Category Archives: Can’t Make Money

Some Catching Up to Do…

Several things have been happening these past few weeks that the blog was down, all of which would normally deserve their own blog entries. With several things coming up in the next month or so, however, I can’t afford to give them all separate blog entries (though I’ll try). So I’ve got to compress all of these items into one blog post just to catch up.

To begin with, This Book Cannot Make Any Money was released in eBook form, as expected (and the print edition was revised). It’s $2.99 (it IS a full-sized book, albeit tiny compared to my normal books), but it is free with Kindle Unlimited.  Reviews would be appreciated!

I may still write the originally planned full blog entry about the process of releasing it, as I mentioned in my last blog, but that won’t be until next week at the earliest. Or I may decide that I’ve done enough on this blog series and be done with it, because I’ve got a lot of other things I hope to cover over these next few weeks.

Second:  Back in November, recalling how long he needed to complete cover art in the past, I contacted Alex Kolesar to see if he could start the cover for the next Law of Swords book, hoping to get it by Ravencon in April.  However, he was unavailable… and, worse, he would no longer be taking on freelance work of any kind.  So… I needed to hunt down another artist.  And it couldn’t be just any artist, but one whose style was similar enough that I wouldn’t have to re-do the first to covers to match.

Found one.

The new artist is Hans “Hanzo” Steinbach, a freelancer currently working with Udon Entertainment.  Some of his past credentials include concept art for big-name video games such as NieR: Automata, costume designs for Capcom’s Street Fighter 4, and general concept art, character design, and illustration work for Tokyopop, Emerald City Games, and Boom! Studios.

And while I’d given up hope of getting that cover by mid-April, I’ve already received and approved his artwork; he works fast.  Again, showing off that artwork deserves its own blog.  I may wait to debut it at Ravencon, next month; we’ll see.  (I’ll note that I still only have a working title for Law of Swords Book 3, so even then it may not be finalized).

Third:  I have a release date for The Merrimack Event‘s audiobook:  April 3rd.  Tantor has even put up a sample of the reading, which you can now listen to from their website.  You might want to go listen.

I will note that, while they have the cover art I bought from Joel C. Payne showing, the final version of the Audiobook will have a different cover on it (this is necessary:  The standard size of an audiobook cover is VERY different from an eBook\Print cover, so it would still have needed to be re-done by someone; Tantor will want to work with someone from their regular bullpen of artists to get that done, so it’ll be done by someone else).  I’ve seen a mock-up of the new audiobook cover; thematically, it looks very similar, but the ships are different, the colors are different, the fonts will probably be different, and it looks like they want to show the same scene as the current book cover, but from a different angle.

Finally, I know I’ve mentioned this before, but I’ll be at Ravencon, which runs from April 20-22nd.  I’ll be giving a reading from the upcoming Law of Swords book, showing off the new cover, and dispensing what passes for my wisdom from several panels (how many?  I don’t know yet.  I haven’t received my schedule.  I signed up to do eight panels, but we’ll see how many I get).  Once I get my schedule, I’ll be doing (as I did last year) a few “Ravencon Panels (I’m not on)” posts, as well as a few “Ravencon Panels (I did do)” posts after the convention.

I THINK that’s it, but it’s been more than a month since my last real post, and a LOT has been happening.  If I remember something else that happened, I’ll be sure to let you all know.

Until next time!

This Book Cannot Make Any Money: Ebook Editions…

Normally, I try to get the eBook out first. With “This Book Cannot Make Any Money,” however, the print edition had a deadline (Marscon) but the eBook edition didn’t, and so I never even started an eBook edition until after the print version was complete. In fact, as I start the eBook version, the ONLY editable file with the final version of the completed book is the one I built in Scribus to produce the PDF. And Scribus can’t compile ePubs (yet; it’s on their ‘to do’ list, according to some of the things I’ve seen in their forums).

So I’m going to use Sigil to build the eBook with. Now, Sigil is most useful for editing pre-existing ePubs, such as those produced by Scrivener or InDesign. It is CAPABLE of producing ePubs from scratch, however… and that’s what I’m going to have to do here… yet as I’m writing this, I have no idea how to begin.

Fortunately, there are several tutorials out there — available both in written format and in video format — which will certainly help.  The problem is, every tutorial I’ve found has had completely different, and frequently incompatible, methods for how to get things done, and yet none of them are telling me everything I need to know.

Even with Sigil, I need to learn a bit of how to code in Cascading Style Sheets to get everything done I want to do.  I’ve never worked with CSS before, however, and I started building this eBook not knowing anything about how to code CSS.

A few days later, and I still know ALMOST nothing about coding CSS, and the process I eventually settled on was tedious and occasionally frustrating, but I’ve figured out enough to put that darned eBook together.

Now, it’s taken me a bit longer than I’d hoped to actually put the book together, but it only took me one day of, quite frankly, fumbling around blindly, guided only by mismatched tutorials, to learn what I needed to get started.  It’s a bit daunting, but it’s doable, and it doesn’t take THAT long to learn the basics.

So, is it practical for someone, with zero previous experience and zero budget, to learn how to make their own eBook?  Well, I’d say “yes,” but then again people have been eating Tide Pods lately, so… yeah.  I would say that it is doable for MOST people, however, if they’re willing to spend a day or two mucking about trying to figure things out.  You don’t need to learn all there is to know about coding .css, you don’t need to know everything about book design, etc., you just need to know enough to get YOUR book ready.

When the eBook is done, I’ll test it on my Kindle (as well as on other devices and through whatever .mobi reading software is available that I can find) before uploading.  I suppose that will be cheating, a bit (if I’m not allowing myself to use things like Microsoft Word for this challenge, I shouldn’t let myself use things like my Kindle, either), but the idea at that stage is to see the eBook across as many devices and platforms as you have access to.

But I’m not at that stage, yet.  I’m about half-way through the eBook conversion process… so I guess there’ll be one more blog on this topic, after all.  Next week (hopefully), the book will be done, and I’ll discuss setting up and uploading a book to Amazon.

This Book Cannot Make Any Money: Book Design (Print)

As a self-published author, I need to take on a number of job titles.

I’m the author, which means I write the books.

I’m the managing editor, which means I am the ultimate word on whether a book is published (you have to know when to reject your own books), I select (and hire) which editor goes with which project (yes, that is precisely what hiring a freelance editor is), and I am the one who has to decide when a book is edited well enough to publish.

I’m the Chief Financial Officer, or CFO.  I set and manage the budget, do the accounting, deal with the taxes, etc.  (Yes, even if you hire an independent tax preparer, you are dealing with the taxes).

I’m the Art Director.  That means I select (hire) the artist to do each cover, make branding decisions regarding artwork (for example, I always use Yataghan as the title font for my fantasy novels), and try to direct the artist to produce the best cover I can get.  Frequently, this includes doing some aspects of the cover design, myself (I usually commission the front cover artwork and do the other aspects of the cover — such as placing the fonts — myself; however, I did let Joel C. Payne handle the font placement for the cover of The Merrimack Event, when he said he wanted to).  Art Directors are often also in charge of hiring the book designer, but more on that, later.

I’m the Marketing and Social Media Director.  I handle my business website (even if you hire it out, you’re still responsible for picking WHO produces your website), this blog, my Author Central profile, my Goodreads author’s page, the mailing list, my Facebook account, my Twitter account, and any of the smaller, newer social media platforms I hear about and might want to try out.  I’m also responsible for getting any promotional materials made (I’ve had two business-card sized and two post-card sized cards promoting two of my series prepared; I’ve also purchased, but have not yet deployed as a marketing tool, a bunch of keychain-sized plushy Fennec foxes for Fennec Fox Press).  And I’m responsible for hiring any advertising services (I can’t really say I’ve tried one that has been very effective, yet (a few which appear to have earned back a little more than I spent on them, but nothing with huge results yet. I’ve also never done a serious price promotion, which may have something to do with that).

I’m the Human Resources director.  That I have no employees (only project-by-project freelancers) doesn’t matter — it’s still my job to my job to scout out all of the people I might hire as a part of any of the jobs above (or below).

And, in my capacity as Art Director, I’ve also picked myself to be the Book Designer.  Even when I do have a budget of more than $0 to work on my book with.  I’ve made a couple mistakes in the past (I still don’t like the font choice I used in In Treachery Forged; it makes the book look like it should be a large-print book when it isn’t, and some of the punctuation looks off), but I’ve never committed any of the “Top Five DIY Layout Mistakes,” and I think I’ve gotten better with each new book release.  So now, by my fifth book (and having collegiate training in the field, albeit years before that first book came out), I have enough experience I can say with confidence that I know what I’m doing with this job.

But, uh, I don’t have any prior experience with Scribus, which I was forcing myself to do all the book design in.  And that proved to be something of a problem.

Now, all of the print book design software I’m familiar with — InDesign, Microsoft Publisher, Quark XPress, etc. — operates the same way:  You create a set of master pages that account that place the page numbers, account for the margins, etc., you use guidelines to show where you want titles placed, etc.  That allows you to make text boxes that “snap to” those grid lines\margins\etc. easily, making your pages consistently uniform.

Theoretically you can do that with Scribus, too, but I had a dickens of a time getting my text boxes to “snap” to the appropriate guide lines.  There was a learning curve, as everything seemed to work in a different way than it does on InDesign or Publisher (I haven’t used Quark recently enough to know if it compares).  I goofed up a few times (there was one section — fortunately caught before I sent it to the printer — where all of the pages were scrambled).

