The books I would love to write

For someone who can barely keep his blog up to date, I would still love to write a book one day. I’ve been told I have a great amount of technical talent multiple times, can sometimes firehose people asking for an intro to a topic, or get a bit detailed. So I’ve got plenty in my head to share, I’m just not sure on how to distill the knowledge I have and what topics people would be genuinely interested in reading. My frontrunners for the topics I’d love to write about are:

Learning DevOps

It’s new to a ton of people and companies alike, it’s not prescriptive, not resolved by tooling, not resolved by title changes, has no certification process, has no manifesto, is implemented entirely different at every company, comes with a culture change that everyone talks about in conversation which is never included in it’s definition, and sometimes a bit hard to explain unless you try to step back in time to explain it with another well established concept like I have with Toyota’s manufacturing methodology.

My exposure to it started while working at Microsoft in the Xbox LIVE team. I saw the evolution the group took starting after the Christmas 2007 outage debacle up until I left in July 2013. That outage was so bad by the way, the dates for when it started and ended were in different years, people wrote songs, gamers filed a lawsuit over it, and Microsoft gave a game away for free to everyone (whether or not they were actually affected by it) on the service to make up for it.

The amount of work the team did over the next year was monumental in terms of what was done infrastructure and software wise. There was a bit of warm breath rolling over people’s shoulders to get all the work done, but the product at the end of it all made us happier. Happy enough to at least pro-actively come out to say we weren’t going to repeat history.

By whatever methods we used to achieve these changes, the way we got there was very painful. The transformation started to slowly shift to a DevOps-like model where the days of twice-a-year planned outages were gone (unless we absolutely had to) and software changes were coming every couple of months. It wasn’t quite there, but we were making some really good progress. Interestingly, the DevOps term wasn’t coined until October 30, 2009, so we were moving in the right direction. I don’t think it was until 2012 that we realized we were just a couple of remaining steps from doing full-on DevOps as it was defined back then.

So that’s one possible topic. And it’s one that hasn’t had too many books published about it already. Although there is already one really well regarded book out there called The Phoenix Project.

Scalable Architectures

This topic has been covered for a while, at least from the view of a software developer or an operations engineer focused on the software itself. It’s also one of these type of topics that puts itself level with the popularity of discussing religion or politics at the dinner table.

I didn’t want to use ‘software’ in the title as some other areas like network, database, DNS, and CDN aren’t covered in as much depth as I would like to see. A lot of large companies have adopted some architectures that are utilized with little fanfare that could also be adopted to smaller companies that have established datacenters.

I can see this being a much more popular topic, but would probably need to call in some favors to get some technical content for the areas which I know just the surface for what is possible.

Or something else?

For those of you who personally know me, what should I write about? Doesn’t even have to be one of these two topics either.