This is where the Project officially begins. It is a time of calculations, creating the first user spaces, generating the ssh keys, the initial barebones design and opening functional accounts for the system.
Time to take a deep breath and make the plunge. There are so many "baby steps" (no pun intended, or is it?) to take and we will have an idea of what we are looking at by the end of this first iteration. But the bottom line is clear, there is no such thing as too hard here. It will be what it will be and will be only as good as it is created. So failure is not an option as much as an expectation, as well as the ability and necessity to overcome that failure.
I have no idea what the folks over at Mycroft have come up with if anything. The last time that I checked, they had about as much of a base as I have. That is mostly because I have theirs to begin with. So I am not looking at coding itself so much in the first iteration as I am creating the Rei user and giving her a good set of scripts to work off of that are persistent and will remain reboot after reboot.
i am borrowing projects
There's no law against it or anything, but I am working on Casglu and my work logic at the same time on the 100 Days challenge, and much of that will be functional on these test systems. These will also be some of the components that Rei will be utilising to perform her functions in regards to the system processes. So I guess I am getting more than one hour a day on this after all.
A concern is exactly how much paperwork should be kept on this because I do not want to be in the position where I spend more time planning than I do actually developing. At the same time, I don't want to fly blindly into things that I will be forced to correct later because I didn't properly think them out.
a lil bit of privacy in my life...
I was able to get the ssh keys generated for Rei, and get an account in on another server. I haven't put zsh on that server yet, and I am undecided if I really want to as the more Spartan it is in construction, the safer it might be. Most of the scripting will be bash, so for that instance zsh isn't at this point really necessary as it is on the primary machine I am currently typing on.
I also got a first round of code pushed up to the code base, well, for the most part. I have one little issue to figure out, but the rest of it got done kind of to the scale where I had imagined.
Baby steps for the baby project. But a good first day.
We'll start things off this way: