The Yaruki Zero book is both a thing I wanted to write and kind of an experiment in learning self-publishing for other, more serious projects. The plan was to put the book up on just about every POD and ebook sales service around, and this is a rather long and detailed journal of what I went through to make it happen, both for my own reference and to help anyone else who’s thinking of trying this stuff out. Later on I’ll see about putting together a follow-up post on how things are working out 6 months or so down the line. If you’re interested in the actual book, check out the Yaruki Zero: The Book page for links to all of the places I have it available for purchase.

Production
I started on the book in mid-January, and had the first draft more or less finished after about a month. That was mixture of copying over existing work and just plain putting most of my free time into writing. I got C. Ellis to do up artwork for the cover, Clay Gardner to do graphic design for the cover, and Ellen Marlow to edit the manuscript. (Also, about half a dozen other people read it over and offered feedback.) Each of these was pretty straightforward, especially since the people I was working with really get my general sensibility. I’ve collaborated on a bunch of stuff with C. Ellis before, Clay just has a way of not only reading my mind but coming up with the stuff I didn’t know I want, and Ellen’s fandom and writing sensibilities line up with mine quite well.

For this book’s interior I decided to keep the layout pretty simple, and just did it up in Word with some use of fonts and such to make it a little fancier. I’ve never been able to figure InDesign out myself, but while Word produces reasonably okay layouts, there are certain things where it just kind of loses its mind. Thankfully I wasn’t trying to do, say, multiple columns for this project. The big thing that I managed to fall afoul of was getting the margins right. Being inexperienced with doing stuff for actual books, it hadn’t occurred to me that the inside margin would need to be wider so that the printed material wouldn’t disappear into the binding. With POD you’re much more beholden to the printer’s specifications,[1] so it pays to read those carefully (though it wouldn’t hurt for them to make those a little clearer than they are). That was how the page count of the book jumped up to 217, and I wound up doing a bunch of small tweaks to fix things.

Print On Demand
Lulu was the first major POD service, and while there are certain glaring flaws with how the site works, it’s not too hard to use on the publishing end, just a little clunky. It’s pretty easy to pick your options and get stuff set up. The big issue with Lulu is that the site is just kind of terrible when it comes to letting customers browse and find stuff. As of this writing if you search just for RPG books you get over 1300 results, in no particular order. You also can’t change the number of results per page, so if you’re feeling adventurous enough to sift through all 1300+ entries you’ll have to go through 130+ pages. There’s some wonderfully bizarre stuff to discover there (The Mister Lincoln eXperiment for example), but by and large if you put something up on Lulu I think it should be because you’re driving traffic to your listing there yourself.

Amazon runs its own POD service called CreateSpace. It has certain major advantages over Lulu. One of the big ones is that the backend will check your submitted files for problems (though running it takes a few minutes). Also, since it’s run by Amazon the stuff you publish will go on Amazon. While the selection of items on Amazon is ludicrously large, it’s also vastly easier to browse and navigate than Lulu. Helping customers find the right products is tricky, all the more so if you’ve literally got millions of items, and I think Amazon pulls it off. On top of that, going through CreateSpace offers Amazon shipping options, which are very likely to be cheaper than Lulu’s for your customers, or even free for those with Amazon Prime. CreateSpace’s printing costs are also significantly lower, which means cheaper proofs and better royalties, plain and simple. They do have a basic approvals process before you can even order a proof, though my experience was that this was very quick and painless.

When the CreateSpace proof came I was really impressed by the quality. What looked like too-small type and cramped pages on a computer screen looked great on paper, and the whole thing generally looked like a real book. I had to comb through it again to check for errors again, with a pen in hand to note stuff down. The next day the Lulu proof came, and while it turned out that the margins weren’t as bad as I’d feared, the differences were pretty striking. The cover especially had duller colors and less contrast in the Lulu version (which is kind of a big deal when the cover design has pop dots in two shades of blue), and the interior paper is of an off-white color that I would not have chosen myself. The overall interior printing quality is pretty comparable, but my mistake of making the margins too narrow made the Lulu proof harder to read, on account of the inside of each page getting a little bit obscured by the binding.

