My goal for Sprint 16 is going to be 12 hours again. It will probably be much harder to pull that off than last week, since I don't have a holiday on Monday to get me half way there. But based on the last couple of weeks, I think it's definitely doable.

I should probably also briefly mention what exactly I'm working on at the moment. I got one chapter more or less finished. I've been writing it using ASCIIDoc, which is a really simple text-based book writing format that can be converted into other output formats using one of a number of tool set. (Usually you go from ASCIIDoc to DocBook to PDF/HTML/ePub/MOBI.)

This book uses a lot of math, and AsciiDoctor, the toolkit I'm using, will use MathJax to convert ASCIIMath to something displayable in HTML. My equations actually all look fairly decent when I export to HTML. And I'm OK with that being a target, but PDF is my primary target. AsciiDoctor flat out ignores your math stuff when converting to PDF. So I needed to come up with a system where I could get the math formulas into the PDF version of the book. I'm basically going to be extracting the math out of the raw ASCIIDoc text, post it to the MathJax website to get an SVG vector-based result, then programmatically use Inkscape to convert that to a PNG. I'll dynamically rewrite the ASCIIDoc text to reference an image, and AsciiDoctor has the smarts to embed the image in the PDF.

It's a little complicated, but I'm not aware of a simpler approach. Other than perhaps to contribute the change to the AsciiDoctor project and make it an officially supported feature or something.

If all goes well, I think I should have a PDF version with math in it by the end of the next week. (No promises though!)

After setting an 8 hour goal for Sprint 14, I then proceeded to not only reach the goal (which I did on Friday) but shatter it. I completed 17 hours during the week. That's as much as all three previous weeks before it. This is the kind of time I was hoping for when I started. I think it just genuinely took this long to really build momentum.

I do feel like I made the right call, lowering my goal week-by-week, from 10, to 8, to 7. Even though this week shows that if things go well, I can approach 20 hours (which I still think is actually very doable still) I don't plan on setting a super high goal for the future.

But it is nice to not only meet my goals, but far surpass them. I'm glad that I'm ending the competition strong.

To top it all off, I was able to do 6.5 hours of work today, on Labor Day. Obviously, it helps that I had the day off. I don't think there's any reasonable way I could have done that on a work day. But it was also a great way to finish things out strong.

I had hoped to have something worth showing off. I don't. Yet. I've said it before though… I think it's important to get things out to people earlier rather than later. My first book was probably delayed by two or four months because I revised so much once I got actual feedback. Had I done that sooner, I probably wouldn't have wasted so much time. In some ways, I think the only thing stopping me from releasing the chunk I've got so far is that I'm worried that people won't like it. That they'll find it incomplete, confusing, etc. And they probably will. Whenever I do release it. Because it will have been viewed by me, and me alone. So that shouldn't stop me from releasing, but it's going to. At least for a bit longer.

Lessons Learned/Retrospective

I'm kind of sad. We've reached the end of the competition. This is actually the first time I've ever been going strong at the finish line of one of these competitions, with the possible exception of the 1st one where I threw away my original plans (a reverse Space Invaders) and did a straight up Space Invaders clone on the last weekend. I've always crumbled apart long before the end. So this is new territory for me.

But having said that, I also wasn't active for the first 70 days of the competition, except some activity during the first week. Work got too busy, and all of this competition stuff took a back seat.

I've found a secret to success, at least for me. Let me walk you through it, in case it is beneficial to you too.

The first big paradigm shift for me was to just realize that what I wanted to do could not possibly be done in the time frame allotted. I was starting on this book stuff at Day 70ish, and under no scenario, including taking a month off of work, was it reasonable to think I could actually write a full book.

The second big shift was to also realize that I could not reasonably predict what I could accomplish. I had some hopes and some thoughts that seemed reasonably accurate (I actually don't think I was wildly off) but there was simply no way to know how far I could actually get. Work life, family life, entertainment, exericise… they all take up time, and it's tough to know how much. So I decided right from the beginning that there was no way I could set reasonable scope-based goals for this project.

