The Art of Not Asking Why @jtr-taonaw - Tumblr Blog

Meta Journal notes in Denote-Journal with Journelly

Wow, I’m so good at overcomplicating things, I already made you go “huh?!” with the title alone! I have some mad skillz!

On with the show:

In the last couple of weeks, I slowly improved and tinkered with my Linux environment (Kubuntu). I’m not sure if it was a single thing that nudged me to do that, but between installing the latest version of Harper and switching to Vivaldi with its great features I keep discovering, I also started using denote-journal.

Denote-journal, from the prolific Prot who also made Denote, was something I wanted to try for a while. I didn’t really have a good reason to, because I’ve been using Journelly for a long time, and on my Mac or Linux desktop, I’d call a capture template that would append to the file, adding my additional entries.

But I always had one major issue with journaling on my iPhone: privacy. My iPhone is owned and regulated (with a system profile) by my workplace. So even if I’m fine writing personal notes on my Mac (I’m not, I’m iffy about it as well, I don’t trust Apple to respect my privacy much more than I trust Google), at the end of the day, these notes also sync to my iPhone.

On a day-to-day basis, for quick thoughts and work notes, Journelly and my iPhone are great, but when it comes to writing longer notes about my future plans, how I spent the weekend with my partners, and basically anything else that involves people whose privacy I respect, I always self-censor.

For this reason, I came up with a way to create private notes with Denote only on Linux. These notes only live on my Linux Desktop, with a few of those syncing to my Android1 (again, not great, but at least it’s not work-managed).

I always feel more like myself on Linux because I am more myself - whether it’s customizing my shortcuts and workflow exactly how I like it, or if it’s the built-in privacy that can be further fortified and inspected. Journaling on Linux, without the world’s biggest nanny peeping over my shoulder, is where I really open up and write my most personal thoughts.

A couple of times I looked at my Journelly notes, those that I wrote on the Mac where I could write at length and use the full power of Emacs, and compared them to my older journal on Linux. Well, there is no comparison. And while creating a private note in Denote in Linux and linking it back to the original works2, it introduces friction that hinders the flow of my thoughts.

So I looked into denote-journal, realized it’s very easy to use, and gave it a try one day, and since then:

Essentially, it’s what I used to do when I started using Journelly: refile my Journelly headers into my journal files when I get the chance. Or, as a matter of fact, I use org-refile-copy, which does exactly what it says, because I want to keep the original in Journelly. Since I now use individual journal files instead of a big file split into weeks (as I did in the past), this process is even easier. The only catch is the images.

Images in Journelly are saved in its /Journelly.org.assets folder, where my journal files can’t see them directly; and even if it did, these images are too big and cause freezes, and also oriented the wrong way, so they need at least a minimal treatment. For these reasons, I attach each image to the correct header in the journal daily file which I moved over from Journelly.

These journal notes are fantastic. I have my quick notes from the day available, but when I want to extend, all I do now is just write a new entry for that day (which looks just like the Journelly entries) and write to my heart’s content. There’s also a bonus: denote-journal allows me to make up for days if I didn’t create a “meta journal note” for the day (so far this happened only once) directly from the calendar, so if I miss a day, I go to the calendar, point at the missing day, and use denote-journal-new-or-existing-entry to take care of things. If I already have an entry, it jumps to it; if I don’t, it creates one.

Between my journal on Linux, the emails I write to other bloggers, my blog posts, and the occasional instructions I write in Denote, I think I write more than I did in my entire life. I’m thinking I need to start capturing it in some sort of book, though I have no idea what it will be about and how to edit my writings in a way that makes sense. This is only a vague concept at the moment.

Footnotes

1 : I keep skirting around this issue so I’ll just mention it quickly: I used GrapheneOS in the past, and it’s great for these kinds of things, but GrapheneOS protects your phone to an extent that certain apps don’t work.

2 : I’ve been doing this often enough that I have a whole tag in Denote called “supplemental” with additional thoughts and notes. In Journelly, I was just linking to those, and writing something like “I have more to say about this” and this would include a Denote link to the Linux-only note.

