Publishers should all have a /covers directory

Here's a lazyweb idea for publishers: I often blog books here and when I do, I like to put up a picture of the cover. Normally I get these off of Amazon or Amazon UK, but it's often the case that Amazon's covers are grainy, missing, too small, or otherwise unsuitable (for some reason, an enormous number of Vertigo graphic novels have the wrong cover on Amazon).

So here's the idea: publishers should create default directories called "covers" at their server-root (e.g.,,, etc) filled with high-rez PNGs or JPGs (or both) named after the book's ISBN -- for Neil Gaiman's Graveyard Book, it would be Tweak your robots.txt file to make sure the search-engines all crawl these directories, so when you search on or for an ISBN, the publisher's high-rez would be right there at the top.

Many bloggers would just embed these images on their homepages (if you're worried about bandwidth costs, I'll personally kick in ten bucks to cover a year's worth of downloads for images hosted on S3), which means that publishers could simultaneously update their covers on (potentially) hundreds of old reviews when a new edition comes out, add banners like "New York Times Bestseller," etc. What's more, you can gather usage stats from your server logs and discover which bloggers are reviewing your books and which of those reviewers gets the most traffic.

With a nice, predictable naming scheme, this becomes a super-lightweight API. Delicious Monster and other services could automatically look up covers in the appropriate publisher's /covers directory. Indie booksellers, school librarians and other people producing promotional materials would have a canonical source. Even a publisher's own PR department could benefit from having an easily accessible, outside-the-firewall, easily accessed, up-to-date directory of cover art.

There you have it -- a practically zero-cost way for publishers to sell more books, get better market-intelligence, and get better control over the collateral used to sell their products online.


  1. “What?!? We can’t just GIVE AWAY jpegs of our covers. Maybe we should charge bloggers a few dollars each to use them!”

    “Excellent idea. Make it so!”

  2. Working for a film distributor we have the same issue with Amazon- you supply them DVD covers and they take ages to appear on the site- and often they are the wrong image or very low quality.

    Seems like a very good idea (also for our film posters) as I often despair at the ability of sites (not just Amazon) to use the correct images with the films.

  3. It’s a nice idea, but the rights that the publisher may have paid for the use of cover artwork may not apply to a freely-available high-resolution version. Remember that the artwork may not have been commissioned for the book — it may be (for example) a work of art appropriate to the content.

    Anyway, for any given ISBN there is no guarantee that the cover artwork will match the physical book. This is particularly true for libraries, which are increasingly pulling cover artwork from Amazon. Libraries often re-bind books for durability, and dust jackets are often discarded. I’ve known more than one library patron reject a book from a shelf because it doesn’t look like it “should”!

  4. Not only that – but for e-book readers, and the iTunes-like services that provide the e-books, this “lightweight API” would become the source for the “album art” for a book.

    I’m sure there are attempts to standardize e-book format – do they allow tagging like the ID3 tags within .mp3 files? Tags would allow embedding the URL of the cover art or embedding the cover art itself, possibly.

  5. This is such a simple and therefore brilliant answer to a common problem. I love how open source concepts sometimes just leak out into the ether and everyone is enhanced by it. Perhaps boingboing should have a subsite for these thought sourcing ideas – I think the audience would be an amazing resource.

    I for one would like an open source billpay platform that is secure and without fees.

  6. @7: You think that there are publishers who negotiate for art to be used in their covers whose license doesn’t include the right to distribute the art for promotional purposes? I find this incredible to the point of not actually believing that it is true.

    That would be like commissioning a cereal box and leaving out the right to give it to grocery stores to stick on coupons promoting the cereal.

    Do you know of any such books?

  7. Ideally, you’d use HTTP Content-Negotiation to support retrieving the ISBN URL with no specific extension, so that the webserver can return the most appropriate format for the client accessing it.

  8. That’s a great idea! It could be helpful for internal use at the publishers as well, where a lot of different departments may need to access the cover image at various times.

    On some servers, directory access may get a little slow with many thousands of files. Also, with the changeover from 10-digit to 13-digit ISBNs, each cover will need two files (or at least two file names).