Truth be told, that's something I've always known about creative projects like book writing and software development. You can fix the scope or you can fix the time period, but you cannot fix both. They are a function of each other. And creative work is beastly tough to accurately predict. Since these competitions are fixed time frame, I probably shouldn't be trying to pin down the scope ("complete this whole game" or "write these two chapters", etc.).

So the final piece of the puzzle here was to change my type of goal. Rather than saying, "Here are the tasks that I'm going to complete," I said, "Here's how much time I'll put in every week." This allowed me to have a concrete goal to make progress towards, without me feeling like I'm not completing everything I need to complete.

I do think this was the key for me, and the next competition that we do, I think I'll structure my sprint goals this way again. I think it works better this way, at least for me.

I don't think this would fly at work, with a customer that is paying us. It probably ought to work. But if I told our customer, "Yeah, it's hard to guess what we'll be able to actually complete. We're just going to make sure we put in 40 hours (or 20 uninterrupted, meeting-free, distraction-free solid development hours)," well… I just don't think they'd be happy with us saying that. Even though that's probably pretty accurate. Customers like being lied to and told exactly when they'll get exactly which feature. But I digress from the original point…

I do think I've found a pattern that has worked well for me lately. I'm looking forward to the next competition to give it another goal. But speaking of that, I'm not sure I need to wait because…

Post-Competition Goals

I plan on just continuing on. My goal for Sprint 15 is 12 hours. That's the highest goal I've set, and I'm mostly only doing that because I've already got 6.5 in. Next week may not be as high.

I had originally estimated a total of 400 hours to complete this book. I'm currently sitting at 40.5. That's about 10% of the way there. I feel like 500 might be a better estimate. Ask again in another month.

I think I can probably commit to continuing on at least through the end of September. If I average something like 10 hours per week for the next 4 weeks, maybe I'll be more like 20% of the way there. While not a guarantee, I want to get an actual early access style release out by then. I said that for this last month too. We'll just have to see how it goes. Again, the primary focus is in spending time with it, little by little, every day and every week. If I do that, then the chapters will fall into place, and eventually the completed book as well.

I'll keep posting here in the forum with future updates.

So my goal for the last full week of the competition is 8 hours. I think that is imminently doable. I still think 10 ought to be pretty easy if I can focus. But 8 is the official goal.

I'm hoping to do even more next Monday, the last actual day of the competition, which is a day off of work for me because it is Labor Day.

My real, underlying goal was to have some small piece of something meaningful to actually show off when the competition ended. I'll admit, with only a total of 17 hours instead of the originally planned 30 hours up to this point, I'm having serious doubts about that working out. I haven't given up hope completely, but I'd probably put it under 50% of that happening at this point.

Last week, I didn't reach my 8 hour goal, but I did get 6 hours total in, and the chapter that I've been working on is actually coming along pretty well. I think it's somewhat reasonable still to think I could have at least this one chapter kind of finished up and ready to go when this competition ends in two weeks.

The chapter I'm working on deals with randomness, probability, and chance. I think this might not be the absolute most important chapter that the book will eventually contain (that will probably go to the chapter(s) on matrices/linear algebra). But it's the place I decided to start. Not sure why, but it felt like a good starting point.

Sprint 13 Goals

Considering I've done 4 hours and 6 hours in the last two sprints, I suppose I should probably aim for something like 5 hours. Or 6. But I'm not going to do that.

I do think if I can get into a rythm, 10 hours should still be doable. Definitely 8 ought to be. There's a huge part of me that wants to keep my 8 hour goal. But the reality of what I've accomplished in the last two weeks probably ought to temper things a bit.

So the goal for Sprint 13 is 7 hours.

I feel like that's a reasonable balance between past reality, and future hopes. It's only an hour a day….