In general, that was my experience with just about all of Scribus’s features.  I could still do most of the things that I was able to do in InDesign (though there are a few InDesign features I haven’t used that weren’t available, such as compiling the book into an .epub eBook), but the methods used to do them were more error-prone, the options weren’t as great, and the controls were different.  It made putting the book together… frustrating, to say the least.  And I got really concerned about what the end result would be.

But, well, I got it done.  If I never had access to InDesign or Publisher again, I think I could live with designing all my print books with Scribus (I wouldn’t like doing it as much, but I could make it work).  And I’ve never used Scribus before I started this project, nor were any of the controls (that I could find) the same as on InDesign.

In other words, even if you have never designed a book before, and don’t have any money, you can (eventually) teach yourself to use this software well enough to do your own book design, just like I had to.  (You just need to learn the basics of book design, which you can do by studying how other books in your genre work.  And if you don’t own any other books in your genre, well, that’s what the library is for).

This (print) book didn’t end up looking as good as I’d hoped, but that wasn’t the software’s fault.  It was KDP Print’s documentation’s fault.

Basically, the problem comes down to the margins.  You need extra margin on one side of the page (the side connected to the spine) so that people can see the whole page when they open it, called a “gutter” margin.

My books usually are in the 400-500 page range, which KDP Print recommends 0.625″ in margin for.  I usually give myself a little margin of error and set it up with 0.75″, and it looks fine.

I wasn’t sure how long this book was going to be, but I estimated it would be between 100-300 pages.  So, I checked the guide, and it said I only need 0.5″ of gutter margin for the outside of that length, and it turned out to be less than 250 pages.  So 0.5″ should be plenty.

But at 0.5″, I’m not happy with the look of the gutter margin.  If it was an easy fix, I’d go ahead and change the margins so that it looked better… but it’s not an easy fix.  Changing the margins changes the page count, changing the page count changes the spine width, changing the spine width means changing the cover, etc., etc.  In other words, I would basically have to start over from scratch.

While not IDEAL, the current gutter is serviceable.  I had a few people look at it, and they thought it was fine.  So, it’s just my own sensibilities as to whether anything needs to be changed, and that’s not worth taking a book that’s on sale off sale for the several weeks needed to make the fix.

Had I been able to order a proof copy instead of having to put it on sale, first, to get my own, I might have changed it.  That is, after all, what physical proof copies are FOR — to catch things that don’t (necessarily) show up in the digital file before the books go on sale.  (Then again, you do have to pay money for a proof copy, so I suppose it would have been cheating, as far as the “rules” set out for this book go).

At any rate, the print version is now available… but, unusually, I still have to do the eBook.  Again using a piece of software I’m largely unfamiliar with.  Which means the next entry in this blog series is going to be… fun.  Heh.

This Book Cannot Make Any Money — How the Cover Was Made….

Well, I’m recovered from the convention (I didn’t get con crud, but I did need some time to get over things) and ready to resume the “This Book Cannot Make Any Money” blog series.

As I’ve mentioned in earlier posts, I’ve already managed to release the print edition of “This Book Cannot Make Any Money.” It might still be further revised; I’ll go into this more on my next blog, but I found at least one issue with the print edition that I would have fixed if I’d been able to get a proof copy first (a week after I pushed it out, Amazon announced that all KDP print authors could now purchase proof copies… just that little bit too late. Sigh), but I don’t think there will be any changes that will affect this blog post.  At least I met my goal by getting the book out in time for Marscon.

So, when we last left this blog series, I was talking about the cover.  I had a plan to produce an “illustrated” cover by blending clip art elements to put together something resembling a custom illustration. Those of you who’ve seen the final result know that I didn’t end up with anything resembling that plan, but let’s discuss what I attempted, why that didn’t work, and what I ended up with.

To start with, I had to figure out the dimensions.  Normally, when I have the budget to hire someone, I only commission the front cover (which is all you need for ebooks) and slip that artwork onto a solid-color background to make the full wrap, so all I need to know for the cover artist is the trim size.  Because I was STARTING with a full-wrap cover, however, I needed to know a bit more than that, first.

I decided to make this a 5″ x 8″ cover, which is smaller than my typical trade-paperback standard-sized 6″ x 9″ book.  I wanted to be sure the book was thick enough to have legible text on the spine, so a smaller trim size made sense.

Because I was starting with a full-wrap cover, I had to build the whole book to find out the actual page count (which is partly what derailed this series, making it more re-cap than follow-along like it originally was intended to be).  In the end, I was able to determine that my book (factoring in not just the text but the front-matter, aka the copyright page, title page, acknowledgements, table of contents, etc. in the front, and the back-matter, aka the advertising in the back) was 234 pages.

Per the formula on both the Createspace and the KDP Print page (I am linking to Createspace because its bookmarked, and at the moment it’s the same for both printers, but it may change in the future; more on that, later), the way to figure your cover width is Bleed + Back Cover Trim Size + Spine Width + Front Cover Trim Size + Bleed.  Similarly, cover height is determined by the formula of Bleed + Book Height Trim Size + Bleed.  Bleed is a constant (0.125″).  Trim size for front and back are always the same (in this case, 5″ x 8″).  Spine width, though, requires you multiply the number of pages by the paper width, so I had to pick which paper I was going to use.

I had an option of white paper or cream paper.  The choice is largely aesthetic, but the cream paper is a little thicker (by some thousandths of an inch) than the white paper.  Picking cream paper (again, I wanted to thicken the spine, but there are other reasons you might pick one over the other), I multiplied the 234 pages by the 0.0025″ paper width to get 0.585″ for the spine.  Please note, different printers calculate spine differently, and they will give you their own formula for determining it.  Sometimes, a “page” refers to one side of the page, as it currently does with Amazon’s two print services; sometimes it refers to both sides of the page (front and back), so I would have had to multiply the paper width by 117 instead, and the paper might have been a different thickness altogether.

Sometimes, the printer may change the page width on you, if they get a different paper supplier.  So, READ THE DOCUMENTATION your printer provides, EVERY TIME you build a cover.  The documentation will also give you such information as how much bleed you need for the cover, how much of a margin you need to give between the edge of the page and the “live elements” (meaning the text and the important parts of the artwork), etc., etc.

Now, I had a concept for the artwork I wanted to put together.  For an unrelated project, I had — entirely with free clip art, but using Photoshop Elements — been able to reproduce different versions of the British Royal Coat of Arms (for purposes of comparison and contrast).

Given what some of the stories were in the book, my plan was to reproduce one of the simpler coats of arms from this project, but instead of using lions and fleur-de-lis I would use other, thematically appropriate silhouettes.

Things started out okay.  I found a decent shield design:

I centered it on a background and colored it appropriately (leaving what I thought was enough room for the title and my name.  I did it using a piece of graphic software I’d never used before, simply by figuring out how to duplicate layers, use the colorify tool, and use the cropping tool.  I made it in GIMP (which was a pain, as it operates VERY differently from graphic programs I’m used to, but I eventually figured it out), and came out with this:

Problem:  The image in my head included such thematic elements as robots, Greco-Roman helmets, a space station, and crossed swords (two sci-fi elements related to the sci-fi story fragments, two fantasy elements for the fantasy story fragments).  I figured the easiest one to find would be the Greco-Roman helmet.  But when I went looking through all the clip-art sites I knew of that had free-for-commercial-use (important if you have a budget of $0, as is the case with this project) clip art, this was the best I could find:

 

And here we run into one of the problems of trying to get cover art when you can’t do art and don’t have the money to hire an artist:  You have to find things needed to execute your vision that are made available FOR FREE.  Sometimes, the best thing you can find is a picture of a helmet where the eyes are pointed one way and the crest is pointed another.  Sigh.  This bit of clip art would be perfectly fine for some art projects, but for a quality book cover it’s completely unsatisfactory.

I didn’t like any of the robots I found (though I did eventually pick one).  I never did find a space station that worked when reduced in size enough to fit on the cover.  And without those sci-fi elements, the swords made the cover a little too fantasy-themed for what I was going for.

I tried various options to replace the elements that weren’t working — the traditional tragedy\drama masks (as one of the stories features a retelling of a particular Greek tragedy), karate silhouettes (as one of the stories features martial arts), and finally forced myself to try slapping in the best of those robots, after all.

And this was the best I could come up with:

Ugh.  That just looks so… cartoony and mismatched.  It wasn’t appropriate for this book cover, at any rate.

So, the art director side of me had to tell the cover artist side of me to change plans.  I usually prefer an illustrated cover, but my one idea for an illustrated cover wasn’t working, so it was time to change plans.  If I have a budget of $0, I can’t afford to be too picky, and one of the advantages of a self-publisher is flexibility; so, even though it hurt, I’d have to try using a photo on the cover.

This book is mostly a collection of of (self) rejected manuscripts (well, the good parts, anyway), so plan be:  A picture invoking the image of writing, with a “rejected” stamp over it.

This proved to be both easier to find the components for and simpler to put together.  There were plenty of pictures that invoked “the writer’s desk,” and while the image is a pure fiction of what my writer’s desk looks like (much too neat, and I usually work on a computer) I found I liked this one:

Put a frame around it (and delete the parts outside of that frame, which again took me a while to figure out how to do using GIMP), slip it onto an appropriately colored background (I picked black), and then all that’s left is the stamp and the text.

