Actually this is the biggest hurdle in leaning how to code. You can blame the huge numbers of “learn to code in 24 hours” articles and videos online and the the influx of “5 day bootcamp” courses. Its like teaching someone the basics of how to drive a car but never teaching them the road rules and never taking them on the road.
A better analogy might be learning a foreign language. It’s like teaching someone all the words in Spanish, but never putting them together in a conversation.
I’d argue that if you say “I know how to code, I know what variables are and how to print text to the console, how do I make an app?” Then actually you don’t know how to code. You might know the basics of a programming language, and that is the first step in learning coding, but there are many steps after that.
I identified this gap a few years ago after seeing a couple of my friends (one finished a boot camp, and one finished a software development major at Uni) both were in this same situation. I determined there is a big gap between “knowing a programming language” and “knowing how to make software”. It’s like going from “I know how to write words” to “I know how to write a novel”. It’s not something that comes easy. It’s something that can take time (often years) to get good at. This is the reason you see requirements like “3 years software development experience” on entry level programmer jobs. The number of people in your situation is incredibly high. The coding bootcamps churn them out by the hundreds every month.
A couple of years ago when I was between jobs, I created a Gumtree ad advertising “post-bootcamp” courses, that aimed at bridging this gap. It was a series of private 1on1 lessons aimed at teaching someone to go from “knowing how to code” to being “software developer” job ready. Lots of people have many different learning styles and different paths they took to this point. The key is focussing not on the giving them the missing information, but teaching the person how to identify what steps are missing and how to find resources to learn them (because that’s the real missing knowledge wink).
Unfortunately I found some people didn’t want to learn how to learn for themselves, and just wanted me to hand them the “secret missing parts” on a platter.
I feel like the average consumer uses their Android TV boxes the way that they come.
The average consumer doesn’t buy android TV boxes from AliExpress. They use a Google Chromecast or a NVidia shield, or Amazon Fire TV.
The people who buy these devices from China are those deliberately looking for specific hardware to use for a specific application.
This is the first time I’ve even heard of CoreELEC despite using LibreELEC. Thanks for mentioning it. I have doubts every obscure cheap Android box is supported though.
LibreElec reduced official support for SBCs with Amlogic Cpus (like the Odroid C2, Odroid C4, and Odroid N2) in 2018, that spawned the fork called CoreElec. Then LibreElec removed Amlogic support entirely in 2019 (they wanted to just focus on Raspberry Pi SBCs). That caused a mass exodus of users and most moved to CoreElec. That was around the same time cheap TV boxes started appearing on AliExpress, and a lot of them happened to have Amlogc CPUs like the s905X, s905X2, s905X3, and s922X, these are the same CPUs in Odroid C2 and Odroid C4 and Odroid N2, so CoreElec was able to add support for most of them.
CoreElec remains tied to LibreElec upstream, receiving the same updates.
They have a comprehensive, accurate and up-to-date hardware compatibility list. They don’t support all Chinese TV boxes, but if it has an Amlogic CPU, there’s a high chance it is supported. If you’re unsure, just look at any one of the hundreds of “will this cheap TV box work?” threads on their forum.
I’ve owned four different android TV boxes from AliExpress over the years, from different manufacturers, different sellers, and different versions of Android. None of them ever came with malware. I’m a member of the CoreElec community forums where thousands of people own android TV boxes, hundreds of different models and hundreds of different firmware versions, and nobody ever once talked about having malware on their device. That LTT video is ill-informed and out of proportion. Anyway, nobody ever buys the Android TV box to use whatever crappy old version of Android they include, they immediately wipe the partition and install CoreElec on it, with kodi and all the plugins you’d ever want.
I have two of them running CoreElec for my media centres, and one with Armbian OS with HomeAssistant installed, running my home automation. They’re the best bang-for-buck ARM powered Linux hardware you can get, miles better than a raspberry PI.
The only good thing to come from this new editor so far is the frank statement by the original Atom Developers (who invented Electron, just to run Atom) admitted that Electron is not a good solution for a code editor, because who in the heck wants to edit their code in a web browser anyway.
Now we just need to convince the devs of Keybase and Obsidian the same.
I started using Trilium in early 2020, with version 0.40.2. Roam had released in 2019 and was growing in popularity quickly, I heard a lot about Roam, it looked cool, so I googled for an open-source self-hosted knowledge base note taking app with similar features to Roam, like notes arranged in a knowledge graph, and a backlinks explorer for each note. The only one that was available then was trilium. Looks like you’re right, the development of trilium was started in 2017, before Roam existed. This is a great interview with the creator, answers a lot of the questions I had. https://console.substack.com/p/console-169
Obsidian didn’t come out until a few months later (and remained under the radar until 2021), all my colleagues and friends use Obsidian now, but I prefer trilium. I had never heard of logseq before I read this thread. Just a quick glance, I see the first 0.1.0 version logseq was in April 2021, just before the first obsidian release.
It’s not really a standalone file format, it’s executable Lua code.
It returns a new item with the given table contents.
That syntax with the keys in square brackets is the “long-form” method of creating a new table, that’s allows the use of spaces and dashes in the key name.
Maybe this is the lua-equivelent of a python Pickle file?
Lol, that reminds me of when I was in Uni, I had a systems development class, they taught in C, all the lectures, tutorials and assessments were done in C. Our final assignment was handed out the week the first Rust v1.0.0 build dropped in 2015. I had been following the hype around the development of Mozilla’s new language, and I was so keen, I asked my professor if I could complete my final assignment using Rust. He said it’s a great idea. Then cut to me furiously trying to learn Rust in just two weeks, so I could even start the assignment, including C interop, implementing functions with c-style interfaces for callbacks, and lots of unsafe blocks for memory manipulation and pointer manipulation. In the end I was just forcing Rust to be C.
