The most important posts in February were life decisions. First, waking up earlier. Three weeks ago I decided to get up at 06:30 each day. The implementation was an alarm, but not in the usual way. I set my alarm on my phone, and put my phone in the kitchen before bed. (I initially thought I would use a separate corded alarm clock, but using my phone is better because I’m forced to put my phone away.) This procedure is helped by a second alarm at 10:00, reminding me to put my phone in the kitchen and go to bed.
The “waking up earlier” change has been successful. My rhythm has stabilized. I have been less successful with leaving work at the desired time. I aimed to leave at 16:30, but have still been leaving around 18:00. The main problem is that evening activities tend to start around 18:00 or 19:00. I wish to find a regular 17:00 activity which forces me out of work.
The second life change in February was to “go paperless”. This is ongoing - I’ve shredded several hundred A4 sheets, with many hundreds/thousands more to go. Going paperless is part of a larger project to organize my home. Don’t hoard. Organize the useful stuff. Remove the crap stuff. Only buy good investments.
A third takeaway from February is that I am not going to become a neckbeard. My company (Pusher) paid for me to visit FOSDEM this month. I wrote a couple of summaries of talks in the “realtime” track, but the main valuable thing I came away with was not any new technical knowledge. It was a warning. FOSDEM was the caricature neckbeard conference. People with no concern for their health or appearance; their life devoted to open source projects which sounded unimportant. I was struck by my lack of care about open source. It scares me a little that my current project - learning C/UNIX/networking - could turn into a neckbeard project. I need to be careful that it does not, and that it is instead focussed on enhancing my career.
Now as for the bulk of February’s posts: they were on C and UNIX.
fork and execve, the UNIX way to call programs in UNIX.pipe system call to call a program with its standard pipes, using fork, execve, and pipe.dup, dup2 and close system calls.mkfifo, the system call for “named pipes”, which are similar to normal pipes but which are accessible via the filesystem path.fork to serve new clients, and the other uses threads, which was my first look at the pthread API.setjmp and longjmp.lsof.I also wrote a few posts on distributed systems: the “happened-before” relation and Lamport timestamps. My intention is to work towards an understanding of consensus algorithms - probably Raft.
A mixed bag. In last month’s review, I said I would …
In February, I’ll learn more electronics fundamentals - voltage, capacitance, resistance, etc. I want to complement this by making real circuits.
This didn’t play out. I didn’t do any of this. I don’t regret it; the things I learned are probably more useful. I bought an Electron device from particle.io, which I want to turn into a bike tracker. I haven’t started with this project.
In February, I’ll explore more C fundamentals - such as compilation in detail (object files, assembly). … I’ll learn more UNIX fundamentals - how time-sharing/processes are implemented, and more fundamental socket/networking programs.
I did some of these. I want to concentrate on the socket/networking programs. I think this is valuable knowledge: for my work now and for my future career.
In February, I’ll learn more WebRTC fundamentals - ICE and SDP. I’ll make a few “hello world” programs. Soon I’ll show the canonical chat program using Pusher as a signaling server.
I did not do any posts on ICE or SDP; those are still the next posts I plan to write. At work (Pusher), I ran a WebRTC evening. We made a clone of Spaceteam. I plan to run another evening in March, probably around adding audio/video to our clone.
Last month, I set these themes: electronics, C, UNIX, networking, WebRTC. These themes are supposed to teach me the fundamentals of real-world programming. They should provide the basis for any future concrete projects/businesses, or for a long-term career in consulting/contracting.
However, these topics alone do not satisfy me. I am also choosing a project for this month as a concrete guide. This project is Vidrio, my screen-sharing program. I aim to have it released by the end of March.
Next technical things to learn:
openssl.lsof posts.I wrote this because I felt like it. This post is my own, and not associated with my employer.
Jim. Public speaking. Friends. Vidrio.