Which brings up fonts.  I needed a font that looked like stamped text, which (fortunately) I found several options for.  After going through the various suppliers I knew who had free-for-commercial-use fonts (Fontsquirrel, DAFont, and 1001 Fonts), I wound up picking Armalite Rifle (which I found by searching for a “stencil” type font).  If you didn’t know you were looking for a “stencil” type fonts, you could just browse the whole collection.  It only takes about a day to go through the entire font catalog of all three sites.

It was at this point that I found a major difference between building a cover with Photoshop Elements and building a cover using GIMP:  Elements has a number of tools for adding text and word art; if GIMP has those same tools, I couldn’t find them.  I wound up having to build the cover in Scribus.

Which worked out fine.  Scribus had the tools needed to color and angle the text appropriately.  It also allowed me to add the title font (I picked Roman Caps, which I found by searching display fonts) and the back cover text (written using the Alegreya font, which I’ve been long familiar with as I use it for the print editions of my fantasy novels).

And, as I’ve already posted, I came out with the following cover:

It took some adjusting, even when I got to this stage, after I saw what it looked like in the KDP Print proofing tool (note:  there are MAJOR differences between Createspace and KDP Print.  For example, with Createspace you can place your ISBN barcode wherever you want just by including a blank box in the right spot.  With KDP Print, there is only one permitted spot to put the ISBN barcode.  I had to flip my barcode and my company logo from my standard book design because of this), but I’d figured out the software well enough to make those changes with relative ease, by then.  In the end, even though it was nothing like I originally planned, I was quite happy with the cover.

The book design, though… well, that’s another story.  But what my next blog will be about, so stay tuned!

This Book Cannot Make Any Money — Cover Art Options

(Sorry for the delay in posting this article; it was ready to go, minus a couple links, on Sunday; website troubles made it so I couldn’t post until today).

After last week’s post, it should be no surprise to learn I’m not done with the edits for This Book Cannot Make Any Money. I probably won’t be done with that until the book is published.  With the methods mentioned last week, I hope to have the thing “fully” edited before my anticipated release date, but only just.  And, as has been shown by my past experience, even a “fully” edited manuscript might have errors which require post-publication edits.

The Merrimack Event, for example, had been edited multiple times over 13 years time.  More than half a dozen people touched that manuscript, making corrections.  I scoured every sentence to the best of my ability all the way up until the minute before I sent it to Kindle.  Yet there were over thirty errors found after publication I had to go back and fix… and I can tell you that most of them (including every then-than error, which is the thing I get the most nit-picked about in the reviews) were in the manuscript from its original rough draft, six people, thirteen years, and umpteen edits ago.

I’m not giving myself thirteen years to edit This Book Cannot Make Any Money.  I’m hoping to have this book out (the eBook version for certain, and hopefully at least a proof copy of the print version, as well) by my appearance at Marscon, all without encroaching on my writing time for the next Law of Swords book (which I also hope to have done by Marscon, but only to send it to my editor and not to publish).

Marscon is… *checks* less than a month away. *gulp* So I’d better get started on those OTHER things needed to publish this thing. With that in mind, lets talk about book covers.

Let’s start by exploring a no-muss, no fuss option for book covers that, yes, would even fit in my $0 budget. Namely, finding a FREE!  PRE-MADE! book cover from either a free (even) for commercial use, royalty and\or copyright-free stock art site (Pixabay, for example, has several fantastic-looking book covers available) or a professional cover artist who happens to have a few free (though this specific example offers paid-for “additions” like a Createspace-ready wrap version or certain types of marketing images featuring the cover).

This option can work for you… and if you have zero artistic talent and NO eye for art, maybe it’s the best option.  There are big disadvantages, however — you’re not able to ask for a free custom cover that actually fits your book (you might get lucky; for example, if your book features a robot detective (heh), a pre-made cover may feature a generic robot that would work for you), you might have to share your cover with other people (especially if you go with something like the Pixabay option), and — quite frankly — the cover you DO find may not fit the dimensions of your book, forcing you to either crop it, stretch it, or try something else.

Besides, I like the branding potential in always having a custom-illustrated cover for each book of mine.  I may not have any money, but I do have some freeware, and a clip-art site.

So, lets get started…

To begin with, you need to figure out what you’re capable of producing on your own.  Me?  I know the software, and I studied design, but I can’t DRAW anything fit for anywhere more public than a refrigerator.  I don’t expect to win any awards for a free cover I’m making for myself, I really hope nothing I make ends up on Lousybookcovers.com.

But, while I don’t like the idea of pulling someone else’s already completed book cover off of a stock art site, pulling ELEMENTS to make the artwork work from a stock-art site like Pixabay or a clip-art site like OpenClipArt.org should be fine.  And with software like GIMP and Inkscape, you can take those elements and blend them together into a cover that will work just fine for your book.

But what should those elements be?  Well, here are a few suggestions:

  • If you are not a skilled artist (which I am not), do not try to mix media.  It’s hard enough to manipulate single-media images to work (I don’t usually frequent Reddit, but it does have some good examples of what happens when you DON’T do it well); trying to get multiple images from different sources and media will increase the difficulty exponentially.
  • The same goes for art styles.  Mixing manga style artwork and classical renaissance artists could become a real disaster if you don’t know what you’re doing.  Even skilled artists sometimes fail to recognize problems in mixing styles.
  • A few authors use book covers which are designed around a single symbol (for example, this one).  This can work, as it does in the example, but only if it fits the story.  As a collection of disparate story fragments from different genre, that’s definitely not going to work with this book.
  • Oh, yeah, speaking of that — it is vitally important your cover fits both the genre of your book and your story.  Sure, maybe you’ve found a truly attention-grabbing picture of two monkeys swordfighting, but if you’re writing a romance novel that doesn’t have a single monkey in it (swordfighting or not), your audience is going to be annoyed at the bait and switch.  They’ll leave negative reviews (hurting your sales) and won’t buy from you again (hurting your career).

Okay, enough don’ts.  So, what am I going to do?

Well, let’s see.  In addition to the poetry (and I’m not going to market this thing to poetry readers), this collection features fragments from stories with elements of sword and sorcery fantasy, historical fiction set in the Roman era, classical Greek mythology, something that would have been set in the Rink of War universe, and a burger-flipping robot detective story.

Hm…

Well, I’ll see what I can do.  Cover reveal in my next blog, complete with an explanation of everything I did to put it together.  With Christmas and a few other events coming up, that may not be next week, but we’ll see.  I’ve got to hurry if I’m going to finish this series by Marscon.

This Book Cannot May Any Money — Editing Phase: When You Have No Resources

This post is a week and a day late, and it turns out I probably could have gotten it out last week, after all.  Ah, well.  On the plus side, the extra day is giving me a chance to make the following announcement:  The signatures have been signed and the contract is now fully executed!  I can now announce that The Merrimack Event will be turned into an audiobook by Tantor Media!  Tantor is a major player in the audiobook world, and I’m looking forward to working with them.  Audiobooks from Tantor will appear in all of the major (and some of the minor) audiobook outlets, including, yes, Audible, iTunes, Hoopla, and others.

But still, sorry about the delay in this post.  I’m trying hard not to let this project interfere with my next novel, which makes finding the time to work on this series a bit difficult at times.  It’s been even harder to manage that with this particular blog (I needed some solid, uninterrupted hours I could work on it for one editorial technique I’m presenting), but I finally managed to find enough time.

Now, if I wanted to talk about editing in general, I’d discuss all sorts of things.  For example, I’d want to cover the differences between developmental editing, copy editing, and proofreading (I did cover that in my Self-Publishing Roundtable).  I’d talk about what to look for in a good editor.  I’d probably talk about style guides, the differences between the Chicago Manual of Style, the AP Stylebook, Words Into Type, and Merriam-Webster’s Guide to Punctuation and Style, and I’d discuss why — if you are self-publishing — you might want to create a house style guide.

But that’s not the case with this series.  This series is covering the worst-case-scenario, the “you have no money, no (useful) friends, no ability to barter for anything, and almost no resources” type of self-publishing.  In this sort of situation, you probably don’t care about the distinctions between those different types of editing, you aren’t going to be employing any editors, and you can’t afford style guides of any variety.  You need to self-edit.

Self-editing is mostly about going back to make things internally consistent.  It’s about making your writing as good as you possibly can on your own.  It’s less about the technical, and more about getting things to flow.  And yet if you do it right, you can produce a product just as good as you can by hiring out the edits… but “doing it right” can be a bit hard to manage.

So, I now have enough content for “This Book Cannot Make Any Money” to declare the “writing” part of this book complete.  If I include the story fragment from the last blog in this series featuring the malfunctioning burger-flipping robot detective with the faux french accent, the book will be somewhere between 40-50k words long… which is VERY short for me, but is a reasonable novel length in some genres (such as certain types of romance novels, pulp westerns, and even some varieties of Mystery novel).  I don’t think I ever came up with a title for that piece, so (as a working title) we’ll call it the “Detective Hummer” story.  Time to edit!