I put only three illustrations in the book, and they’re kind of random and just because I felt like squeezing them in rather than for the purpose they serve. Nonetheless, in the initial proofs they came out looking grainy, and it took me a while to figure out why. The default PDF export settings downsample all images, so my 300+ dpi illos were going to the printer basically at web resolution. (Conversely, for the ebook I had to actually downscale the images so they’d show up more smoothly on e-readers.) I think tweaking the PDF export settings helped, but it still wasn’t where I would’ve wanted it.

Ebook Conversion
I own an e-ink Kindle (a 3rd-gen specifically), and although I’m not about to give up paper books, I am actually pretty fond of it. There are books that I don’t feel any particular need to own in print (like when I decided to read up on iPhone app design, exactly the kind of book that’s guaranteed to become obsolete), and sometimes I just like the instant gratification of being able to download a book and sit down and read it in a format that doesn’t shove light into my eyes. PDFs are kind of a mixed experience for this purpose, because my Kindle is pretty phenomenally awkward for reading anything that doesn’t fit nicely onto its screen. For PDFs of RPGs I basically can only read games formatted for digest size or smaller, and then they can’t be with too small print. Tablets can handle PDFs a lot better, but for e-reader users a reflowable native file format generally provides a better reading experience than a PDF. The only thing with that is that converting files to e-book formats is kind of terrible, and I suspect that if someone manages to make a painless way to do that they might just get rich. (Instead, there are lots of services that will be happy to charge you $60-150 or so for professional ebook conversions.) Aside from PDFs (which these days Word and for that matter OpenOffice can now spit out with basically no trouble at all), the major ebook formats are epub (used on the Nook and Apple’s iBooks) and mobipocket (used on Kindle). Both are XML formats, and although going from epub to mobi is pretty simple, converting documents to epub is kind of an arcane art.

Since my friend Steven Savage has self-published several books[2], I turned to him for advice on how to go about it. He recommended using LibreOffice to produce a file that could then go through an ebook converter program called Jutoh. Trusting his advice, I went ahead and spent the $39 to register Jutoh, plus 99 cents to get the “Creating Great Ebooks Using Jutoh” Kindle ebook, though it’s also available on the web for free. They do a good job of laying out how to use Jutoh and include good tips for non-obvious stuff about them. As things stand my verdict is that Jutoh is well worth it, and superior to the likes of Sigil and Calibre for its intended purpose (though Calibre is great for doing quick and dirty conversions for your own use). For me Jutoh had a little bit of a learning curve, and the accompanying book was invaluable for sorting out that stuff. Its big strength is that it has a good editor that lets you more easily do all the tweaking that you’ll inevitably have to do.

I found that Jutoh actually supports Word docx files fine, so I ended up skipping the LibreOffice step of converting to an odt file. I did however make a new docx and go through to edit it to minimize elements that would either be difficult for Jutoh to handle or just not come through in the ebook. One of the big ones was that I converted all of the tables in the book into lists, since the developer basically admits that Jutoh’s handling of tables is a kludge at best. I also changed the base text to 12 point Times New Roman for compatibility purposes (it had been in 10.5 point Goudy Old Style) and the headings (which used a couple different fancy fonts I like) to Arial. Apparently text you put in 12-point type shows up in the e-reader’s default format, and anything above or below that will be proportionally bigger or smaller. My book also has lots of hierarchical headers, and I wanted the chapters and articles to have page breaks after them, so naturally I went through and added page breaks. Except, ebooks don’t support those kinds of page breaks, so I had to go back and remove the page breaks (which uses the same command as inserting them) and split each section into sub-sections (basically you split once, then cut the second section, then use the Paste as Child command, and finally go and split the child section as needed).