    @7: There should be no issue here: the publisher can post the highest-res version that they do have the right to distribute.

  9. Cory, you should definitely check with Tim Spalding at LibraryThing as mentioned above. He’s working on more or less that idea, with an open API. If you can’t find contact info for him, lmk and I’ll pass along his phone # to you…I don’t think he’d mind. :-)

  10. If I were a jr. level consultant working at my company, I would somehow find a way to involve Active-X in the solution.

  11. “Anyway, for any given ISBN there is no guarantee that the cover artwork will match the physical book.” Cover art is inherently commissioned for promotion. Plus, it makes blog posts look more spiffy. Mostly, it’s a lack of people thinking to do this sort of thing.

  12. Who pays for bandwidth? What happens when a publisher goes defunct?

    Given that book covers typically fall under “Fair Use,” I think that Wikipedia might be our best hope in this case.

  13. I love the idea, especially using the unique ISBN to associate the art.

    BUT, I would prefer to see a single site hosting all the art in one place rather than having to chase down the URL for each publisher.

    This would be mutually beneficial to all book and music publishers, so you’d think perhaps an organization such as the RIAA, MPAA, American Booksellers Association, or better yet, a site with a more universal scope such as Wikipedia could provide the space.

    Then it would be a simple matter of typing a commonly known url/ISBN.png.

    And why stop at books, music, and media? Why not have an artwork site where you can get images of any product with some sort of standard code including UPC, etc.?

    You’d need a prefix if you went that route url/ISBNxxxxxx.png for ISBNs, url/UPCxxxxxx.png, etc.

  14. All the more ambitious extensions to this idea seem like good ideas to me, except that they require a lot of meetings, agreements, and discussions between competitors, likely requiring board-level approval, etc. Likely ETA: 2015.

    By contrast, a cabal of three people (webmaster, art director, head of PR) at any given publisher could make this happen tomorrow.

  15. LibraryThing does just such a thing (as the 5th commenter came forward with.) Indexed by ISBN with small, medium and large covers.

    Their URLs look like:

    Those closing numerals are the ISBN, yeah? They give you JPEGs. It’s quite complete; a surprising wealth of obscure covers are scanned. For my part, I use this to show me the correct covers when I’m sifting through used books online. There’s been some question as to the accuracy of matching ISBN with cover, but I’ve received a small pile of books through Abe Books recently and they all matched the covers I got through this script:

  16. Yes, like @5, I was going to suggest LibraryThing too…I have personally scanned covers to put in my library there — but I wonder if the images are still too small for “bloggers”? — If you search for a popular book with lots of editions you can see a variety of covers with the different ISBNs.

  17. @David Bruce Murray: Yeah, wouldn’t it be great if there was some sort of super Library of Congress site for book information? I would love to see some basic publication info for every book that exists in one easy location. Images, or lack thereof, can really make a difference in gaining people’s interest.

  18. In the interest of full disclosure, I work at Random House. It’s not as simple as a /covers directory but we have an archive system in place here:
    You can freely download cover art in a variety of formats and sizes. This system has some advantages over a /covers directory or an ISBN system: search capabilities (who can rattle off an ISBN, even of a favorite book?), format and size choices, metadata., etc. All of this is available free of charge. It’s too bad so many commentators piled on to lament the absence of such a system, when there is one in place. We want the covers freely distributed to get the word out about the books! However, publicity departments like to be asked for cover art because then they know that coverage is coming and can report that to sales, marketing, and editorial. Anyway, I hope this is helpful.

  19. Even a publisher’s own PR department could benefit from having an easily accessible, outside-the-firewall, easily accessed, up-to-date directory of cover art.

    I used to build and maintain promotional Web sites for authors, and obtaining usable cover art from publishers was often something of a headache for a variety of reasons. Having a charmingly simple system like this (or LibraryThing) would be a welcome improvement over the existing piecemeal method of distribution.

