editions in 2-3 digits for works with hardcover sales in 4-5 digits. (Obviously things are different for pure e-publishers.)

2. The appearance of a book cover in a JPEG image on a screen is different from a glossy chunk of card, as you noted: this has caught me out in the past when my publishers send me covers to ogle. (The Ace cover of SATURN'S CHILDREN looks dire on-screen, but was designed for metallic ink on a glossy slip cover and somehow looks less eye-manglingly bad IRL. But I shudder to think how it'll look as an ebook.)

3. The cover illustration is a marketing tool, pure and simple. Dead tree books have covers to protect the thin paper pages, and we've taken to puting eye-catching pictures and exciting blurbs on them to make the readers pick them up. But is driving up the file size of an ebook really a good way of marketing them? Is giving a reader in an online storefront a mess of graphics that bear only a dubious relationship to the content they're searching/browsing an effective way of helping them choose what to read? I think we need a re-think here.

4. Different countries, different cover art conventions. US book design looks garish and cluttered from a British perspective, for example, but a model of staid sophistication compared to the Czech market. As ebooks become a world market (cost of shipping: zero) are book covers going to change to reflect this?

[kaigou.livejournal.com]

Not many authors have graphic design training, and only a tiny subset of those have actual cover design experience.

Which normally is fine: the majority of trad publishers have some nice in-house artists and art directors. Epubs? So with the not very much -- and that's the amusing part. Yes, they're proud of their work, but this takes on new meaning when (if I were one of their potential authors) I'd be saying, "look, I'm willing to pay for Anne Cain, anyone, just ANYONE other than your in-house team, PLEASE. ANYONE." Yes. It's that bad.

I'm not saying there should be a redesign -- if I see a book's cover in the bookstore, then that's what I'll look for in the ebook. A redesign would probably be very confusing, when the cover is part of our visual tab on a book.

Nor am I saying that tpub covers need work; what I am saying is that epubs need art directors good enough at the underlying concepts to be able to move away from taking all the art direction cues from tpubs. There's no doubt in my mind that covers are one of the most powerful marketing tools, especially for impulse buying (and epubs are nothing if not big honking impulse buys, if not the epitome).

The thing is art direction is a skill, and like any skill, you can be taught the underlying concepts like composition and typography, or you can learn it by mimicking what someone else did that worked. (This is true of many skills, it seems.) As I see it, what art direction there is at many epubs consists of someone who is not a trained art director doing their best to mimic tpub art direction -- and it just doesn't work.

As ebooks become a world market (cost of shipping: zero) are book covers going to change to reflect this?

Depends on the publisher, I suspect. Tpubs are totally hung up on selling ebooks ONLY to the country that has distribution rights, while ebooks are "hey, whereever you are, baby." But again, that's talking about tpubs; a less-savvy art director need only mimic the conventions of his/her home country.

For epubs, the home country is the entire internet, and the internet is developing (slowly but surely) its own conventions. I just think epubs need art directors to catch up with that.

Okay, not all of them. But boy do some of them badly need more expertise. Badly. Very very badly. *scrubs eyeballs*

[tiercel]

You know what drives me nuts about urban fantasy? The sexy women covers, as exemplified in your first set. I'm really, really tired of that shit - especially when the sexy woman on the cover doesn't bear any resemblance to the main character. (See the Negotiator series by C.E. Murphy, in which the main character is black and the woman on the cover is very, very white. ARGH.)

And of course, the worst covers ever, in which the poor authors got saddled with some terrible, terrible shit: http://www.likesbooks.com/covers.html

[kaigou.livejournal.com]

I think the CE Murphy example is among the most egregrious, but epubs aren't much better (and in some cases, worse). That, though, is because (as I understand it) the majority of tpub covers are using models + artists to either manipulate/overpaint a photograph or to create an entire artwork from model sketches. That's a massive investment of time and money in a cover, which epubs don't have -- so epubs work with the models they can get (hence the big use of Poser for awhile, there). It's a lot harder when you've got a smaller budget to be able to find, let alone pay, a model who fits the description just so.

Which means that I tend to give more leeway to some ebook covers (in terms of the 'that doesn't look anything like the guy/gal' syndrome). It also means ebook covers, done on a budget, must emphasis other elements to get around limited budget and model choices. Photo manipulation is one, another is limiting the amount of actual (recognizable) face seen. If you look at the ebook covers above, three don't even show a face at all. The tpub books? I'd call that three-and-a-half.

(I would have used other covers but I had those handy, and they illustrated what I'd realized. But... sexy woman on cover is pretty much tiresome on me, too.)

[autopope.livejournal.com]

There is, sad to say, a problem with racism and cover design.

It's not that the art directors and artists and publishers are racist. But, by way of an example: a friend of mine -- a British fantasy author -- had the first book in her latest series bomb really badly on release in the US. Non-white protagonist, cover with representation of said protagonist ... publisher did the marketing right, but sales were inexplicably w-a-y down on what had been expected. The UK cover, in contrast, was a lot more abstract (as British covers currently tend to be) and sales were fine, on track with her previous novels.

Obviously lots of readers did manage to buy a fantasy novel with a black protagonist on the cover. But a percentage (I can ask her when I see her next week) who had been expected to buy her books passed on that one. And there's a -- very bitter -- deduction to be drawn from this experience: that black protagonists bring out unconscious (or conscious for all I know) racism in a large sector of the fantasy book readership. (EDIT: or at some other point in the distribution chain between the publisher and the readers. Pointing no fingers here: insufficient data.)

[kaigou.livejournal.com]

The racism elsewhere is more subtle, but it's there. I've seen at least three books in the SFF genre that (from the teaser) have Asian protagonists... like the one with the Japanese protagonist but that's a Chinese model if I've ever seen one. Or the Chinese-American protagonist with the Korean model on the cover. Maybe no one else would notice, I suppose, but I do, and it throws me, and then it annoys me.

Yet another reason to argue for more abstract covers, I suppose. Skip the entire question of whether the person on the cover looks anything like the visual in my head (even when the cover art is a close-enough in the important ways) -- and don't forget the annoyance of covers that look like illustrations of fascinating scenes... that never actually happen in the book. AAAAUUUUUGGGHHHH.

[christeos-pir.livejournal.com]

What immediately struck me is how they use practically the exact same picture over and over. At least do it on purpose, people!

Anyway, there hasn't been a decent illustrator since Pauline Baynes. };->

[kaigou.livejournal.com]

Which one -- you mean the two in the center, in the first set of pictures? I didn't realize until just now that I'd selected two from the same series, which apparently has chosen to brand itself by using similar pose and same model for every cover. Then again, I think most of Louis L'Amour's covers were pretty interchangeable, too.

Just think what Pauline Baynes could do these days, with a tablet and a big honking professional version of Photoshop!