The ebooks also needed proofing, and for a different set of issues than the print book. A few general errors slipped through in the conversion process, and since page numbers aren’t a thing for ebooks I changed those into cross-reference links that needed checking. There was also the matter of the moving target of multiple devices, so I was checking things on my Kindle and iPhone, plus I borrowed a friend’s Nook and asked a few other people to look it over. One thing I quickly found was that anything in bulleted or numbered lists was simply not displaying on the Kindle or Nook (everything was fine in iBooks though), and I ended up just changing all the lists into paragraphs to get stuff to display correctly. I skimmed the book with the Nook (which is quite nice BTW) highlighting things as I went… and then found that it hadn’t saved my highlights for whatever reason. Another round of checking on my Kindle (which did remember the highlights at least) and I decided I was done, or at least sick enough of looking at the thing to pretend.

Storefronts
That left the matter of getting things up on various sites for sale.

Getting the book listed on the Kindle store was just a matter of uploading the mobi file and entering a few things. You have to wait for approval, but for me that happened in a matter of hours. Once CreateSpace approved my new upload it was up me to proof it and give final approval, which I did with the online proofing tool. The thing that they had not advertised quite so readily is that while products show up in the CreateSpace store immediately, the site told me it would take another 5-7 business days for my book to go up on Amazon. Of course, when I searched Amazon for it the next day it turned out that it was already available, and a day later it had connected the two listings on Amazon.

Lulu has a couple of issues with how it handles ebooks that you’ll want to know about. The big one is that while it has an option for expanded distribution of epub files to get them into the Nook and iBooks stores, you can only actually select said option when you initially create the listing for the ebook. (That they do this is the major reason I put up with Lulu for doing the ebook version in the first place.) I ended up deleting my ebook listing and redoing it because of that. It also lacks the ability to have one listing with multiple formats. Although I’d have happily provided the PDF and epub (and mobi) to people all at once, on Lulu they have to be separate. It also has a rather stringent epub checker, and I had to fiddle with it in a bunch of different programs before I discovered that tweaking the table of contents settings in Jutoh did it. I used this EPUB Validator site to check the epub many times over by the way. I was very much not in the mood to deal with another set of submissions to Barnes & Noble’s PubIt! and Apple’s iBookstore (and they’re lower priority anyway), so the part about Lulu submitting to them for me automatically was one of the few things in its favor. From what I’ve read B&N takes 1-3 days to approve submissions, while Apple can easily take 2 weeks.

For DriveThruRPG I decided to take after Green Ronin’s publication of Hobby Games: The 100 Best, offering a single inexpensive package with DRM-free PDF, epub, and mobi files. I find DTRPG kind of mixed in terms of the ability to browse, though certainly far better than Lulu, but it’s pretty much the definitive RPG ebook sales site, and for example Clay found that sales of OVA were pretty reliably consistent through there, even when he wasn’t doing any marketing to speak of. It apparently has a body of customers who are willing to browse and try stuff out, which is kind of awesome. One Book Shelf (parent company of DTRPG and several related sites) also has POD printing. At this point DTRPG’s POD offerings basically amount to a massive amount of White Wolf stuff and a smattering of indie stuff. The backend for publishing stuff on OBS isn’t as polished as Amazon’s, but it’s got good functionality overall. OBS’ actual printing is through Lightning Source, and when you click on the link to get a template for the cover it takes you to LightningSource’s page for such, where it will only generate it if you give it an ISBN, something that the DTRPG interface tells you that your cover must not have. My submission got approved within a day or so, but there was one aspect of it that was kinda dumb. You can only pay for proofs through your seller balance, and if you don’t have a balance you have to buy a gift certificate and transfer the funds to your seller balance and use that. Not big problems, but something that needs smoothing over. On the plus side, the approval went through very quickly.

Update (3/16/2013): Comparing Proofs
From when I ordered it to when I got it, the proof from DTRPG took about a full week to arrive. I was happy with the quality of the printing, though the paper and the cardstock of the cover were thinner than either CreateSpace or Lulu. Not a deal-breaker, but not what I would’ve gone for myself. As you can see in the photo below the cover of the DTRPG one was even brighter than the CreateSpace one, and the Lulu version still looks just awful. The interiors are pretty comparable, though the pictures look a little faded in the CS one and the Lulu one has that cream-colored paper.