  20. This is the kind of cool, innovative idea that I like to see; practical, implementable, and extraordinarily useful. Most everyone that deals with media files feels a jolt of geek joy when the correct media art comes up for whatever file you are accessing; I remember coming up with the idea for CDDB (Gracenote?) long before it actually came online (easy to say in retrospect, but true– it was a pretty obvious idea), and how cool it was when the included Windows CD player would pull back an accurate track listing when you stuck a CD in!

    I’ve always wondered why newer formats like DVDs didn’t have specs for that info built into the media. Do BD discs have that capability? My only player is a PS3 and I haven’t had a chance to inspect the file system structure, but there is at least icon art that comes up (but is that from the Internet, or embedded on the disc?)

    I’d love to see it happen, but content providers are not likely to go for it– “but they could use those images to print out and sell copies of our books” or something else that really doesn’t make sense, but is along those lines.

    I think what would be ideal would be a wiki-style CC licensed “info-base” that covered all media and could be a jump-off point for further research for any work (books, movies, music, spoken word, etc) but I doubt that cover scans and lyrics and whatnot could be considered fair use in such a setting, and marketers would probably not like everything getting out of their control, so another good idea squashed by our ill-conceived 21st century notions of intellectual property.

    Think of it, though—being able to reference all related works by an author, director, actor, musician, or even, say, director of photography. You could browse a tree of intellectual precursors and descendants to a school of thought, maybe even buy related content. We need a consistent “semantic” web for media content that works the same for all content. All this is possible now, but the information is not connected to the medium in any consistent way, for example: the crappy nature of bonus “CD-ROM” content on DVDs. Or, even when the info is connected, it’s only a small subset and researching the bigger picture just doesn’t pay off in the long run. Providers like Amazon are about the closest thing to this kind of integration, but even there the info is not complete and is slanted towards forcing you to make a purchase.

    Anyway, thanks, Corey, for yet another forward-thinking idea!

  21. I blog about Star Wars books (published by Del Rey, a division of Random House) and the RHI Archive that Modano (#23) mentioned has been a great resource. A search is really a must – mostly I deal with books that aren’t actually out yet, and while I can get the ISBN from’s online catalog, it’s so much easier to just do a search on ‘star wars’ or an author name – sometimes just at random to see if anything new shows up.

    Definitely bookmarking the LibraryThing project, though. Never know when such things will come in handy.

    @Futurenerd: Great idea. Sometimes you might not want to use the current cover, but the one you most identify with the book. (Jezebel’s Fine Lines, for instance.)

  22. This would be great for academic libraries as well. We have to pay for a third party service in order to get cover images in our catalog. This would remove the need for that service as we could just set up a bit of java script on our library server to grab these images.

  23. I must say this is a good idea, a pretty good one, but I would even thank more a CD cover site.

    Yeah, iTunes can give you cd covers, but not all them are there; and usually the ones you can find on google images are not the right ones, or are not well scanned.

    I collect books and music, and I would love to have access to all my music covers without having to use the scanner…

  24. posted by edwinx2:

    I’ve always wondered why newer formats like DVDs didn’t have specs for that info built into the media. Do BD discs have that capability? My only player is a PS3 and I haven’t had a chance to inspect the file system structure, but there is at least icon art that comes up (but is that from the Internet, or embedded on the disc?)

    DVDs can include a (short) title, and an icon for the disk.

    In my experience, perhaps half of mainstream DVDs include the movie title in a format my changer will pick up, and even fewer include a useful image.

  25. I think this is brilliant, and I can see how tying it to the publishers websites will provide useful feedback for them. If they have access to data that tells them what books are being talked about, it could help in getting books reprinted (or any number of other reader-friendly outcomes). I like that RandomHouse’s website lets you search by ISBN or author or title, but I think that if I’m doing a google search I’m looking for the ISBN, because I’ll probably have the book handy. Definitely gonna bookmark the RandomHouse website and the LibraryThing.

  26. As a book jacket designer, I find this idea intriguing. However, as for “publishers could simultaneously update their covers on (potentially) hundreds of old reviews when a new edition comes out” troubles me to no end. There are good covers and then there are bad covers. Add an associated archive, I’ll hop on board.