Well, hello, Spider-Noir, welcome to Binge City. Where have you been? Nice. 📺

Enjoying Vivaldi

So far I’m enjoying Vivladi. Thanks to Jack, I realized I was stuck in a “Vertical Left” present, which - as he says - makes the address bar too short and weird and also hides some of the buttons. At least for now, as I’m learning it, I want to see everything.

The Tab Stacks are better than I realized. I use the accordion mode to get stuff out of the way when I’m just browsing, but when I’m doing one of the usual activities (like blogging from Linux, as I do now), I work within that stack. I really like the fact that my blog’s stack has my blog’s colors. It makes it stand out from the rest and invites me to write a post.

Also, split-screen view is especially useful when blogging: I write on the left, while I have references and other tools I use (like Inkwell, or a site I’m quoting from) to the right. It may seem small, but the fact that I don’t need to switch tabs makes blogging much easier; on Safari, I drag a tab out of the pack to create its own window, but it feels forced, it slower, and it doesn’t dock next to it as nicely.

There are probably a million other features I don’t know about. In fact, let me ask you, if you use Vivaldi (especially on Linux): what have you discovered? What’s one thing or maybe two you use often and consider a must-have?

It seems like the the windows app (RMD) is messing up Emacs. It constantly freezes.

The issue seems to be with copy-paste. pboard (clipboard) doesn’t handles RMD’s clipboard well.

Screw RMD. I will just use my old setup (run locally from the Mac). Every day I likr Microsoft even less.

Sometimes you just gotta laugh at yourself.

Interview with an Emacs Enthusiast in 2023 with Emerald McS., PhD

People don’t quit Emacs. They just die at some point.

Brilliant.

Play Subnautica again: why cheating sometimes make things better

I mentioned I picked up Subnautica again, and it’s more fun than I remember it was my last time with it. Why? Because I’m kind of cheating. Hang on, hang on, hear me out….

Subnautica is not an easy game for the easily distracted. For those of us who constantly forget what we were doing because there’s a new shiny thing to focus on, Subnautica can be brutal. A large part of it is survival and crafting, which means my internal monologue goes something like this: “I need some copper for the… oh wait, if I get more silver I can craft the wiring kit for that torpedo, but I need aerogel… what do I need for aerogel? Oh, gel sacks, right, OK, where’s my Seamoth, ha, I forgot I docked it outside instead of the moonpool… oh crap oxygen! swim up! up up up up…! Whew! Wait, where am I? Where’s the damn Seamoth? Ah! Here you are… good, now what was I doing? Silver? Or was it gold? I wanted gold for the CPU so I can craft… oh right! I need copper.”

The game’s a lot of fun, and it rewards you for exploring, but when your brain constantly works like the above, you tend to miss a lot. This is why I started looking online for things like where to find X, and watching a few helpful YouTube videos. I also learned which plants I can grow around my base so I can craft what I need more easily, instead of running around and forgetting what I’m looking for. Because I played about 2/3 of the game before, I know the basic plot and what’s going on, so I’m taking my time now preparing for each new horrifying level and enjoying building my base and putting more details into it.

Another very helpful aspect of searching for answers online is finding the parts for blueprints I need. The game gives you recipes, or blueprints, for different tools. Some of those are required to get to a certain point; otherwise, you can’t progress. Certain materials are not available to you until later in the game as part of how it’s built, which makes sense, but it means I can tell when something fairly basic is missing. In order to find blueprints, you need to scan certain wrecks and parts in the game, and you need to find where these parts are. They tend to be in certain areas, however, where exactly in those areas is random.

For example, I needed a propulsion cannon to push crates that blocked my way to a place I needed to go. I knew I needed to go there because I was at a point in the game where this is indicated by the audio logs I picked up, and also, I played this part before. I was looking at the ingredients needed for the cannon and noticed they’re very basic, the sort of things you get really early on. However, I only had one part of the cannon scanned; in order to get the full blueprint, I needed to find one more part, which I never did. Because I knew I needed this cannon at this point, I looked online to see where this part is located. It still took a bit of time, as some YouTube videos are more helpful than others, but I was able to find it because I knew exactly where to look - and even then, it was hiding in a tough spot to reach (again, not by design).

