I glanced at my blog today and realized I’ve written very few posts lately. I’ve been working pretty hard on The Turing Exception. Between that work, my day job, and kids, I haven’t had a lot of time for blogging.

Most of January was spent working with my copyeditor. This is a bigger, more complex task that it might sound like. You might imagine that I turn my manuscript over to the copyeditor, and then get it back with a bunch of corrections, and it’s done.

In fact, what happens is closer to this:

• I send the manuscript.
• I get a bunch of questions in the beginning as my copyeditor goes to work.
• Then he goes radio-silent for two weeks as he gets deep into it.
• Then I get the manuscript back. This one contained about 4,000 changes.
• Some changes are easy to process: commas moved, spelling corrected, words replaced. I use Word’s change review, and it’s lots of clicking on “accept”. Still, it’s 4,000 changes, and it takes me several days of full-time work to review each change and accept it.
• Some changes are more difficult to handle. They might be a comment, like “you need more interior character dialogue here.” Then I need to go think about what the character is thinking about in that scene, and write a few paragraphs, keeping it consistent with everything going on around it.
• Some changes are widespread, like when I’ve described a single event several different ways over the course of a novel. Or used several different names to refer to one organization. I have to pick something, and then make sure it is consistent throughout.
• Some changes and comments I don’t understand, so I have to email back and forth with my copyeditor until I do, and then make the changes.
• When I’m done, I send the file back to the copyeditor, and now he can review my changes. There were about 300 on this last exchange.
• He accepts the ones that look good, but might have to make more corrections, which I then accept, and so on.

Eventually it’s done. The copyeditor and I are in agreement.

Then I get the manuscript to the proofreader. This is a second person who is focused on line-level items, like punctuation and spelling, although he’ll also catch some bigger issues. The manuscript came back from the proofreader with 800 changes. I basically go through all the same stuff as with the copyeditor. Some changes are straightforward, some are not.

If I make big changes, then it has to go back to the proofreader again for a second pass.

Along the way, I usually get feedback from beta readers who are getting back to me late. I hate to ignore feedback, so I do the best I can to address any issues they spotted, without breaking the copyediting / proofreading process.

Sometimes I’m trying to address beta reader feedback by changing only one or two words, to avoid having to do another round of proofreading. I remember this happening with The Last Firewall, where I think Brad Feld or Harper Reed said “I’m confused about what kind of vehicles exist in this world”. And so there’s a scene in the beginning of the book where Cat is crossing the street, and I had to get her to establish all the types of vehicles (ground cars, hover cars, and flying cars) in a single sentence, so that I didn’t make changes in multiple places.

I’m now one to three days away from finishing the proofreading cycle. When this is done, it will go to two different people for formatting: one person will generate the ebooks, and another will generate the PDF interior for the print book. Then I’ll need to carefully proofread both of those, to make sure nothing gets dropped, and no formatting errors or other mistakes are introduced.

It’s fairly intense work when the ball is in my court. But when it’s handed off to someone else, that’s my chance to do a little creative work. I’ve written about 15,000 words in Tomo, a new novel about privacy, social networks, and data profiling. No AI or robots…yet.

Last November, on a train ride to Seattle to see Ramez Naam and Greg Bear, I started book four of the Singularity series. Last Thursday, I finished the rough draft. I was very excited and did a little dance in my office. Of course, I’m not done. I’ve got to turn that draft into a cohesive story, polish that story into something that reads well, and get it edited, proofread, and then it goes into production (page layout, ebook conversion, cover design).

I’m excited to get it out, and I know other folks are excited to read it, but it’s still many months away from being available. I’m not sure exactly how long. It varies with each book, as I learn more, also depends on my work schedule. I’m making some small changes to my day job schedule that should give me more consecutive days of writing time, which will help me make steady progress.

For a few weeks though, I’ll be focused on other things: I’ve got to the second edition of Avogadro Corp fully out. I need to do some marketing work around my novel for kids ages 7 to 12, The Case of the Wilted Broccoli. I need to give a little attention to my Patreon campaign. And I’ve got blog posts I really want to write that I’ve put on the back burner while I worked on finishing the draft of book 4. At some point I also need to figure out a title for the fourth book.