  27. The RH database would be easier to use if it let you search by exact title. It seems to “or” words so if the title has a common word, you get a bazillion results. And no author search. (There probably are workarounds, but I’m too tired tonight to read a huge helpfile.)

    I wish there was some mass database with the kind of metadata that would let publishers avoid the copycat covers you see so often – the same stock image slightly photoshopped. It happens all the time and it’s just annoying as hell.

    LibraryThing rules. Tim Spalding is God. But nicer.

  28. @10: It’s nothing to do with promotion, it’s more to do with the fact that you’ll essentially be redistributing a high-resolution image of the artwork for free. If all you’re doing is providing a low-res thumbnail then fair enough… and this should be good enough for most applications.

    I can’t name a specific book because AFAIK this has never been tested in court, but I’d be willing to bet that owners of works who grant permission for them to be used as cover artwork wouldn’t extend that permission to unlimited distribution of a perfect digital copy of the original.

  29. This is a good idea, but I suggest a minor twist. Instead of /covers/[ISBN], it should be /isbn/[ISBN]/cover.png

    This way, you can add other things about the book (sample.pdf? reviews.txt? authordata.xml?) and they will be easy to discover: just go to /isbn/ and look at what’s there.

  30. @40: You’re confusing the image of the book cover with the original artwork that was used to create that book cover. And there’s a world of difference between the resolution required for decent print reproduction and the resolution of even a large cover image on the web. We may use a 30-meg scan of a piece of art to create a book cover, and obviously we’re not going to distribute that scan. By the time it’s part of a large cover image for web use, the resolution is 1/4 to 1/10 of that original scan, and often has text overlaid.

  31. Another post from Random House here — it seems that Boing Boing is popular over at RH.

    S3, eh? Remember, RH sells direct, so Amazon is a competitor. We would have to be very disciplined in order to avoid leaking referrer data between our Amazon hosted covers and our e-commerece servers and I don’t want to have to to worry about that.

    For the record, we have covers in three places, and we’re not blocking anyone from including them in there blog posts. They are:


    Of course, each of those start with “”, and you can use either a 10 or 13 digit isbn, with or without hyphenation. Respectively, the folders contain images 150, 100, and 450 pixels tall. If you want higher resolution images, I would recommend using the teams search engine link posted earlier.

    In addition, many of our titles are enhanced with an embeddable widget which allows users to view some pages of our books. If you see a “Browse and Search” widget on a book’s page, click the “+” button to reveal html code which can be used to embed the widget in your blog.

    And one final choice; Our dynamic image resizing tool is available to give you a cover in the size you want.


    “height” can be replaced with “width”, and “png” can be replaced with “gif” or “jpg”.

    I might think about putting a covers directory on the root if some sort of standard emerged as to what sort of file should be placed there, but our doc root is already pretty crowded and our rewrite rules rather extensive.

  32. We run a small web design business and my partner has just blogged about how useful we would find this (see

    Our experience has been that publishers are obviously happy to let us reproduce book covers in their entirety, but we’ve been told to change web pages where we have included a detail from a cover. This is how, I presume, they can protect the original art work.

  33. Cory D has a good point. Over the years, my own books have gone into various editions and appeared in different media and Amazon not only generally get the covers wrong, mixing up movies, graphic novels, books, records and audio-books, they even get the categories wrong. Several of my sf/fantasy titles, for instance, have been categorized as ‘military history’ while literary novels and thrillers turn up labelled as ‘sci fi’. In my experience they are NEVER responsive to attempts to straighten them out. The frustration is considerable. Publishers don’t seem interested in backing me up, sadly. The only way I can protest is to use any supplier of books and movies BUT Amazon for my own reading and viewing. I’d rather not be so negative.

  34. Hi All,

    Great suggestion. I’m the Web Editor at a small, independent book publisher in Vermont, Chelsea Green Publishing. We’ve been publishing book about sustainability and green living since 1984.

    Per this suggestion, I just set up a /covers directory on our web site.

    Thoughts? Improvements?


Comments are closed.