If you ask me, the randomness of where these parts can be a bit of a design flaw, especially when something is trivial to the game, which is exactly why this sort of “cheating” is justified. I got what I needed, and the game flowed exactly as it was designed, with everything else clicking into place. But not all cheating I do is the “good kind” like the above; some of it is cheesy. Subnautica is a scary game, the sort of game that freaks you out with the unknown. And it’s good at freaking you out, believe you me.

Because I played the game before and had already seen most (maybe even all) enemies, I took a peek at some fighting strategy videos against the hardest creatures to take some of the sting out of the fear. I wanted to be ready and enjoy without being scared all the time, ok? Being a bit scared is enough. It also doesn’t help that the game only has exactly one save game slot per game either, so if you really mess up somehow, you’re screwed. Fortunately, that doesn’t happen too often… but it can. Which makes me worry. Which takes away some of the fun element. So yeah, some fear control goes a long way. Exploring is still fun and stressful in the right amount that pulls me back.

See? Sometimes a bit of corner-cutting helps.

Trying out Vivaldi on Linux

The last time we left off the browsers saga, I was trying Zen on my Mac, which started to annoy me pretty fast, so I got back to Safari. Meanwhile, on my Linux Desktop, Zen was my browser until a couple of days ago.

Zen has been working nicely, but over time I’ve become more aware of its glitches. Most of those are unique to me, as I like to overcomplicate things.

The main issue I’ve had with Zen is Flatpak. Usually it’s a good way to install apps, but with something as integrated with your system as a browser, this can cause problems. Case in point: KeePassXC, my password manager of choice. KeePassXC requires integration that flatpak blocks by design. To get around that, I tried to install Zen directly from a tarball. This is a manual installation, which means I need to download and upgrade my browser manually. Browsers get updated very often; I simply stopped updating Zen, and that’s just plain bad.

From there, things started to feel a bit flaky. For example, I never got my tabs, bookmarks, and passwords to fully sync between the Mac (when I had it installed) and Linux. It seemed to always freeze halfway, with some items synced and others not. Meanwhile, when I wanted to start using my Linux environment for writing and blogging more, I just couldn’t get Grammarly to work on Zen. It refused to show the sign-in pop-up window, no matter what I tried, from allowing pop-ups to reducing security to basically zero.

Meanwhile, its main features (like spaces) didn’t interest me. On Linux, I have this problem resolved with activities, and for work, Edge works best. I have Edge on my Linux desktop too if I need to use it, and it works better on Linux than it does on the Mac.

It feels like Zen is meant to be a Safari replacement, but I don’t need one. Where I need organization and a browser running in full screen with tabs organized is Linux, which is why I kept putting up with Zen until I tried Vivaldi.

Vivaldi has a .deb installer that you can integrate into apt so it updates smoothly with everything else. Meanwhile, Grammarly worked as soon as I installed it, and KeePassXC has a dedicated Vivaldi extension that integrates without a hitch. That’s already three strikes against Zen without even breaking a sweat. But what really made me smile were its tab stacks.

Tab stacks (or tab containers, or tab groups, or whatever you want to call them) exist in different browsers, but Vivaldi lets me use a CSS file to customize these containers with the color and fonts I want, along with whatever CSS decorations I want. It feels a little bit like a hack, but it’s not: you need to turn on the flag to allow you to change Vivaldi with your own CSS, and then you just create the CSS file with the choices you want. It might be a bit of an overkill, but having a tab stack dedicated to my blog and Micro.blog with my blog colors is a very nice touch. This CSS option exists for every part in Vivaldi, if I feel like customizing something else.

While workplaces got deleted fast (a simple right-click on Zen; it was a flag hack), other features I thought I wouldn’t care about, like the built-in email client in the browser, are getting more interesting. It’s actually nice to have the same email interface for my several email accounts without the heaviness of Thunderbird, and I think using their calendar might come next. It’s easy to add and remove buttons I don’t need, so that Kagi’s assistant has a dedicated button, for example, while something they think I need to use (like Notes) is gone with a quick right click. I appreciate applications that let me decide what I want instead of deciding for me even more these days.

Vivaldi also has an official Android version, where I actually need a better browser (I’ve been using DDG’s browser, which is meh), which means I could sync tabs between my desktop and my Android if I want to. It’s not that I really miss these features, but it’s a nice head nod, a way to make things a bit easier.