But back up a moment, here. Usually, when writing, instead of stopping mid-story and re-writing, I note issues that I want to “fix” before I let anyone else see it, or things which I want my editor’s opinion on.  In the case of “Detective Hummer,” I had these notes:

  • This story wasn’t started with any form of plot or plan in mind; I started building one as I continued writing.  I need to be sure to check that there are no “plot-breaking” points in the story.  (This turned out to just be a story fragment, however, so that isn’t a problem).
  • Originally, I started describing Hummer to establish the ‘sci-fi’ elements for the story.  A malfunctioning robot working as a burger flipper was just supposed to be a set piece; it wasn’t supposed to become the main character.  I should see if I need to add more details about Hummer (and comparable details to the other burger-flipping bots) so that readers will have enough details to figure out what’s going on.
  • I rushed, a bit, through that scene in the police station.  I need to be sure I included enough elements to explain why Hummer was able to get into Evidence Lock-up without being stopped.  I also need to make sure, in a scene that by necessity had very little dialog, that there are enough details and things so you don’t violate the (often over-applied, but there is a kernel of wisdom in the saying) principle of “Show, don’t tell.”

Now, there are undoubtedly more issues than that, but those were the ones I remember coming up as I was writing.  Ideally, I write these down, and then set the story aside for a couple months so it goes “cold” for me.

But I don’t have that luxury with this piece — this is a demonstration piece, so I need to have this story fragment ready to go by the time my blog on Book Design comes out in a few weeks.  Rushing through edits only gives you bad editing, so I need to start editing NOW!

All right, time to check the editing budget.  I need to hire a good editor who works fast!  *glances at the budget*  Oh, right.  The budget is US$0.00.  Uh…

Well, that’s okay.  I’ve been able to negotiate a trade of services for editing, before (babysitting for editing in a couple cases; I nearly traded one of my mother’s quilts in exchange for a line edit of a book, once).  No problem!  Oh, wait, that takes a while to negotiate, and we’ve just established I don’t have any time.

Well, they’re never as good as a human editor, but there are some pieces of editing software like Hemingway… oh, wait, that costs money.   There’s Grammarly, too, and there’s a free version… that doesn’t include any of the elements that might make it more effective that LibreOffice’s native spelling and grammar checker.  To get the effective parts of that software, you need to pay for the premium version, and in the budget, we have… (checking again) $0.  *sigh*

Well, what about replacing the editor with a team of quality beta readers?  I know a number of successful writers who do that!  *checks rules of this blog series*  Oh, sigh.  I have to pretend I don’t have any friends or family who have any experience in this whole “writing” process, which eliminates the pool of people from whom I’d get beta readers.  *sigh*

Okay, pure self-editing it is!

Several years ago, I wanted to build my own houseboat, hoping I could use it as a writing getaway (I never finished this project, mind you, so don’t ask for pictures).  I recall reading a chapter in one boat-building book that was talking about boat motors.  It went into all of these details about what the ideal boat motor was (which included general advice like, look for a motor built in the 1980s or earlier, because the boat motors manufactured today aren’t built with proper marine-grade materials… or something like that).  At the end of the chapter, however, it started talking about options for if you COULDN’T find an appropriate boat motor.  It ended by saying (paraphrasing, because I don’t have access to that book now) that if necessary, there were ways to modify most lawnmower engines so that they could work as trawler motors.  While he didn’t recommend it, the author felt the need to tell you how to use this option because it was better to use a lawnmower-converted into-a-trawler-motor than to not build your boat for lack of a motor.

In that spirit, I DO NOT RECOMMEND pure self-editing.  It is the worst possible option that I think the self-publisher can take when it comes to editing… but I’d rather see a good book self-edited than not see the book at all (and self-editing is better than no editing).  So, I’m going to go through several options, and I would suggest you use several of these techniques if you find yourself stuck self-editing your own work.

Now, the first thing to do is just to re-read the story yourself, keeping those initial thoughts (mentioned above) in mind.  This is FAR better done if you actually have time to let the story get “cold” in your mind, first, but as I said before I don’t have time for that; I started self-editing almost the moment I ended the last blog, and I’m still a couple weeks late.  Still, you can catch some things, especially if you already have in mind what you’re looking for.  You will not catch everything, but you will catch a number of things, and your story fragment (or book, or whatever you’re editing) will be all the better for it.

Now, because people tend to see things different on a screen vs. in print, print out a copy of your story and read it again on paper, making notes on paper you’ll transcribe back to the computer later.  You’ll see things you missed the first read on the computer… (though the computer read could show you things you miss in print, too).  Again, pay special attention to those things you’d already identified as needing work when you were writing it.  You still won’t catch everything, but with each successive pass there will be fewer things you need to fix… (but a caution:  You may introduce new mistakes trying to fix the old if you aren’t careful, so BE CAREFUL).

Next, we’ll try the “reading backwards” method.  This is a technique that may help compensate for not having the time to let the story fragment get “cold.”

Now, you might be surprised to learn the technique does, in fact, work the way it sounds.  Sort of.  But this doesn’t mean you should be reading your sentences backwards, word for word.  If I had to read the whole story as “.hour an than more little in began Burger Bungalow at shift Hummer’s.  system transit public the to back hurry to had it then And” I would not only find no mistakes, I would soon be driving myself crazy.

No, in this case “Reading Backwards” means to read, paragraph by paragraph (some people do it sentence by sentence, but I think that’s a bit extreme), the manuscript in question.  This helps isolate each paragraph, and it reduces the “I meant to write it this way so I see it this way even if its wrong” effect that most self-editors encounter.  Or so I’m told — I’ve never noticed, one way or another, whether it helps with this.

But I do think it can help if you read your story this way, correcting whatever grammatical or typographic errors you find along the way, then think really hard about what the context was needed in order for the paragraph you just read makes sense.  Then scan back a few paragraphs and SEE if that context is there.  Maybe you did write the context down earlier in the story, but if it isn’t in the text recently enough to be fresh in your readers’ minds they are going to get lost, and that needs to be fixed.

Finally, I present two variations of a method for the final copy edit.  Both work the same, but there are slight differences.

The first is reading aloud to yourself.  You are checking both for grammar and flow with this one — if you stumble while reading (at least, if you stumble for reasons that can’t otherwise be explained by a sudden allergy attack, the phone ringing, your water heater exploding, that kind of thing), check to see if something is wrong with either the grammar or the flow and fix it until you can read it without stumbling.

This method catches most errors… if you can read your own work properly.  The problem is that most people (even reading aloud) read what they INTENDED to write, not what they actually wrote.  Letting your writing get cold helps with this, but it doesn’t completely fix it.

Which is why some people try a variant method:  Getting your computer’s pre-installed screen text reader to read it out loud for you (I tried using Narrator, which comes with Windows 2007; a horrid  voice which I can only get to work by copy-pasting the story into a text document, but its (barely) functional.  Most operating systems come with one, but if yours doesn’t there are plenty free versions for download — I haven’t tested them, however).

You will hear your text AS IT IS ACTUALLY WRITTEN, instead of as you intended to write it.  That is immensely valuable, and will make errors stand out dramatically.  When you DO find an error, pause the narrator and make the correction.

It is also time consuming, sometimes hard to understand thanks to the synthetic voice, and easy to lose focus on.  It’s great for short works, but if your book is, say, 150,000 words long (three of my novels have been that long or longer) you’re probably going to have to endure this for DAYS.   The computer narrator is accurate — more accurate than most human readers — but the synthetic monotone delivery will drive you crazy after a while.  Some people can stand it, others will start getting easily distracted, lose focus, or even fall asleep during the reading.

The synthetic monotone delivery has one more drawback, as well:  It will help you find grammatical errors, but it might obscure errors in the flow and tone of your delivery.  So, this method of editing isn’t flawless.

(Note:  It was originally my intention to have a full-on demonstration of the above technique which I’d share on this blog, but setting things up so I could record Windows Narrator reciting the Detective Hummer blog entry — as I’d planned — proved to be technically unfeasible.  Sorry about that; this was supposed to be the centerpiece of this blog post, and was partly why it was delayed until this week, but it just didn’t work).

Ideally, if you want to fully self-edit your work, you would apply a mixture of the above techniques, refining your story each round through.  Not all of these techniques will work for everyone, but a blend of the ones that do work for you will improve your book immensely.

So, how effective is all of this?  Well, I’m not applying all of these methods to this project — I can’t invest the time for it, and I think I can get something good enough using just a few of them.  I think I’ll combine the basic re-read technique with the reading backwards technique (paragraph-length) and, if I can find the time, one of the last two variants (In my attempts to record a sample of the narrator, I caught two minor mistakes, but geez that thing gives me a headache; I’ll probably stick to the self-read aloud version).  Is that enough?  Well, you’ve seen the before with my last blog post.  You’ll be able to judge the after for yourself when I complete all of this.

Before you can do that, though I need to publish this thing, and before that I need to do something about cover art.  Hm… I wonder what the topic of my next blog is supposed to be?

This Book Cannot Make Any Money — Editing Phase: The Before

I’d love to start this post off with a follow-up to last week’s post on my possible audiobook deal, but I still can’t give any more details.  Things are progressing behind the scenes, however, and I wouldn’t be surprised if I have some news by next weekend.