It did work in the end, and I did get an A, mostly because the professor couldn’t understand any of the Rust code.
I was using Inconsolata for about 5 years, then switched to Inconsolata-g when that came out, for another 5 years. But it’s a pretty old font and is TrueType and it’s hinting is bad, so doesn’t render well on Linux and it misses out on a lot of new font features.
In 2019 I went hunting for a new favourite font, and tried out a whole bunch, giving each one a week in my IDE to really get to know it. During that time I realised I had a bunch of basic requirements for a font that some do better than others:
Similar characters should be distinct: eg, uppercase O and number 0. Uppercase I, lowercase l, and number 1. It’s weird how many popular coding fonts fail to make these clear.
Not too wide, and not too narrow. You’d think monospace fonts are all around the same size horizontally, but a standard 80-column slab of code can vary greatly in screen space width depending on the font, some are much too wide. Consolas is an example that is too wide. I like to have the option to tile three code panes side-by-side on a 1080P screen.
Easy to read. For some reason a lot of coding-specific fonts affect my ability to quickly and easily read the code, and some give me a headache.
I realised that my use of Inconsolata for such a long time in the early stages of my career definitely shaped my preferences. I was looking for something similar to Inconsolata. That was when I discovered Fantasque Sans Mono. It’s a kind of weird looking font, maybe a bit too playful for a serious coding font, but I found I could read and parse code much faster (maybe it helps with mild dyslexia?), each letter is very distinct from every other. It has elements of handwriting, it has elements of a dyslexic font, it has similarities to Inconsolata.
I’ve been using Fantasque (with Nerdfonts mixins) for 4 years now. Since then there has been a renaissance of code fonts, like Jet Brains Mono, and Fira Code. I like those, they are good fonts, but I keep going back to Fantasque, it feels so comfortable to use.
I used be a computer technician at a small town computer shop around 2008-2011. More than half of our customers were over 60.
Sometimes I needed to take some tech support calls, and sometimes I needed to make house calls to troubleshoot the folks issues.
Literally every support call started with “Why doesn’t the email work?” while the actual problem ranged from ISP issues, and modem faults, PC faults, Windows configuration errors, to dead monitors or a broken mouse. Literally any computer fault could be described as a failure to access their email.
I am still using the last (ancient) version of Azureus (Vuze) because it has great plugins. I have the plugin to allow it to use the Mainline DHT, and I use the i2p plugin too (because the default i2p torrent UI is bad). Azureus is (was) the only client that can download/seed a torrent on clearnet and i2p at the same time.
Having said that, I didn’t know that BiglyBT has i2p support, so I need to check that out.
I agree with that, mostly. However I find I don’t really ever need to add or edit content on mobile. I only use the web app on mobile to lookup something when my laptop isn’t at hand. There is the official Trilium Sender app for Android that allows you to forward text, pictures, links, etc from your device to your Trilium server, then you organise the content when you get back to your laptop. I find that fills any gaps in functionality. I hate brain dumping or editing long or complex paragraphs of text on my mobile anyway.
Another vote for Trilium.
A couple of years ago Roam Research was trending, I read some articles and reviews about it and I liked the concepts it introduced. I looked for a free, open source self-hosted cross-platform alternative to Roam and found Trilium.
Its native on Windows, Mac, and Linux, while it doesn’t have any Native Mobile apps, the webapp works on great on mobile and can be installed to your phone launcher as a PWA.
It does everything I want, and I use it a lot. A bunch of my colleagues have been recently moving from Evernote or Notable, over to Obsidian, and I understand Obsidian is the new hot thing, but I think I’ll stick with Trilium.
My advice would be to try out a bunch. Note taking is surprisingly nuanced and personal preferences play a major role. Try each one for a week or two, and see which best matches your workflow and your requirements.
Sounds like you, like a lot of others, have come to docker from the perspective of “it’s like a mini virtual machine”. Maybe you’ve used VMs before, like virtualbox or VMware or EC2. Maybe you have experience with setting up a cluster of VMs, each with their own OS, own SSH client, own suite of applications, and an overlay network between them all. Maybe someone told you “you should use docker instead, it’s like mini lightweight VMs”. And you’d be right to assume this is the wrong perspective to approach docker, because it leads to the problems you have faced.
Instead, try to think of docker containers as standalone applications. They don’t contain a kernel, they don’t have SSH, no Nano or VIM, just simply the Application, in a container, with enough supporting filesystem and OS libraries to make the application work.
That perspective is what helped me to get better at docker, I know it’s not exactly answering your question, but it might help.
Wow, I was looking for something exactly like this only last week.
I use alacritty as my terminal emulator, and I normally have three or four alacrity windows open to different remote hosts. I was thinking how good it would be to have a GUI tool that lists all my remotes and allows me to open alacritty directly to a SSH session on a given host, and I even thought it would be great if that GUI tool had file management capabilities, so I didn’t need to have FileZilla open for each host too.
You’ve made my day. I’ll try it out on Monday.
TL;DR
It is good for writing cross-platform scripts. It’s easier and more powerful than Bash but not as heavy and cumbersome as Python.
Raku is the new name for Perl 6. They changed the name because there are backward incompatible changes in the new version, and they wanted to try to move away from the negative associations people have with Perl.
Perl has always been better than Bash at scripting, and Raku added some modern features to take that advantage further.
Trillium is a full featured configurable and programmable self-hosted note-taking app that can be easily configured to suit the use case you’re describing, it does categories, tags, links to other topics etc.