My current plan is to assume you know arithmetic and the basics of algebra (for example, you aren't completely out of your depth looking at a simple formula like y = mx + b, or a^2 + b^2 = c^2, and that if I said, "Solve this equation for x, or for b," or something like that, you wouldn't be completely lost. To be clear, I care less that you recognize those equations for what they are (the equation of a line, and the Pythagorean theorem) and more about the ability to do basic manipulation on those equations to get various things to happen.

To the pseudocode question, again, I am hoping that early readers can help direct me in that regard. My gut feeling is that code in a language you're familiar is way clearer than pseudocode. Not everybody knows C#, but the overwhelming majority of people who read a book written by me probably are. And most people who don't probably know C, C++, Java, or maybe VB.Net. Knowing any of those (especially the first three, which have the same general syntax) would probably understand the C# code better than the pseudocode.

I think my preference would be to do the code samples in C# rather than pseudocode, and then if there's enough interest in other languages, have a different version for other languages. Again, that's just "what feels right at the moment" not an authoritative, final answer. I hope readers can help direct me on that.

Looking forward to seeing how the Math book progresses. I am very interested in a game math theory book. The only book on Game Math I have is one on AI and it turned out to be well beyond my comprehension of mathematics. :|

For examples, Will the book use C#/MonoGame or will it use pseudo code?

My goal for Sprint 11 was to put in 10 hours on my book. My actual results was 4 hours. I really wasn't even close.

The problem I faced was that as I started to work on the BCL book I described earlier, I quickly lost interest. Perhaps I just started working on the wrong chapter. Maybe changing to a different section would have been a good choice.

Instead, based on some conversations with my wife as well as some other people in the chat room, I decided I should probably just work on the book that is more likely to help people (and as a result, bring in money). So after getting bored/disinterested in the middle of the week, by the weekend, I was working on the game math book instead.

I totaled 1.5 hours on the BCL book, and 2.5 on the game math book.

Even with that, 2.5 hours over a Saturday and Sunday feels… small.

If I don't loose interest, I'm sure I can beat the 4 hours that I did last week. But perhaps I should also be honest with myself and lower the goal number down from 10 hours as well.

So my goal for Sprint 12 is a total of 8 hours. I'm still hoping I can get 1 hour in every week day. If I do, that means I only need 3 over the weekend, which ought to be very doable. (Though note that I only got 2.5 in this last week.)

I don't have anything worth showing off yet. 4 hours split over two books just doesn't get you very far. There's a non-zero chance that I'll have something to show for it next week though. Especially if I can hit my 8 hour goal or my 10 hour stretch goal.

The book I'm planning on getting started on is a book that does more of a deep dive in the things available in the Base Class Library, and how to use them. It's sort of an extension of The C# Player's Guide. That book focused on the language itself. This one would focus on all of the material that is already available for you to use, and guide you through how to use them. This includes things like List, Dictionary, Threads, Nullable, Streams, IEnumerable, etc. Before long, I'll have a more fully developed list, and I welcome feedback on what should be included and what shouldn't.

That topic might seem somewhat boring. Indeed, it might turn out to be. I think I can make it interesting though. And if I can't produce something interesting and meaningful over the next few weeks, I'll switch topics and go with something else. (Next most likely candidate is a book about game math and physics.)

My only real goal for Sprint #11 (I think that's where we're at) is to put in 10 hours into the book. To make things a little more interesting, I'm only going to count things in uninterrupted 30 minute blocks. Two minutes here, six minutes there, etc. doesn't count. That's not enough time for me, personally, to get focused and engaged enough to get stuff done. So chunks of at least 30 uninterrupted minutes or it goes to waste.

I'm hoping I can do an hour every day during the work week, which would leave only 5 hours total for the weekend. But if I haven't done anything by Saturday, or even by Sunday, I can still most likely get to the 10 hour mark anyway.

Hopefully I'll have some more details for you in a week!