This is the weekly thread to discuss what you have done recently and are working on this week.
Please be descriptive and don’t hesitate to champion your accomplishments or ask for help, advice or other guidance.
In an effort to decrease my smartphone dependace, I’m starting a project to print out all relevant information I need on a receipt using a thermal printer. The goal is to use a raspberry pi to gather stuff like
Has anyone here have any experience with Adafruit’s thermal printers? Is there one should pay attention to, or some common mistakes one can easily avoid?
I’ve not used Adafruit’s printer’s but I have used the things they appear to be based on. I found this quite handy.
I’ve seen a few of these projects before - I can’t find the source for this one, but there are some photos on this account: https://twitter.com/paultag/status/966786313662935046
Drafting two articles for beginners:
Hopefully your tmux article will finally be the one to make me switch from screen. Old habits die hard.
If screen works for you, maybe you don’t need tmux. :)
I’ve never tried screen. tmux(1) was my first terminal multiplexor on macOS and now on OpenBSD I don’t have screen(1) in base.
This week, I’m mostly:
If I get all that done I’m working on migrating Rawhex over to static hosting and then I’m going to write up the gitea install as there were a couple of gotchas.
Incidentally, does anyone have any pointers to how I could provide some updated docs for an OpenBSD port? Do I just contact the maintainer or is there an official process?
I’m working on a website of Ted Nelson words (some people call them “Nelsonisms”) with definitions. He coined a lot of words, some stuck better than others. https://fangl.es
I also set up my own gopherhole with pygopherd on FreeBSD that im going to expand a bit. You can view it via HTTP proxy here, and there is a file for how I did the setup if you want to make your own, http://gopher.modem.city:70
shouldn’t your gopher server listen on port 70, and your web server listen on 80?
pygopherd actually appears to multiplex HTTP and gopher traffic on the same port! I don’t have port 80 open to reverse proxy it, but you can successfully open both HTTP and gopher connections on 70 :)
Ah, good to know! 70 outgoing is blocked at $work, but it’s not a big deal.
If you can’t access port 70, there is always the floodgap proxy :), https://gopher.floodgap.com/gopher/gw?a=gopher%3A%2F%2Fgopher.modem.city
Kind of meta-productive: put online a paper-reading log with notes on papers I’ve read, so I can find them again and remember what I found useful/interesting in them. Mostly for my own reference, but maybe interesting to others.
I realized I had mostly been externalizing this kind of thing on social media—Tweeting out papers, submitting them to lobste.rs, etc.—which is fine, except when it comes time to find a paper again. Trying to remember keywords I used in my tweet so I can use Twitter search to dig up a paper I remember reading isn’t a great UI for searching research notes. Hence I started a local log in March, and now got around to rendering it to HTML, plus backfilled it with as many old Tweet threads & Facebook notes as I could find from last year.
re paper-reading log
I thought about doing that myself for the CompSci stuff. I look at way more than I post. Someone might find some of it useful but I wouldn’t want to flood an aggregator with it. It’s cool you’re doing it. Some interesting stuff there. :)
I’m preparing to start first end-to-end test in federation for my current project and finishing up the ORM switch (Pure Postgres ORM instead of a generic one). Also I’ll have to finish up IndieAuth to get some other features working and tested. Additionally both points above will require me to flesh out the event system I’m working on, which would make a lot of future work on federation easier by having the endpoints simply send (and optionally wait on) events to create comments, responses, posts, etc.
Otherwise I’m digging into LISP again as part of a course assignment at Uni. It’s been a while since I seriously wrote lisp but I enjoy it.
At work I’m going to be on a hackathon project to improve our CI/CD system.
Outside of work I’m reading up on Prolog and learning the McClim GUI library for Common Lisp.
Customising an existing CI/CD or improving an in-house system?
Detail work on my personal site, namely working on weird viewport size reckoning / scrolling behavior on mobile browsers. Also considering dropping in ServiceWorker caching and friends for offline access / buzzword practice.
Just a heads up: it scrolls really weird on iOS. I suppose it might be because of your implementation of the “sticky” header? I can’t look at the code right now, but I’d recommend either making the header sticky using position: sticky; top: 0, if that is the problem, or using -webkit-overflow-scrolling: touch, which would make it scroll more naturally in its current state.
position: sticky; top: 0
Thanks—I was avoiding position: sticky on purpose, but -webkit-overflow-scrolling:touch looks promising.
Update: That fixes scrolling, but breaks sub‐header re‐expansion when scrolled to the top. iOS sure is a treat to develop for when you don’t own a Mac.
It feels a lot better now! Another problem, however, is that Safari’s bottom bar obscures the bottom of the page – for example, the submit button on the contact form. If you wait a second at the bottom and try to scroll again, you can reveal it, but it feels a little wonky. If you don’t want to redo the scrolling mehanism, a simple solution might be to just add a small margin at the bottom?
Tested drm-v4.15 for FreeBSD, turns out it works — even with AMDGPU DC! (On Polaris, but looking forward to someone testing if Vega works.)
Updated weston-rs to use libweston’s new head-based output API.
Updated the SDKs/dependencies of freepass’s Android version (which doesn’t really work yet, but now I can work on it again maybe).
Its tempting to try and add some self-hosted servers because of this thread.
My biggest issue is that the bigger servers don’t usually explain what they are doing, even when its simple. So each one is a lot of reading to try and configure to have it do what I actually want it to.
I’m still working on open source project, nuster, adding major feature.
Still working on my rewrite of my text-browser game PDB. It’s getting almost where I was on the rails version, I’m maybe at 50% done.
Maybe try to work a bit on Cardshifter (Open-source Card game like Hearthstone) since we started to play test it, a lot of issues have surface.
I am working on a small side project in Haskell. A clone of PostgREST with less ambiguity in the interface and EXPLAIN-based VIEW introspection. So far it has been a lot of fun, especially the part where I’ve embedded Aeson in my own query parser.
I rather like the idea of using a single DB schema for the public API, letting an universal application server to take care of details. PostgreSQL is a very powerful piece of technology.
I have started hacking on the ODMA sample code. My eventual goal is a Notepad++ plugin for the document management system that I use at work.
Mostly database migrations to fix “broken” or badly formatted data (eg where the legacy 0 date was being used in mysql instead of NULL) and add foreign keys, so that we can be (more) certain about the integrity of the data after migrating to the new cluster.
So far in one day that’s meant:
filing a bug and initial PR to add SSL (really TLS) support to the python migration tool we use;
tracking down an issue when installing percona cluster after having had percona server on the machine.
The migration tool shenanigans has reminded me that I really need to write my own migration applier (thats all the python tool does: apply up or down scripts and record/remove that state in
A db table) to give better feedback. It’d also reduce the number of weird pypi messages I get on setup all the time. So maybe that will happen this week too.
And finally I discovered why the migrations were giving weird errors. The migration tool we (currently) use, uses glob() to find the SQL files to run, but (before I patched it) wasn’t sorting the returned results, so they were being run out-of-order.
This was the straw that prompted the camel to build his own back brace… I think I mangled that analogy. Anyway, I’m sick of patching what seems like rudimentary functionality in someone else’s tool, so I will be writing a new one. Watch this space: https://bitbucket.org/koalephant/mallard
Primarily? Trying to keep my head on straight.