So, we will see how this holds up. Maybe I’ll go back to Firefox in 6 months…? Who knows. For now, Vivaldi is my default browser on Linux.

Correcting photo orientation for org-mode in Linux

Here’s a niche problem: When viewing iPhone-captured photos in org-mode on Linux, they always appear in landscape orientation, even if you took them in portrait.

The reason is that the sensor of the camera on the phone is physically embedded in landscape mode, so all photos are in landscape mode; when you hold the phone in portrait mode (which is how you hold it most of the time), the phone detects that and implements a fix in the EXIF data file. Essentially, a software fix to a hardware design issue.

Since most photos I take are usually in portrait orientation, it means I need to twist my neck and view images at a 90-degree angle when I look at my Journelly entries.

The source of the problem seems to stem from how org-mode interprets EXIF data in the photo: it doesn’t. It relies on other parts of Emacs, which in turn rely on parts of the OS, to do the job. On Linux (at least on Kubuntu, which is what I use these days), those parts don’t handle EXIF orientation information. Why and how, I am not sure, it’s more digging than I have the time for right now… but anyway - there’s a simple fix.

Auto-orient, which is part of ImageMagick’s mogrify tool. And if you use Emacs on Linux, good chance it’s already installed:

This operator reads and resets the EXIF image profile setting ‘Orientation’ and then performs the appropriate 90 degree rotation on the image to orient the image, for correct viewing.

To execute on an image file: mogrify -auto-orient <file>.

And because I use Emacs, of course there’s a dwim-shell-command solution:

(defun jtr/dwim-image-auto-orient () "Auto-orient images based on EXIF data using mogrify." (interactive) (dwim-shell-command-on-marked-files "Auto-orient images" "mogrify -auto-orient '<<f>>'" :utils "mogrify" :silent-success t))

That last part,:silent-success, closes the empty buffer that pops up after successful execution, as mogrify doesn’t really produce an output window. So, it will just bring us back to Dired.

Traveling to a remote site for work today, got a chance to enjoy the beautiful weather and take a breather.

A good weekend

Last weekend was good. I came to a realization not too long ago that most of my weekends are good, which is a good thing in itself, but I wanted to ramble a little about why.

My sleeping issues (I should coin the term somehow. “My Sleeping Issues” sounds like a terrible album name stuck on repeat at this point) meant that I woke up at 4, walked around for a bit, and went back to bed around 6. This pattern has been going on since last Thursday, when I pushed myself to be in bed at around 23:00. Seems like I sleep 5-6 hours no matter what I do, so if I go to sleep earlier, I wake up earlier without enough sleep. The good part? Going to bed around 6 again means I’m not a complete zombie the whole first half of the day, which definitely played a role here.

First off, there was a lot of Subnautica. With Part 2 out in early access, I saw Subnautica: Below Zero was on sale on Humble Bundle and snatched it. I never finished the first Subnautica (it was too scary, and yes, I need to get this post over to the new blog), and I wanted to try again. Man, Subnautica is a treat. It was easy to get hooked again. This time around, I watched plenty of YouTube videos to help me out with tips, and building my underwater base was fun in a relaxing kind of way. There’s something about being able to organize a stressful situation that doesn’t let you breathe (literally in this game) into something safe that makes sense to you. And, yes, that’s also a metaphor.

I mentioned before I want to make some video content for gaming, but it’s not easy. Or rather, it’s a lot. I am no stranger to OBS Studio, but I had to re-learn to do a couple of things, and even then, going back and watching an hour or more of content and starting editing takes a lot of time. It’s a serious commitment. I haven’t given up on the idea, but I might change it around a bit toward a mix of writing and video content. I’m not sure yet. There are too many good games out there, and I don’t think there are enough places reviewing them from a personal perspective (by the way, let me introduce you to Laura Michet’s blog. She is a game writer/narrative designer, and she writes often).

Then I had a nice hangout with my partner, who works as an EMT. They are awesome in general (obviously I think so), and they recently got themselves a car, and driving in Manhattan is… well, let’s just say parking is one tough SOB boss. Imagine working a 9-hour shift in an ambulance, driving around (on a weekend, because during the week they’re busy with their other job), just to drive another two hours, the last hour and a half(!) of it is in order to find parking. They pretty much showed up just to crash and fall asleep, but then the next day, we went to the Farmer’s market together and drove around and walked the parks of upper Manhattan. It’s a sort of buffer, a break that makes my brain slow down a bit and appreciate my life outside of work and tech.

Which means that when I got back to my /own/ tech on Sunday, I was refreshed and full of energy. This weekend I installed Harper on my Linux desktop. Working through it but especially writing the post about it requires geek brain energy, which I just don’t have during the week. I really enjoy these projects; I learn more about stuff I’m passionate about, and it’s very rewarding.

I also wrote a lot in my journal, which I’m reorganizing, and it made me realize that, hey, I write a lot. Between my posts, emails to other bloggers, journaling, and experimenting with TiddlyWiki, there’s a lot of keyboard typing going on. This, as usual, makes me appreciate Emacs and org-mode, which has its own dedicated category. But that’s what happens when you can customize everything about a tool you use all the time: it’s a game in itself.

And now, with coffee done and this draft more or less complete, it’s time to get in the shower and quickly get some groceries before the work day gets the chance to kick my ass back into gear.

Installing Harper on Kubuntu: The Right Way. Maybe.

I recently installed Harper on my Linux Desktop to work with Emacs, but since I’m running Kubuntu, I ran into difficulties. In short, there’s no Flatpak or Apt option when it comes to Harper.

After a few interesting changes to the way I journal in Emacs (this is something I hope to discuss soon), I decided to go for the full version, and that meant installing Rust and Cargo.

This was another “programming quest” I didn’t know how to start in the past. I used Claude.ai to guide me, but as usual, I asked a million questions about everything, so I can explain it again here (this is my test to myself). So if you’re new to all of this like I am, take the explanations with a grain of salt, and if you’re an experienced developer who understands Rust (and curl, for that matter) feel free to reach out and educate me further.

Alright, here we go:

To install Rust and Cargo with it:

curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh

The above is a bit more complex than what’s in Cargo’s documentation, but based on a quick search, it is what’s directly recommended in rustup, which is where you install Rust. The idea is the same as other curl installations, with a few more options for added security and to ensure we’re getting what we really want:

Run curl, but restrict it only to https (no http):

curl --proto '=https'

Somewhat redundant: curl will usually refuse anything lower than 1.2 by default. This forces TLS 1.2 as the minimum. This is good practice and also what they tell us to use, so why not:

--tlsv1.2

options for silent mode s (so don’t show us progress and status), but show us if we get errors S, and if we get a 404 error or similar, just stop silently f (otherwise it will pipe it into the sh command at the end):

-sSf

Then we have the URL to download from:

https://sh.rustup.rs

and finally we pipe it | into a shell sh command so it runs as a script as intended here. If you go to the above URL directly, it will download a shell script - so this is how we get it and run it in one go:

| sh

Because we’re about to run commands for Rust, it’s a good idea to add it to our source environment, the same as editing ~/.bashrc manually and adding . "$HOME/.cargo/env". Without it, we’ll have to specify where Cargo is installed for the next commands

source ~/.bashrc

At first, I installed what was available on crates.io. Crates, as I learned, is the official repository for Cargo, our “app store” for Rust, (or Elpa for Emacs). The individual packages are called “crates”. Makes sense now, but before it all looked like a bunch of command-line voodoo to me.

However, apperently what’s available on Crates is not up to snuff. The official repository for Harper is at https://github.com/Automattic/harper, and it specifies version 2.3.1, whereas the one available in Crates is 2.0.0. We are still using cargo (it’s the “installer” for Rust), but specify to get what we need directly from there:

cargo install --git https://github.com/Automattic/harper harper-ls --locked

The git option tells Cargo we’re installing directly with git, which is what we’re doing here; the locked option is specified in Harper’s documentation, and upon some research, I learned this forces the exact dependency versions specified in Cargo.lock. Without it, cargo might choose newer dependency versions that were not tested or are not specified in the documentation.

