This is the weekly thread to discuss what you have done recently and are working on this week.
Be descriptive, and don’t hesitate to ask for help!
Inspired by reading OpenBSD Daily blog post, I’ve installed OpenBSD in virtualbox. This is my first try with *BSD. So far so good. I have started by reading the basic commands implementations.
Make sure that if you use a remote terminal to talk to it, you set the font to Comic Sans.
Getting ready to go from breadboard to protoboard on a RPi ZeroW based lake temperature monitoring system. The nodes use cheap Dallas one-wire based temp probes to measure the temp of the air, the surface, and 1m down. Results are reported over 802.11 to an MQTT broker for monitoring. Its working, just need to get it all to fit in a weatherproof case. We’re expecting rain later in the week, so I see an afternoon of soldering and hot-glue in my near future.
I’m having a hard time getting signal down to the water due to a line of sight issue. I’m thinking of setting up a BATMAN mesh to reach the probe stations, but it’d be simpler to have a single hop. Can anyone recommend decent access point hardware that boots OpenWRT, can run on POE, and can survive either outside or in an attic? If I had an access point up on the roof, I’m sure it shine down to the lake…
On a different subject, The Last Outpost MUD is turning 25 years old on July 9th. Stop by and try it out if you feel like playing a little bit of retro text adventuring!
Finally got a real job after bouncing around for the last year, as I wanted a bit more structure in my life. Greenfields multi-DC stateful systems are fun :)
For fun, continuing to work on my rusty lock-free flash-sympathetic b-link tree and planning out a new project that uses symbolic execution + linearizability checking + writing distributed algorithms as state-machine transitions. When you have an algorithm described using this state-machine library, you can then check it for linearizability using symbolic execution and then plug the same code into a high performance RPC system. I have no idea at what phase this effort will break down, but I’m super interested in seeing how far it can get.
That’s some neat stuff. Keep at it and submit any interesting developments. Sure some Lobsters would enjoy reading it.
Keep posting about ways people increase confidence in systems! Your posts are definitely spiking my curiosity in this area :)
Appreciate it and will do!
Converting an application from a manually-deployed process to a Docker-based continuous process.
Personal: I’m trying to get back in shape by following the (in)famous Couch-to-5K program; I’m at the end of the fourth week, and to be honest I’m feeling better than ever!
I’m trying to write more on my blog, but finding something interesting to blab about is harder than ever nowadays.
University: I’m studying Turing Machines and finite-state automata theory: very interesting arguments, but hard to understand. As a project for another exam, me and a colleague wrote a small paper about differences in approach and performance between Go and C+MPI; turns out the good ol’ C approach to parallelism isn’t always the fastest option available (which was an unexpected conclusion, to be honest).
Hacks: since the cryptocurrency mining world is getting interesting again, I might publish a service I’ve been working on for some time that could help miners to watch on their rigs more easily… But first, exams!
Great job with your consistency on the C25k program. You’ve already cleared the hard part.
I am writing up my internship report, so pretty boring work. The report
needs to be done on friday, since it’s a necessity for my Bachelor’s
which I want to register on this friday as well. Unfortunately I am not
close at all to the minimum limit in pages so there’s going to be quite a bit
of overtime this week.
Maybe I’ll fiddle a bit more with D3 if I have
time, it’s a really interesting approach to visualization.
I am working on a search engine to visualise all of world’s knowledge with user curated mind maps.
Rewriting a crappy interface to use the provider’s updated API.
Personal: working on myself, trying to eat healthier, be active, etc. I’ve been playing around with more Python and trying to find more things to do with it. I’ll probably find something to add to my once-simple-blog-app-now-quasi-framework in PHP.
Work: Converting our site to pure HTTPS. There’s a lot of hard-coded URLs in the codebase so this will be a headache. As I clean up other people’s messes, I do wonder about the ones I left behind.
Contract: push my PoC and dev into production. Just about done this work, although it often gets side-lined by my actual job.
I’m out in the world looking at jobs, and it turns out that Toronto is pretty heavily into neural networks and other forms of machine learning? This is convenient. Otherwise, just trying to get my knee better so I can start cycling again so I can start losing weight again so I can start adjusting my diet to fix my mood. For want of a patellar tendon &c. &c. &c.
Looking for a new job after layoffs.
If anybody’s looking for a web developer that can work out of Las Vegas, drop me an email.
Since last week I had a talk accepted at the AISB Serendipity Symposium that’s this Thursday, now actually making the talk. Getting accepted to conferences where submission is by abstract feels a bit like going into debt: if they accept this 250 word proposal, now you owe the fleshed out version! My friends in the humanities and social sciences are used to it, but computer science conferences usually have you submit full papers up front, so you’re not promising you can deliver a full paper before you have at least a draft of it.
In other research stuff, working on a paper to submit to an interactive narrative conference, a Rube-Goldberg-machine generator. Hopefully will have something more to share about that soon.
In leisure activities, trying to take more advantage of living in Cornwall when the weather is nice by doing day-hikes on portions of the South West Coast Path. Although I don’t have a car, so it requires strategically planning start and end points to link up with transit. Fortunately the bus service is reasonably good relative to the sparse population, not too frequent but at least extant even in many remote areas.
P.S. - I often notice more people posting in than upvoting the weekly what’re-you-doing thread. If you find it useful, consider upvoting it so it gets more visibility!
Following up on last week:
In learning about the CE[S]K machine I came to the realization that it is ‘merely’ a formalization of a trampolining interpreter, which are common in Lisp implementations, and show up elsewhere, including Smalltalk. Because the CEK machine has small-steps semantics, it effectively reifies the entire control mechanism, thus allowing first-class continuations to easily fall out. The nice thing about this realization is I have more terms I can Google. Adding the store component to the CEK machine lets me implement garbage collection, which is really neat.
In the past, I’ve implemented an interpreter using denotational semantics. This alternate approach to interpreters is equally mind-expanding and beautiful.
I’m working on a Rust implementation of this. While it can still be considered computational substrate (e.g. nothing immediately usable), it is pretty neat to watch my machine tick through reduction rules to find an answer.
I am presenting a poster at EuCNC in the really Northern Oulu Finland. If any lobsters are in town message me and we can meet up for a beer.
I am trying to get some work done while on the road, I have managed to finish analysis of a tcp modification for a colleague. I am hoping to do some work on a wifi driver now the tcp stuff is out of the way.
I’m reviewing/grading proposals submitted to the OCaml Workshop and the ML Workshop (ML is a family of programming languages including OCaml, ML, F# and, depending on the context, may even encompass aspects of Haskell, Rust and Scala), which are going to happen consecutively on September 7th-8th in Oxford, UK, colocated with the International Conference on Functional Programming. The OCaml workshop is more application-oriented (cool new libraries or tools for the OCaml ecosystem), while the ML workshop is more theory-oriented (it’s about research ideas that would apply to several different programming languages). Both have incredibly exciting submissions this year, and I’m already eager to actually see the talks.
Easy reading week. Main goals are to complete Deaths End by Cixin Liu and Mindshift by Barbara Oakley.
Also sometime during the week, the construction of a new dip station for calisthenics