Bootcamp - Investigation One
in Metis
I gave the presentation for the first of two Investigations today!
The investigations have started.
No… it’s not the Spanish Inquisition.
Every student in the cohort has to give two presentations on “Investigations” at some time during the Bootcamp. I believe it was the first or second day when the instructors pointed us to a schedule on Google Docs. We had to sign up for two timeslots.
I dutifully ignored this.
I mean how could I even think about researching some random topic in the midst of all the pressure related to the projects!?! and the Challenges!!
Somewhere in the past couple of weeks, I caught a whiff or a conversation between Alice some of the students. I didn’t quite catch what prompted the discussion. But it seemed to be something along the lines of someone commenting about all the things to do and the struggle of the pressure relative to scheduling and prioritizing.
Alice provided this order of prioritization:2018-01-24-Project-Luther-Begins.md
- Projects
- Blog
- Challenges
Now… that doesn’t appear to mean just jettison stuff… No, we got a brief (and supposedly one-time) bit of chastisement from David this morning on submitting things on time as expected. And none of that could have pertained to the Project since we all do that stuff (MVPs, standups, Presentations, etc.). The only exception so far has been Amy who was out sick on the day for Presentations on Project Two and she gave her presentation belatedly after I presented my Investigation.
So… you can run, but you can’t hide. David pestered me last week so I finally signed up for Investigations. I chose one early date and one late date since I plan on one “easy” and one “hard” topic. OK. Truth be told, you wait to be almost the last one to sign up, you no longer have choice of slots.
Docker was on the list of suggested topics, so I jumped on it given my love for things related to virtualization and my experience playing with containers of various OSs.
I stayed with reveal.js and like before I tried to explore some new functionality of reveal.js while doing so. I used the “cube” transition which, I discovered, locks you into that for the full presentation - no per-slide overrides. It seems to have to with how both sides of the transition have to work with the format.
But the real fun was incorporating “shellinabox” which permitted me to include a terminal within the presentation for the purposes of a live demo. This worked well. It was “cool” to have the terminal be a part of the cube transition. What did not work as well was what would seem to be simpler - just incorporating websites in iframes in a slide. I mean… that’s a precursor or dependency of the functionality for shellinabox!! But I just could not get it to work. I did manage to use something so I bounced from the presentation to website and back rather seamlessly… by using some hack for debugging that’s been left in reveal.js as an undocumented feature. So I did keep in the spirit of not exiting the presentation. But still… sheesh!
Another “fun” aspect was that earlier in the day Alice walked the group through installing postgresql in EC2 instances on AWS. From there we were supposed to be able to pull the data into local Jupyter notebooks. We had all sorts of problems because of some odd postgresql issues. It was funny to use as a demo an example of firing up a local Docker with postgresql and have it all work immediately. Of course, this was a bit of apples to oranges since the work to get that to work with Juptyer would be very similar.
This was the “easy” investigation. I have high hopes for the next one to be much more interesting.