But enough about that.  I still need more content for “This Book Cannot Make Any Money,” and I also need a “before” example for my post on editing.  So let’s try to write a short story that’ll help fill things out (keep in mind I have no outline, no plot idea, done no research to prepare for it… at the time I’m typing this, I’m not even sure what genre I intend this story to be.  Furthermore, I’m usually terribly slow on short stories, and I only have a couple hours of time each day — usually when also eating dinner or sitting with family who INSIST on talking to me while I’m trying to get things done — to write it in.  And it’s completely unedited — that’s the whole point, after all — so… don’t expect much):

HRC-2057-BB, a.k.a. Humanoid Robotic Chef 2057-Bungalow Burger (known to most of the staff as “Hummer,” both because of its full name and because of a particular noise from a minor electrical short it had developed in its first year of operation) had a lot of unnecessary programming in it.

It was originally supposed to be labelled HRC-2-CdlC (or Human Robotic Check 2-Creme de la Creme), intended for a high-end gourmet restaurant, before said restaurant went bankrupt.  It was instead sent to one of the three thousand Bungalow Burger fast food restaurants.

Bungalow Burger had added its own food prep and customer service protocols to Hummer when it had been purchased, but had never bothered to delete Creme de la Creme’s food prep and customer service protocols.

That, combined with that very old short circuit, led to some very strange conversations with its human co-workers, sometimes.

“Zey want zer burgers well-done?  Sacré bleu!  Ze flavor of zis exzellent meat will be utterly ruined!”  Zzzzt!  Hummmm….  “The customer is always right.  Two well-done Bungalow Big Burgers, coming right up.”

“Belay that, Hummer,” that particular Bungalow Burger’s human supervisor, Jeffrey Davis, said.  “Give that order to Zipper.  We’ve got a visitor who needs to talk to you.”

Zzzzt!  Hummm….  “You zay we ‘ave a viziter?  Do ze want to zee ze master chef in action?”

Davis smacked Hummer in a particular spot on his frame, nearly dislodging the silly chef’s hat he’d been ordered to wear whenever cooking.  Zzzzt!  Hummm….  “Drop that horrible fake French accent.  The police officer who wants to talk with you won’t appreciate it.  He wants to talk to you about one of your orders from a few days ago.”

“Law Enforcement Compliance Protocols engaged,” Hummer said, its voice losing all of the personality pre-programmed into it.  “I will comply with any lawful orders.  Take me to the law enforcement official.”

“Huh.  That’s a new one,” Davis said, chuckling.  “I’m guessing that’s installed in case Federal inspectors come to check up the restaurant.  Well, this has nothing to do with inspections.  Come on.  He’s waiting by the maintenance hub.”

The kitchen, if you could call it that (Hummer had been programmed with the ability to form culinary opinions, including on the quality of the kitchen equipment; both of its personalities had frequently debated with its human controller whether a flat-top grill and a bank of deep fat fryers constituted an actual kitchen or not), opened in the back to a bank of charging stations for the robot chefs and janitors that made up the bulk of Bungalow Burgers’ staff.  It was also the place they all went for maintenance, and an informal meeting room between robots and human whenever something needed to be discussed out of hearing of the customers.

A meeting in that room, per every protocol that had been programmed into it since it was brought into Bungalow Burger, meant that Hummer was to remain silent until and unless asked questions.

Davis led Hummer into the room and stepped aside.  By law, the human owner of a robot (or, in this case, their licensed representative) was both permitted and required to be present when questioned by law enforcement, so he would be sitting through the entire interview.

“Wow,” the policeman said.  He wore a name-tag labeling him as Officer Kaaya.  “Your humanoid chefs look really… human.  More than I’m used to seeing from a fast food joint.”

Hummer would never pass for a human on the streets, with obvious hydraulics making up its ‘muscles’ and metallic mesh for a ‘skin,’ but he did have several cosmetic adaptions that made him look more human than the track-motored industrial robots like Zipper, like humanoid-style walking legs and a molded body frame.

“Most of ours aren’t,” Davis said.  “One of the ‘bots we started this restaurant with was damaged a couple years after the first batch came in — a maintenance tech dropped one of its circuit boards into one of the fryers.  This was a higher-end model, bought as a replacement at a discount.  It’s got all kinds of advantages over our other ‘bots — various sensors like infrared thermometers to check the food as its cooked, UV decontamination lights to help ‘clean’ his workspace, simulated emotions for customer service, and all sorts of high tech gizmos that are completely useless for a glorified burger flipper.  Fortunately it can do that, too.”

“So, he’s less likely than any of your other ‘bots to mess up an order, then?”

“I wouldn’t necessarily say that.  He sometimes forgets he’s a burger flipper, not a fine dining chef.”

Officer Kaaya raised an eyebrow at that.  “Okay, that’s odd… but it’s not why I’m here.  Robot HRC-2057-BB, you cooked a meal two weeks ago corresponding to this ticket, right?”

It took a moment for Hummer to recognize the faded characters on a photo of an old receipt.  Culinary-grade humanoid robots were required to store data on all orders for at least two weeks for the purpose of tracking down sources of food poisoning.  They were sorted by a tracking number, which Hummer was quickly able to interpret.

“Two Bungalow Burgers, medium rare, one onion rings, one fries, and two drinks,” Hummer recounted.  “Was there a problem with our service?”

Were Hummer’s video sensors not so heavily calibrated, it might have missed the hesitant tick of Kaaya’s nose at the question.  The body language interpretation subroutine programmed into Hummer’s old front-of-house protocol interpreted that as an indication he was about to either lie or misdirect to avoid answering.  That protocol was dormant at the moment, however.

“Not at all.  However, these two customers of yours were found murdered later that evening, so we’re trying to track everything down about their day that we can.”

“Murdered?” Davis exclaimed.  “You didn’t tell me this was a murder investigation.”

“I didn’t?  Oh, my apologies.”  Even Hummer could tell that was insincere.  It did not know why Kaaya would have failed to mention the crime to its supervisor, however.  “I would still like to know, HRC-2057-BB, if there was anything unusual you might have stored in your memory bank about this order.”

The Law Enforcement Compliance Protocols demanded that Hummer comply, so he searched his memory banks.  It had perfect recollection of cooking that particular meal, and it went through things from start to finish.

“No anomalies in my routine are detected,” Hummer said.

“Maybe not with your routine, but with the ingredients themselves?  In particular, the burger meat?”

Zzzzt!  Hummm….

“Oh, no,” Davis said, his head falling into his hands.

“Incroyable!  You dare inzult our ingredients?  Zis ees an outrage!  We have only ze finest prime grade meats available anywhere in zis restaurant!” Hummer exclaimed.

“No, we don’t,” Davis said, smacking a particular spot on Hummer’s torso.  Zzzzt!  Hummm….  “We use cheap, utility grade beef.  Hummer sometimes forgets we’re a fast food joint, and not the high-end fine dining restaurant it was originally programmed to service.”

(Author’s Note:  In terms of weird research, I had to look up the cheapest grade of meat acceptable for commercial use to write the above line, which led to several articles on interpreting USDA Beef grades.  Fortunately, that was a quick bit of research; let’s hope that this little tangent doesn’t make it impossible to finish this story in time)

Kaaya’s lips twitched.  “I think I’m starting to see why you were able to get this guy at a discount.”

“Actually, the short developed after we got him,” Davis said.  “Still not sure quite what’s causing it.  But his memory is still accurate enough to pass certification, and his cooking skills are… well, better than any of the burger flippers we’ve got, here.  I just wish I knew what the restaurant that we got him from was thinking when they gave him that ridiculous faux-french accent.”

Davis had not quite been accurate in his description of Hummer’s special abilities.  He did not have ‘simulated’ emotions, he had ‘synthetic’ emotions, though the mistake was understandable.  Robots with simulated emotions were designed to react in specific ’emotional’ ways (for example, arranging their facial features to resemble a smile or a frown) to specific external stimuli.

Synthetic emotions would start much the same way, though with a wider range of reactions to a significantly wider range of stimuli.  Then, using heuristic analysis of the situations which prompted those stimuli and an adaptive artificial intelligence, they would start applying those emotional reactions to different stimuli.

In other words, robots with synthetic emotions would learn emotions.

Hummer had been programmed to have a strong sense of pride in its cooking.  That pride in its cooking had also grown to be pride in itself, being the best cook in the restaurant by far.  Having its trusted supervisor describe its accent as “ridiculous” hit those heuristic pride emotions hard.

But after having slipped its Law Enforcement Compliance protocols once after an emotional reaction, Hummer had to be careful not to react again, instead choosing another seldom-used emotional state in its place:  Stoicism.

“So, what was wrong with our meat, Officer Kaaya?” Davis asked.

“The couple who died was poisoned with a neurotoxin similar to what’s found in box jellyfish, laced with some synthetic controlling agent I’ve never heard of that delayed any reaction to said neurotoxin for several hours.  Our coroner was able to determine, after investigating their stomach contents, that the bulk of the neurotoxins were concentrated in some burger meat they had ingested about two hours before they died.  From this receipt, we’re fairly confident that said burger meat came from your restaurant.”

“Has anyone else died?” Davis asked.

“We’d have shut you down long before now if they had,” Kaaya said.  “No, the poison must have been added to those specific burgers and no others.  And, given the amount of time needed for that poison to act, we figure it must have happened while the couple was here… meaning it was done by either the staff or another customer.  Tell me, Mr. Davis, how many human staff are employed at this Bungalow Burger?”

“Uh… just three of us.  The day shift manager, the maintenance tech, and me.  Labor costs are so high, nowadays, that even high-end robots like Hummer are cheaper than employing human staff… or so that’s what corporate says.  Things are automated enough that we don’t even need to be here every day.”

“And your shift starts at?”

“Four in the afternoon, usually,” Davis said, then gave a start.  “But it wasn’t us!  Like I said, we don’t need to be here every day; none of us were here in the restaurant when this couple came through.  We were four hours drive away.  The regional manager had a meeting of all of his franchisees, upstate.”

“And it’s unlikely your ‘bot, here, tried to kill them,” Kaaya said, shaking his head.  “We’d like to check your sales records for that day, see if there are any connections between your other customers and the victims.”

Letting out a deep breath, Davis nodded.  “Of course — we’re quite willing to co-operate in any way we can.”

Hummer knew that wasn’t corporate policy — Bungalow Burgers did not permit the release of customer information, even to law enforcement without a warrant — but the Law Enforcement Compliance Protocol (which were added to its programming at the builder’s discretion, not the restaurant’s) prevented it from saying anything.

But, underneath all the protocols dictating its outward behavior, Hummer’s synthetic emotional matrix was suggesting it take radical action.  It had pride in its food — even if all it made, in its current employment, was burgers — and the thought that someone had used its cooking to kill someone hurt that pride.

Hummer needed to do something.


Technically, there was no law or corporate policy requiring that robots remain in their owner’s establishment when not engaged in normal operations.  The one concern might be keeping batteries recharged, but most modern robots had batteries that would allow them to continue normal operations for up to three days between recharges.

However, there had never been any reason for Hummer — or any of Bungalow Burger’s other robots — to leave the restaurant, either.  Leaving the restaurant was not a behavior it had been programmed with, and so Hummer never left.

But Hummer had synthetic emotions.  In order to make synthetic emotions work, there had to be a mechanism for an emotional reaction to override normal behavior.  Usually, this was limited to something simple, such as laughing at something its heuristical analysis determined was ‘funny’ during a period in which its normal programming said it should do nothing, but the level of basic programming it was permitted to override increased the greater the emotional response.

Currently, Hummer’s emotions were so engaged that it could even override the valid orders of a law enforcement officer.  Hummer didn’t want to believe that its burgers were the murder weapon for a pair of humans.  It was also curious — could it even tell if they were?

Among the ‘various sensors’ Davis had explained Hummer was equipped with were tools to detect contaminants in the food, both biological and non-biological.  They were supposed to be sensitive enough that it could tell if there were enough traces of shellfish on a mixed-use cutting board to cause an allergic reaction if food was prepped on it.  Most of Bungalow Burger’s burger-flipper bots didn’t have any sensors at all, so there were no protocols requiring Hummer to use that sensor, but the pride in his work that his original programming demanded of him was such that he used those sensors, anyway — every time he cooked, he made sure there was no contamination in any food he prepared.

So, if the burgers were poisoned, Hummer should have detected it… right?

But perhaps the poison was something that its sensors wouldn’t detect.  Hummer did a search through all of the documentation available to it, but still could not tell if this ‘box jellyfish toxin’ would be registered by its contamination sensors.  The only way to be sure would be to experiment and see.  The problem was attempting that experiment required a sample of the toxin, which should not be available in the Bungalow Burger restaurant.

The police would have a sample of the poison in evidence lock-up at the police station, however, which was why, after the store had been closed up, Hummer had left the Bungalow Burger for the first time since its arrival, several years before.

Directions to the station were easy enough to find on the net, but walking on bipedal legs it would take hours to get there.  Fortunately, public transportation was still running, and it was no longer strange to see a humanoid robot using public transportation.

Of course, once it was at the police station, Hummer’s problems weren’t over.  Robots weren’t allowed inside the station, unescorted, without identification declaring they were in the employ of the police, and all robots were hard-coded with warnings not to take orders that would have them stealing or copying those identifications.

But Hummer hadn’t been ordered to do so — it was acting on its own, in response to his emotional prerogatives.  Hummer hadn’t known the term ‘loophole’ would ever be relevant to its programming, but it was capable of exploiting one when it needed to.

There was a public charging station for robots of all designs only a block away.  The statistical probabilities of finding a robot with the necessary identification to allow them entry into the Police station was highest at that charging station.

If Hummer could even comprehend the concept of ‘luck,’ he would think himself lucky to find a lone robot at the charging station bearing an identity transponder that would allow it access into the police station unsupervised.  All it took was disconnecting the other robot’s battery and towing it down a nearby alleyway where no-one might stumble across it.  Hummer then disconnected the transponder from the police robot and mounting it into its own frame.

And into the police station Hummer went.


Hummer was not doing a good job of ‘blending in’ as he made his way to the evidence lock-up.

“Since when have we had bipedal robots?”

“I’ve never seen that ‘bot before — is he new?”

“Does that ‘bot have a spatula in its holster instead of a service weapon?!”

Hummer could hear these and other questions, but at least no-one was curious enough to follow or investigate it.  Evidence lock-up was down several flights of stairs (Hummer was programmed to navigate stairs safely, but had never tested that feature of its bipedal legs until that day), in the basement, near the coroner’s offices and the morgue.

Of course, then Hummer had to figure out where the sample of its burger had been filed.  That first involved finding any case files featuring its restaurant (this wasn’t the first time a Bungalow Burger was involved in a crime, locally, but it did appear to be the first time it was involved in a murder), then figuring out how the evidence filing system worked.

Initially, Hummer tried to plug in to the network directly, but the firewalls wouldn’t let him in.  However, the in-station computers were set up to bypass those firewalls automatically, and it was easy for Hummer to get onto those computers and give himself access.

After that, it was a quick search, and there it was… locked up.  It wasn’t hard to bypass the security on the electronic lock, however, and finally Hummer had its sample.

The contamination probe was quickly inserted and… yep.  It could distinguish the poison from the original (now spoiled) meat… and from the stomach juices and other assorted tidbits which contaminated it.

There is was.  Absolute proof that it wasn’t Hummer’s cooking that killed the couple.  That should allow the satisfaction circuit to kick in, and it could resume its days at the Bungalow Burger without a problem.

Zzzzt!  Hummm….

“Hey, does someone smell something like wires burning?” a voice called from outside of evidence lock-up.

Fine dining chef persona engages or not, Hummer knew it needed to get out of there, and as quietly as possible.  After replacing the evidence and re-sealing the evidence locker well enough to hide the fact that a robot had ever been inside it, Hummer took off, back up the stairs and out of the police station.  Its mission was complete.

Heading back to the alleyway where Hummer had left the deactivated police robot, it contemplated the truth of that statement.  Yes, logically, the synthetic emotion of pride should be satisfied by having confirmed that the deaths weren’t its fault, and emotional subroutine ‘satisfaction’ should be engaged.  It was not engaging, however.  Along the way, ‘pride’ brought out another, rarely used emotion into the mix:  Curiosity.

The meat had not been contaminated when Hummer cooked it.  It was contaminated when the couple ate it… which, given the timeline mentioned by the coroner in the file, had to have been consumed before the couple were last seen in the Bungalow Burger that night.  Davis and the Bungalow Burger’s other humans had been away during that period of time, as well.  Officer Kaaya had been right — whoever had murdered the young couple had been a customer at Bungalow Burger, as well.  But which one?

It would take a real investigation to figure out which customer was the murderer, however, and Kaaya had already marked the file as ‘closed, unsolved.’  Apparently, following through to investigate each of those customers was too much for him.

“Sacré Bleu!  If zomeone iz to zolve zis crime, it must be me!”

Which meant Hummer need to learn how to be a detective.  Fortunately, there was a certain captive police robot down a certain alleyway which was available to learn methods and techniques of investigation from.

And then it had to hurry back to the public transit system.  Hummer’s shift at Bungalow Burger began in little more than an hour.


…And that’s it for this story fragment.  No, I am NOT continuing this, at least not any time soon; I’ve already got far too many things to do, first, and I’m not really equipped to write a mystery novel right now… not even one featuring a burger-flipping robot detective.  But I will be demonstrating self-editing techniques on it in my next blog post (probably the week after next, as next week looks pretty busy), and then the (edited) fragment will be published in “This Book Cannot Make Any Money.”

This Book Cannot Make Any Money — Getting Started: Content

In the first part of this series, I set up my computer with a bunch of free software for the production of a book without any budget. Today, I deal with the book’s content.

Now, if you’ve been following this blog long enough to remember the original post on this concept, you’ll recall that I planned the publication of “This Book Cannot Make Any Money” as a sort of tutorial (or, rather, to borrow a term from the Youtube gaming industry, a sort of “Let’s Play“) for self-publishing a book.

But I have more than one reason for pursuing this blog series:  I want something I can use to test out the KDP Print service, and I don’t want to use one of my more substantive books.  I need something new, something unconnected with my larger series, but I don’t have time to write an entirely new novel.

I do have time, however, to compile a bunch of material from my past writing can’t be used elsewhere. Things like poetry (much of it written in high school), a short story (nearly a short-short) that doesn’t fit with anything else I’ve ever written, fragments of other stories which I’ve rejected, myself, for one reason or another, etc.  Maybe even a few of my past (or future!) blog entries to add a little more to it.  Anything I can find, from my old material, that wouldn’t be a COMPLETE embarrassment to publish (though some of those bits of high school poetry are pushing it).

And many of them have one thing in common:  Someone (and in some cases, it was me), somewhere along the way, said these bits and pieces can’t make any money.

So why am I bothering with a blog on this book’s content, if the content is already taken care of?  Well, there’s more than one thing to mention, here.  For one, once you’re done writing your book, even before editing it, you need to evaluate the content and judge whether you’ve done a good enough job to actually publish your work.

Writers have a reputation for being neurotic.  No wonder!  We have to be egotistical enough to believe that our writing will interest others while still being humble enough to allow constructive criticism.  A lot of would-be professional writers lack the confidence to believe that what they’re writing is worth publishing, and so never get published.  A lot of writers grow an ego so large that they never accept criticism, and therefore produce low-quality work.  That sort of dichotomy is a veritable breeding ground for Imposter Syndrome.

When you’re self-publishing, it can get even worse.  You need to believe that your writing is good enough to sell, but you need to keep your ego in check enough to maintain quality control.  Now, I have occasionally had quality control issues (as all of the reviews for The Merrimack Event warning about my “then-than” issue might suggest; I’m working on that, people, but so far I’ve only found five.  From the reviews, I would think I had a lot more than that, so I’m still looking for more before uploading a revised version.  If you could point specific instances out for me instead of just spouting that I have “lots” of then-than errors for the hundredth time, I’d be grateful), but mostly of the mild and technical variety.

That sort of error should be found and cleaned up, preferably (sigh) before publication, but that’s not a substantive issue that would prevent the work from being published.  What you need to look for is:

  1.  Do you have a proper beginning, middle, and end?
  2.  Do you have enough story and character development to support the plot?
  3.  Do you have enough conflict to create dramatic tension?  (This is needed EVEN IN COMEDY, if you are producing comedy-with-plot)
  4.  Is your plot premised on bad research?  If so, are your readers likely to call you out on that?
  5.  Opposite problem:  Have you added so much conflict that your plot is becoming convoluted?  Must your characters rely solely on luck to succeed?

If you’re capable of giving truthful answers to the above questions, you can evaluate your own work to determine whether it’s acceptable for publication.  (If you can’t, you need to find someone who can to read your manuscripts for you)

All of these problems can be fixed; the question becomes how much time will it take to fix them.  If you figure you’ve got the time to fix it, go ahead and FIX the darned thing and then publish it.  If you figure the amount of time needed to fix it is more than its worth, you need to be able to reject your own work.

Much of the (non-poetic) material intended for “This Book Cannot Make Any Money” comes from my older work that I rejected for one or more of the reasons, above.  Which doesn’t meant the writing is horrible; just that it would take me far too long to revise the whole manuscript into something I could publish.

What I’m going to do is add the parts of each manuscript which aren’t going to require too much time to fix up, leaving the rest behind.

But, as I said, I had something else to cover in the content portion of this blog series:  Not all of the content has been compiled, yet.  There’s still more that has to be written.

Even with a collection of previously written material like this, I still need to add enough structure to turn it from a loose collection of random writings into, well, an actual book.  Some of the material is handwritten, and needs to be typed up.  Some of it will need to be cut and trimmed down to just the acceptable portions.

But to start with, I need to make the content fit into some sort of structure; that turns it from a random collection of my outtakes into an actual book.  That may include writing some (small) bits of content to make things fit.

For example, I’m going to need to split the poetry into three sections.  It naturally fits into two — High School poetry, and haiku.  So, reluctantly, I’m going to spend some of what is normally my blog-writing time over the next few weeks writing poetry.  Bleh.  Maybe I can make it something silly, like limericks.  If I can come up with limericks connected to my existing writing, that might even be fun.

Of course, that’s not all of the “extra” writing I need to do.  It may only be a couple paragraphs here or there, but I still need to write introductory pieces for each section.  And an introduction to the book itself, which may be lengthier.  And, well, there’s one other thing that I know hasn’t been written, yet (but we’ll cover that later).

So, even though the bulk of the content of “This Book Cannot Make Any Money” is already complete, I still need to fire up that LibreOffice suite I installed last week and get to typing.  Fortunately that’s pretty straightforward — no-one here needs me to demonstrate typing, do they?  (I’ll note that I would be a terrible typing tutor; while I don’t quite use a “hunt-and-peck” style of typing, I never learned touch typing, either.  My method works for me, but that’s after more than thirty years of experience in typing “wrong.”).  The only note I think needs to be made here, for those following along at home, is that I recommend saving any files created under the old .doc extension (often called the Word 97-2003 format); I’ve found it to be more universally compatible than the newer .docx format.  I’ve encountered the weirdest glitches when trying to work with a document copied and pasted from a .docx file.

Another thing:  A lot of self-published writers recommend using Microsoft Office’s styles extensively.  I can see why — much of the software designed to automatically convert your eBook from .doc (or .docx) to .epub uses elements from those styles to determine things like chapter separations and the like.  Since I’ll be building the .epub file manually, however, I won’t bother with that.  I stick to the default style and worry about formatting the book later.

There should be at least one more piece of content I need to write that hasn’t been mentioned yet, however:  I need something unedited to use as material for a before-and-after type demonstration for a future “This Book Cannot Make Any Money” post:  The Editing process.

So, next week should be… well, whatever I come up with for that unedited material (I’m still not sure what that’s going to be; it should be short, however —  a thousand words or so, at most).  Then we’ll talk about editing your work without having the budget to hire an editor.

As a note, I am hoping to have something to announce in the next few weeks (possibly, though not likely, as early as next weekend).  If I do make that announcement, I’ll be bumping that week’s blog forward a bit.  Again, this PROBABLY won’t bump next weekends blog, but I figured I’d mention the possibility in case it did.

This Book Cannot Possibly Make Any Money — Getting Started: Software

As was in my “Future Plans” post, I’m currently working on three writing projects simultaneously:  The third Law of Swords book, the Fennec Fox Press House Style Guide (which is typically added to only as issues come up) and — in those times when I NORMALLY work on this blog — a book entitled “This Book Cannot Make Any Money.”

Work on “This Book Cannot Make Any Money” won’t prevent this blog from being written, however.  Instead, it is intended as its own blog series, allowing me to go through the process of self-publishing a book in a tutorial form, or (since I intend to actually publish the end-product of this series) maybe more like a “lets play” (to borrow from the Gamer vernacular) of self-publishing a book for my blog readers.  This is my second try at this kind of project; the first time it got bogged down and eventually swallowed by the need to deal with other things, but this time I have a more developed plan for how to handle this.

So let’s begin.

Once you’ve completed your first draft, I recommend setting a budget based on your projected worst-case-scenario projection for sales. I should add the caveat that I mean REALISTIC (not optimistic, not pessimistic) worst-case scenario.  If you mishandle things, yes, it is possible to never sell a single copy of a book, but that’s a pessimistic projection.  From past performance, I would project the worst case scenario for any of my sci-fi\fantasy genre novels as two hundred fifty ebook sales.  At $4 profit per sale (when I set the eBook price at $5.99), that means I could set a budget of $1000 and realistically expect to break even in a worst-case scenario.

But this isn’t one of my genre novels; this is a collection of material that I’ve read, or for which I’ve been told, or which I’ve even decided for myself “Cannot Possibly Make Any Money.”  With that as the premise for this project, I (at least for purposes of this blog series) project earning… no money from this book.  So my total budget is zero dollars, of which I can spend zero dollars on software, zero dollars on the cover, zero dollars on the editing, zero dollars on the book design, and zero dollars on marketing.  Okay, that was easy!

The rest of this blog series is going to be on overcoming the obstacle of having zero budget when self-publishing; how, with no budget, I can acquire the necessary software, create an original cover, get the book edited, and (easiest of all, though you might not believe it) market that book without spending one penny.

I have, or can and have borrowed from my mother:

1.  Microsoft Office Suite 2007 (IIRC, it was bought at an extreme discount through a program my workplace at the time was offering)

2.  Adobe InDesign v.6 (received as a gift; the last non-cloud version of InDesign.  I recommend sinking the costs of any software you buy in a purchase rather than creating a recurring cost by leasing it over the cloud)

3.  Scrivener, purchased during one of their half-price sales.  (I think that, with NaNoWriMo just around the corner, that’s about to come up)

4.  Photoshop Elements v. 10 or v.15 (I may be purchasing Photoshop Elements 2018 soon; v.10 came packaged for free with other software, and v.15 is borrowed from my mother)

5.  Corel Draw (whatever the latest version is; it’s on my mothers computer)

…and probably a few other pieces of software I’ve bought for my writing business (or my mother has bought for her quilting business) that I’m not thinking of right now.

But, since we’re maintaining the rule that I have zero budget for this project, I’m going to pretend I haven’t bought ANY of this, yet, and find substitutes.

I do have to make certain concessions for the series as a whole before we begin:  I have a blog, access to the internet, etc.  My blog is on a paid-for site, but its using a resource that is free and can provide a free host if necessary (WordPress).  These things I could manage to access from my local library, but the library usually won’t allow you to install software on their computers.  They might, if they’re equipped well enough, have some similar software installed on their computers you can borrow, but you can’t count on that.

So, it is a bit of an assumption that — even with zero budget to produce your book — you own or have access to a computer on which you can access the internet and are permitted to install software.  If you don’t, well, I’m sorry, I’m not sure what to suggest.

So, with the limitations of zero budget (minus that concession), what options in the software department are there?

