Initially I had thought this was going to be something of a sleeper course. I had expected it to be mostly reveiw, maybe some theory and details that I never learned. I started this course at the same time as the Python Automation course, and for the first few months I was taking both courses at the same time. But I found that I didn't have enough time to dive deep as I wanted on the course material and still have time to complete the week's coursework. So I delayed finishing this course until after I finished the Automation course. If anything this course was as challenging and practical as the other.

I now (somewhat) grasp the network protocol stack. I'd stumbled around the OSI model for years, but only barely understood it. I still lack much more than a rudimentary understanding, but it makes a lot more sense, that it is just protocol on top of protocol, abstraction after abstraction. Terms that I used to use interchangeably like dataframe and packet have subtle differences that are worth knowing.

I had barely understood hashed values prior. Now I use hashing in my scripts to confirm that the files prior to and after transfer or compression haven't changed or been corrupted. I have since learned that other protocols seem to handle this well most of the time, but now I no longer have to wonder, I can confirm it. And hashed values help in other areas I never would have guessed. I have users who capture various versions of the same files, making this exponential mess of files named the same in various directories and states. At scale this becomes a real challenge, but with hashed values I can identify which are copies of the same thing and which are not.

My career has largely been about working with individual hosts, one at a time. Our entire LAN is only about fifty devices, mostly workstations that are fairly easily maintained without needing centralized tools. But what can be done by one person on a small LAN doesn't scale well unless new tools are used, things like Puppet or Chocolatey, to make admin more productive. Leveraging VC like git with Puppet to treat devices as ephimeral is nothing short of amazing.

And the above is just a sample. I learned in practical ways how public-private key encryption works, things like Kerberos, and how a VPN actually is built. Much of these I had some practical knowledge of prior, but were largely magical black-boxes that I couldn't claim any real knowledge of. And as prior, the Google courses were taught largely by under-represented people, or at least folks that didn't fit the stereotype of white dudes working in IT. This is important to me for two reasons: first, representation should be important to everyone, but second, it is to me as someone who hasn't taken a typical career path. I have no college degree, and I was undiagnosed ASD for much of my life. Seeing others that are not as represented empowers me, that we don't have to assume that the industry will be the same as it always has been.