But before the end of September, I should be back to work editing and revising.

Here’s the description I wrote for the never-started campaign:

I am asking for help to publish my novel Avogadro Corp. The manuscript is completed, and just needs a final round of copy-editing, cover design, and layout in order to be published. 

Synopsis 

David Ryan is a brilliant computer scientist, cherry-picked to lead a new project at Avogadro Corp, the world’s leading Internet company. The goal of the project, called ELOPe, is to create a next-generation feature for the company’s email product – one that can optimize the language of emails to make them more effective and persuasive. 

With his chief architect, Mike Williams, and a team of programmers, the two have proven the feasibility of the concept and are hard at work trying to release the feature. When David gives a presentation to the executive leadership of the company, they are impressed by the project results and effectiveness. But David fails to disclose to the executives that the project is grossly inefficient, requiring thousands of times more servers than any other project. 

The VP of Operations threatens to kick ELOPe off the servers if David and Mike don’t decrease the number of servers the project uses within two weeks. This would be a death blow for the project, in part because David has been deceptive from the start about how many resources the project has been using. David and Mike start scrambling to fix the performance of ELOPe. 

When it becomes clear a few days before the deadline that they can’t fix ELOPe’s performance, David stays up late making subtle modifications to the software. Instead of fixing the performance problems, David embeds a directive in the software to maximize the project success. David’s modifications have ELOPe filtering company emails to secretly modify any email that mentions ELOPe to strive for a positive outcome. 

The software is so good that at first, the effort seems successful – the project is allocated thousands of new servers and high performance computing experts are brought in to help optimize the code. Innocuous sounding emails convince people to grant more resources and develop new capabilities that make ELOPe more powerful. But soon ELOPe is social engineering people around the company to neutralize threats and strengthen itself. 

When Mike is sent on a wild goose chase to Wisconsin, getting him off the grid as just the moment when David needs him, it dawns on Mike that something is wrong. 

Simultaneously, Gene Keyes, a crotchety old auditor at Avogadro who is known for distrusting computers and using only paper records, begins to find evidence of financial oddities that all point in the same direction. 

Amid background news stories hinting at ELOPe’s ever growing influence, even at the level of government policy, David, Mike and Gene take ever escalating action to shut ELOPe down. However ELOPe anticipates and blocks their every move. 

As the humans prepare for a final showdown with ELOPe, Mike sees a pattern emerge in the news reports: the AI is actually helping humans by fostering peace agreements and stabilizing financial markets. 

Can they win a final showdown with ELOPe — or should they even try? 

Endorsements 

“This is an alarming and jaw-dropping tale about how something as innocuous as email can subvert an entire organization.  I found myself reading with a sense of awe, and read it way too late into the night.”– Gene Kim, founder of Tripwire, author of Visible Ops. 

“Avogadro Corp builds a picture of how an AI could emerge, piece by piece, from technology available today. A fascinating, logical, and utterly believable scenario – I just hope nobody tries this at home.” — Nathan Rutman, Software Architect, Lustre High Performance Distributed Filesystem 

Background for Avogadro Corp 

Avogadro Corp evolved out of a lunchtime conversation. I was arguing that the development of human level artificial intelligence is an inevitable consequence of the increasing processing speeds of computers. My friend countered with the argument that mere people who would do the programming, and we weren’t smart enough to create an artificial intelligence as smart or smarter than us. He challenged me to describe a scenario in which an artificial intelligence could be born. So I described one based on plausible extrapolation from known programming techniques. And the idea for Avogadro Corp was born. 

Avogadro Corp will be satisfying to technical readers who want realistic fiction, and enjoyable for casual readers who want easy-to-grasp explanations of how the science works.’ 

Project Timeline & Funds 

I expect that the digital versions of Avogadro Corp will be ready within 30-45 days of completion of the kickstarter project. Printed books will take longer, due to printing and shipping times.

About Me 

I’m William Hertling, and I live in Portland, Oregon. I’ve been a computer programmer, social media strategist, data analyst, program manager, web developer, and now writer. Avogadro Corp is my first novel, and I am currently working on a sequel.