In place of Microsoft Office:  Anything that I would normally do with Microsoft Office, I will instead — for this project only — do with the LibreOffice suite.  Now, both Microsoft Office and LibreOffice are suites of tools, but to replace the ones I actually use for my publishing work, I only need LibreOffice Writer (for word processing, replacing Microsoft Word) and LibreOffice Base (simple database software replacing Microsoft Access.  I use Access to maintain some of my notes, such as character records, which need to be kept across books of a series; due to the nature of “This Book Cannot Possibly Make Any Money,” however, I won’t be using it for this book).  LibreOffice is available for free (it better be, or I’m already breaking the rules), and will work with Windows, Macintosh, or even Linux.  (You do need to download the correct version for your operating system, of course).

An alternative to LibreOffice is Apache OpenOffice.  LibreOffice was, in fact, originally OpenOffice, but (skipping one long, complex, boring story to explain why) they split up into two organizations developing similar suites of software from a common ancestor.  LibreOffice is generally considered to be the better option, containing much of the original design team, but some people still prefer OpenOffice

To replace InDesign I’ll choose Scribus.  Scribus is also free, open-source software designed specifically to do, well, the same things InDesign does.  It’s been going strong for many years, now, and most of the bugs are already worked out!  (A word of warning:  They recommend that you install ‘Ghostscript‘ first.  I made the mistake of not doing this the first time I installed Scribus, and it caused several problems with my initial set-up.

Outside of Scribus, the only other free software I can think of that works as a replacement for InDesign is… InDesign.  A couple years back, Adobe offered a free download of a no-longer-supported earlier version of InDesign (in fact, a whole suite of programs InDesign was part of a package of), version 2.0.  It’s a bit hard to track down, and requires a software key (they provided one for the public domain at the time) which may no longer be listed anywhere, but if you can find it you can get the entire Adobe CS software suite for free.  Because of its obscurity, however, I’ll stick with Scribus.  (Scribus has a few more modern features, anyway).  If you do have the budget to BUY this sort of software, however, I wouldn’t recommend the current, cloud-only version of InDesign; instead, I would go with QuarkXPress.  A bit expensive, but it has a lifetime license (and thus is a sunken cost).

Scrivener is an odd one.  It’s a word processor designed specifically for creating books, but the Windows version (which is the only one I have) is missing several key features available on the Mac version.  Scrivener has promised a new, updated version soon (Scrivener 3.0) which should EVENTUALLY bring them up to near identical versions, but even with that the Mac version will be the first release.

The long and the short of it is that I only use Scrivener for eBook building, after the book has been edited.  Since that’s all I use it for, I will compare it not with other word processors but rather with other eBook-making utilities.  I’m at least somewhat familiar with Sigil, so that’s what I’ll be using, but I understand Calibre is popularly thought to be more intuitive and will likely have more tutorials for its use.  Nevertheless, I’ll be using Sigil to produce an ePub, which I will then convert to .mobi for uploading to Amazon.  (Calibre can do the conversion itself; Since Sigil can’t — at least not as of the latest version I’ve downloaded — I’ll instead be using a simple tool called ePub to Mobi).

The graphics suites are all that we still need to worry about.  As a substitute for Photoshop I’ll be using the popular (though a bit tricky-to-use) GIMP.  As a substitute for Corel Draw, I’ll try Inkscape (an open-source vector-based graphics utility I first saw in a package of “best open-sourced software” back in 2009.  I’ve often installed it but never used it, so this will be a bit of an adventure).  Not sure if I’ll need both of these programs, but at least I’m set up if I do.

Okay, software is taken care of.  Next time on “This Book Cannot Possibly Make Any Money,” I’ll start using these bits of software to ‘create’ the book’s content (which is already written… or is it?).  See you then.

Future Plans

I had hoped I would be able to say “the print edition for The Merrimack Event has been released!” by now… but I’m still waiting on a second proof copy (hopefully one without a mangled spine, this time). I do have several milestones to note: I reached over one thousand sales and one MILLION page reads (actually, it’s about one and a quarter million page reads, at last check). In less than a month.

Uh… okay, that was a little unexpected.

So, my original plan, as far as future book releases go, hadn’t factored in The Merrimack Event. As fed up with it as I was by the time it was released, I was convinced it wouldn’t sell, well, anything. I was publishing it to get it out of the way, so I could move on to my other books.  I spent more to get it out the door than my other books, so my expectations could be summed up as “I hope it will break even or something.  And it won’t be holding up my other books, any more!”

Well, it did that… and a lot more.  And now I have to figure out where to slip a (still untitled) sequel into the “to do” list.  I have a plan for a sequel… uh, somewhere (it’s been thirteen years since I last looked at it, but I do remember that I’ve preserved it across several computer moves).  Even if I can’t find that outline, however, I can come up with a new one; I’ve just got to figure out when to get it started.

The original plan was to finish In Division Imperiled (or whatever I call the 3rd book of the Law of Swords series), and then move on to By Claw and Arrow (the sequel to The Kitsune Stratagem.  I need to re-launch this book; it doesn’t seem right that my best-written book (both my opinion and by several objective standards) should also be my worst-selling book.  Getting the sequel out there would be an opportunity to do that).  After that, I was going to polish off To the Rink of War, turn it (and the unpublished serialized short sequels) and re-publish it as a novel.

I also had the idea of putting together a couple shared-world anthologies for a couple of my books, but that would have required some changes to Fennec Fox Press‘s business model (I’d be going from a sole proprietorship to a LLC, I’d have to change my accounting system so I can preserve money to pay other authors, etc.), so they were in a more nebulous “later.”  And, some day in the middle of all that, I was thinking of putting together that “This Book Can’t Make Any Money” blog project as part of a self-publishing tutorial on the side.  There were also a couple supplementary works planned that would slot in as they were ready — for example, the Fennec Fox Press House Style Guide, which is currently (in software parlance) in an Alpha version, but would need to be completed before I could even consider an anthology, and would be nice to complete before I send ANYTHING out for editing, again.

But now… all of that is out the window.  Oh, I’m still doing all of that, but now I need to slot in a sequel for The Merrimack Event.  And if I slot in a sequel for The Merrimack Event, I’ll have enough books in the list to need to account for book four of Law of Swords.  And as urgent as keeping those two series going is, maybe I’ll have to set aside those anthology plans until I’ve cleared up some other parts of my schedule.  And…

Well, anyway, I revamped my “order of production” schedule; see what you think.

  1.  In Division Imperiled (working title):
    The manuscript for this is already half-way done (or, well, I’m somewhere in the middle of it.  It’s gone pretty far off the trail set by the original outline, so I’m not sure exactly where I am in the story).
  2. The Fennec Fox Press House Style Guide
    The editor for In Division Imperiled has become overwhelmed with work, and may not be available for that book, so I might need to find another one.  If so, I’m going to need to have this ready for them.  This is a small thing, and can be worked on concurrently with In Division Imperiled.  It may wind up being completed first.  If released to the public (instead of just sent to the new editor with the manuscript), it would be a free download off the Fennec Fox Press website.
  3. By Claw and Arrow (Inari’s Children, Book 2)
    I still want to re-launch The Kitsune Stratagem, and getting this book out there is a big part of the plan for doing so.  So, while I’m anxious to get The Merrimack Event’s sequel out there, I’m still planning to get this out as quickly as possible, too.  If I start working on this and it gets bogged down, however, I’ll swap this with the next book in the queue.
  4. Shieldclads # 2
    Um, since I don’t even know where I put the outline for this, yet, I haven’t worked out a title for it.  But here is where I hope to slot it in.  Here is also where my original scheduled plan starts to diverge from the new one.
  5. This Book Cannot Make Any Money
    Another side project that can be done alongside other books (since most of the work will actually be done in the time allotted for working on this blog).  This could actually be ready any time before or after this point, but I’m guessing that I’ll have it done by this point.
  6. Law of Swords, Book # 4
    This was GOING to be The Rink of War, the novel-length version of the short story\novelette, To The Rink of War.  Instead, I have to juggle in the sequels to my more popular series, so here’s where Law of Swords 4 goes in.
  7. ONE OF:  Rink of War OR Nine Tales of the Kitsune
    Nine Tales of the Kitsune is the first of my planned Anthology projects.  IF I think I can generate the interest from other authors without too much trouble (one of the things I hope to do at my upcoming convention appearances is network with other writers), I may get this set up for this slot.  Otherwise, Rink of War (mentioned above) will be bumped here.
  8. Shieldclads #3
    Juggling two successful series is going to be difficult, especially with my other projects included.  This project and the next might wind up flipped, depending on how things work out.
  9. Law of Swords, Book # 5
    This should CONCLUDE the Law of Swords series.  I may revisit this world again, but with the series ended the schedule will be freed up for more “new” projects.
  10. Inari’s Children, Book # 3
    Current plans have this as the concluding book, but I’m not happy with the outline for this one.  If the relaunch of The Kitsune Stratagem is successful, I’ll rewrite the outline spreading the story into at least four books; otherwise, I’ll revamp it to conclude the series here.

And that’s all I can queue up at this point.  I still have more books planned outside of what you see here (including more Shieldclads, an anthology and possible sequels to Rink of War, some supplementary material for all of my series, and another sci-fi series dealing with a chubby pilot, his mind-reading girlfriend, and a space racing jalopy), and it’s possible one of those won’t let me go until I slip it in somewhere, but for the moment that’s as far as I have planned.

Edit:  Spammers are really going to town; I already have to shut down the comments on this one.