Posting about this whole thing on Story Games led to me starting a thread on Comparing POD Services, where Johnstone Metzger posted about his very different experiences self-publishing his book The Metamorphica (which is really neat, incidentally). CreateSpace’s cover printing apparently doesn’t handle a mostly white cover with subtle pencil art terribly well, and the shipping costs are surprisingly bad for Canada. (On the plus side, Lulu’s sales/revenue tracking page is actually very easy and simple now.)

I really don’t know what I can even do about international shipping. It’s getting to be more of a problem, especially since prices went up on January. For Channel A, Asmadi literally had to charge $25 for shipping, effectively doubling the price for people in places like Europe. To some extent that’s something our European friends have to be used to by now (it took a while for my jaw to stop hanging open whenever I read the prices of video games in Euros or Pounds), and pretty much the only thing I can do on my end to ameliorate that (aside from making e-books in the first place) is to get a local retailer/distributor who can buy in bulk and sell at a more reasonable price (like when Leisure Games in the UK made it drastically easier for European fans to get Maid RPG).

It’s definitely premature to draw conclusions from a whole two weeks worth of sales, but so far customers overwhelmingly prefer CreateSpace for POD. On the other hand e-book sales through DriveThruRPG are taking the #1 spot for my book. It’s hard to say how much is that it’s the venue focused on the target audience and how much is the fact that it lets me offer the e-book solution with the best value, but it is the clear winner so far.

Conclusions
I’ll be looking forward to trying this out with a novel, or anything else with less formatting elements to worry about. While I would like to see more RPGs available in epub/mobi formats, I have a hard time blaming anyone who doesn’t want to go through the hassle of making them. (I’m thinking about it for Golden Sky Stories anyway, but we’ll see.) I’m glad I wasn’t gunning for a set launch date for this book, because the disparate approvals and other delays basically made it impossible to get the different versions at the same time. Lulu lets you do it instantly, Amazon has a short delay, and with iBooks it’s whenever Apple gets around to it.

CreateSpace is excellent, and I think I’d like to see more RPG people discover it. As POD solutions go, it provides good service and more importantly it gets over the hump of high shipping costs that are a problem with a whole lot of other POD services. Having the price a customer pays go up by 50-100% because of shipping kind of sucks, and overcoming that is undoubtedly a big part of why Amazon is so successful.

I don’t think I’ll go quite as crazy with redundant sales venues in the future because it generates a good amount of extra work, especially since no two places seem to use quite the same format for covers, even for ebooks. The biggest thing is that, although it’s hard to say how much is my own talents, how much is the technologies, and how much is the people I can ask for help, making a book wasn’t too that hard. I do have some skills that a lot of people lack,[3] but at the furthest extreme you could write up a book in OpenOffice, use a well-chosen cheap or free image for the cover, and get a book up for sale with relative ease. The results of this democratization can be terrifying at times (as the Lousy Book Covers Tumblr amply demonstrates),[4] but for people with talent and dedication the tools have really arrived in a way they hadn’t ten years ago.

This blog post and a little bit of social media stuff will be about the extent of my marketing for this book, since this project is aimed at a niche audience and had practice for other things as a major goal. It’ll be interesting to see how it goes from here on out, but I’m pretty happy to have finished something and put it out there.

[1]My limited experience getting quotes for Golden Sky Stories was that although traditional printers require larger print runs, they can actually be very flexible on a lot of things. Getting GSS printed in a funny Japanese size with a dust jacket on a softcover book is actually doable, and not all that expensive.

[2]Mainly his geeky career advice stuff, though since he’s the guy behind Seventh Sanctum I’m hoping to get him to collaborate on a book of random tables.

[3]I know so many people who are vastly superior to myself at stuff like Photoshop and graphic design that it’s easy to forget that I nonetheless have the benefit of an A.A. degree’s worth of training in this stuff.

[4]On the other hand, the depths of CreateSpace and the Kindle store do turn up works that, if nothing else, have an amazing, raw sincerity to them. Sometimes they seem to be a lot like the book equivalent of the song-poems that Penn Jillette finds so fascinating.