Finally, in Emacs, we want to tell eglot where to find Harper:

(when (eq system-type 'gnu/linux) (add-to-list 'exec-path (expand-file-name "~/.cargo/bin")))

In my case, since I use the same config on my Mac, I want this to run only on Linux. On my Mac, Harper is installed without all these shenanigans directly from Homebrew, which also keeps it up to date. This is added to the same config block I specified in my earlier post. It now looks like this:

(with-eval-after-load 'eglot (add-to-list 'eglot-server-programs '(org-mode . ("harper-ls" "--stdio")))) (setq-default eglot-workspace-configuration '(:harper-ls (:dialect "American" :linters (:LongSentences :json-false :AvoidCurses :json-false))))

;; Besides choosing American as the language, I also want to ignore long sentences (the main issue is that it hides other errors nested in those), and I also want Harper not to tell me when it thinks something is offensive. I'm a big boy/an old fart. The full list of these options is in https://writewithharper.com/docs/rules. It needs to be nested inside the :linters option. (when (eq system-type 'gnu/linux) (add-to-list 'exec-path (expand-file-name "~/.cargo/bin")))

In the future, when I need to update:

cargo install --git https://github.com/Automattic/harper harper-ls --locked

Now, Harper works as it should on my Linux Desktop. Another geeky weekend project.

With recent CSS changes to the blog seemingly holding for now, I’m going for another big one: switching from Hugo 0.91 to 0.158 (the latest available on Micro.blog). Works ok on my test blog… what can go wrong? I’m just waiting for the backups to complete 😬

Started fixing up my CSS.

Earlier, I said I need to move over all my custom-made CSS changes to custom.css from the Tiny Theme’s original CSS file. That’s what I just did. Now let’s see what broke in the proccess.

Using Denote for Email: A manual workflow

As I started to write more emails to other bloggers, the annoyance with macOS' built-in email client grew. It wasn’t just the fact that it has small text that’s hard on the eyes especially on the harsh white background anymore; it just started to feel restricting.

Emacs is my natural writing environment for longer texts, like blog posts or the kind of emails I end up writing.

I’ve considered mu4e before, but setting it up seems a daunting overkill: the place I would benefit from mu4e is work, but I’m blocked by Microsoft-only 2FA authentication, so I have to stick with Outlook; meanwhile, for the three or so emails I write to other bloggers, it doesn’t require such heavy lifting.

One day about two weeks ago, I just fired up Denote, and suddenly it clicked. Denote, when you invoke it for a new note, asks for a directory - so I created an email directory in my parent Notes folder, and started writing. For a title, I use the subject, and the keyword is reserved for the recipient.

Now my eyes thank me again, as some of these emails can take an hour (and more even) to write. Links are a breeze to include, and quotes - which I use heavily in emails - are just a keyboard press away. It also looks nice when I go to the email directory and see all my drafts there, organized nicely as Denote knows how to do.

Since Denote doesn’t handle emails, for this I simply export the org file to HTML, and then with Dired (which opens in the same directory as the note I’m writing by default), I open the HTML file with my browser. From there, I copy-paste into Apple Mail, which acts as a proofread enhancer with Grammarly going to work there (this is something I’d miss if I were to use mu4e, though I could probably use Harper).

It’s a bit of a manual process, and I do need to delete the HTML files from the email directory every now and then, but for now it’s fine. It’s probably easy enough to create some shortcut that will open these HTML files directly with Mail instead of copy-pasting1 though.

Footnotes

1 Opening an HTML file with Dired with ! open -a Mail would make sense, but it opens Mail with the HTML file as an attachment, not as the body of the text.

Journelly and OSM for Emacs are good together

I mentioned OSM for emacs briefly before, but I haven’t played with it much. That’s because the maps never showed up correctly in the buffer: the map tiles were not aligned correctly and some appeared blank.

As it turns out, someone else had this problem and also found the culprit: visual-line-mode. I have it turned on by default as the majority of my work in Emacs involves org-mode and I need my lines wrapped in the buffer. With visual-line-mode disabled, OSM works as expected, including zooming in and out. Good stuff.

Now that I fixed OSM, I was wondering about something else I wanted to do for a while: having Journelly’s latitude and longitude fed automatically to OSM in Emacs, so I can view the location on a map.

Journelly captures locations and weather information for each note and stores those under properties, like so:

PROPERTIES: :LATITUDE: ##.########## :LONGITUDE: ##.########## :WEATHER_TEMPERATURE: 62.1°F :WEATHER_CONDITION: Cloudy :WEATHER_SYMBOL: cloud :END:

The OSM function that calls for those is osm-goto.

So what we need is a simple function to feed the properties values directly:

(defun jtr-goto-from-properties () (interactive) (let ((lat (org-entry-get (point) "LATITUDE")) (lon (org-entry-get (point) "LONGITUDE"))) (if (and lat lon) (osm-goto (string-to-number lat) (string-to-number lon) osm-default-zoom) (message "No LATITUDE/LONGITUDE properties found on this entry"))))

This is an interactive function that I use when I’m standing on the header in Journelly I want to see on a map. It’s quick and works well. Now I can use my Journelly entries, which are already in org-mode, as a base for a post with a map tile inside Emacs. OSM doesn’t have a native function to export an image, but since I usually want to annotate the image anyway before I make a post out of it, a regular screen-capture app is a good solution, at least for now.

In other geek news, I found out about What the Cable a week ago through Mastodon:

GitHub - darrylmorley/whatcable: macOS menu bar app that tells you, in plain English, what each USB-C cable plugged into your Mac can actually do github.com

GitHub · Change is constant. GitHub keeps you ahead. https://github.com/darrylmorley/whatcable

It’s a nice little app for macOS that gives you a bunch of information regarding what your USB cables are connected to, what kind of cable, and if they’re optimized. nicely made.

I wrote about Harper before, but I wanted to expand now that I have it working on Kubuntu with a couple of more options.

Harper is good in two scenarios for me: first, when I want something quick and I don’t feel like starting a browser with Grammarly in it, and second, when I write a personal email and the idea of my words going to some AI grammar bot somewhere makes my skin crawl. Otherwise, for my blog (which is public anyway) and work email (I don’t care about those) Grammarly is definitely better.

The issue with Linux is that the makers of Harper geared it toward macOS (Homebrew) and Arch Linux, among other things. It was made for programmers by programmers, and these guys don’t bother with Ubuntu-like distros. Fair, but up until recently it meant I had to jump through hoops.

The quick and easy route in Ubuntu distros (which is what Kubuntu is) is to use snap. I know, I know. I didn’t want to either, but since the Harper makers don’t bother with flatpak, the other option was to install a Rust environment, which is a big overkill just for an app inside Emacs I use like once a month or so or less. I don’t like snap and I don’t use it, but I made an exception here.

Now, that snap “shortcut” is by a guy who works with Ubuntu (I think) and maintains his own package for it, so it’s on the edge channel (not stable) and seems to be a lot behind (version .49 to be exact, and Harper is currently 2.2.1!) so if Harper is your choice of grammar check, and you use it daily, I’d suggest against what I’m doing below. I’d install as needed in that case, Rust and all.

With that disclaimer, let’s move on: sudo snap install harper --edge.

Now in Emacs, in Linux, we want to tell it where harper is:

(when (eq system-type 'gnu/linux) (add-to-list 'exec-path "/snap/harper/current/bin"))

When tells it to add the snap path only when running Linux, since I have the same config for both macOS and Linux.

That's all... After that, I've added some of Harper's flags, or linters. Here's the whole code as I have it in my emacs org settings: (with-eval-after-load 'eglot (add-to-list 'eglot-server-programs '(org-mode . ("harper-ls" "--stdio")))) (setq-default eglot-workspace-configuration '(:harper-ls (:dialect "American" :linters (:LongSentences :json-false :AvoidCurses :json-false)))) ;; Besides choosing American as the language, I also want to ignore long sentences (the main issue is that it hides other errors nested in those) and I also want harper not to tell me when it thinks something is offensive. The full list of these options is in https://writewithharper.com/docs/rules. It needs to be nested inside the :linters option. (when (eq system-type 'gnu/linux) (add-to-list 'exec-path "/snap/harper/current/bin")) ;; on a mac, Harper is installed via Homebrew - on Kubuntu, the best option is snap - the harper team does not do a package (flatpak) unfortunately, and I don't want to install Rust just for harper. So.. meh. I did sudo 'snap install harper --edge' for this.

Trending Blogs

Recently Viewed Blogs

The Art of Not Asking Why