I have a post in one section of my site, say notes and miscellanea
That is my canonical post.
But I also need it to show up in the work section of my site, possibly with a another name.
A Hugo alias in the yaml front matter does not work, since both sections exist and Hugo gets confused.

Solution

1. Make sure layouts/redirect/single.html exists within your Hugo folder structure and it contains just this line …

{{- template "alias.html" (dict "Permalink" .Params.target) -}}

2. Create your orginal post as usual
3. Then go create a post in your other¹ section with the following metadata:

---
title: <Your title here>
#date autogenerated by my archetype. if missing, add date
date: 2025-11-06T08:30:54+05:30 
type: redirect
target: <url-to-redirect-to>
summary: <Optional text, that will appear in the sectiones listing page>
---

4. The date field is critical. The post might not appear where you expect it, if it’s absent.
   If not date, then you might need to use weight. One of the two is needed.
5. In my case, the metadata looked like this:

---
title: "The Plan for My New Hetzner VM"
date: 2025-11-06T08:30:54+05:30
type: redirect
target: /nm/2025-11-06/
summary: Alone and helpless, like you've lost your fight 
---

6. Done! You can see an entry for “The Plan for My New Hetzner VM” in my work section as you see in the pic below. Clicking that will redirect to the actual post in my Notes and Miscellanea section!

7. Tada 🎉

I’ve never updated my Emacs packages until recently, because Emacs is where all my writing happens, and so I’m justifiably paranoid.
But then some packages stopped working, due to various circumstances¹ and an update solved it.

So I’ve decided to update my packages once a quarter, so that I don’t lose days yak shaving when something goes wrong and I handle breakage on my terms and not the machine’s.

As far as package management goes, I want to keep things simple.
In fact, I still haven’t graduated to use-package or straight.el because my package needs are few and conservative². And so, while there are automatic update options out there, I’ll just stick to updating them manually, every quarter.

Ergo, this is the checklist I’ll use next time onwards …

1. Stop emacs user service, systemctl --user stop emacs
2. Backup emacs folder in ~/.config
3. Start emacs manually (not the service).
4. M-x package-refresh-contents
5. M-x package-upgrade-all
6. Problems? Quit emacs. Revert backup folder.
7. In the end, start emacs user sevice, systemctl --user start emacs

There’s an Org mode task, scheduled quarterly, so that I won’t forget.

For e.g. https://janusworx.com/blog/using-hugo-variables-to-help-with-mailto-links-in-hugo/ …
was now at, https://janusworx.com/work/using-hugo-variables-to-help-with-mailto-links-in-hugo/

After searching a bit, I found that Hugo supported aliases. For me, it would redirect the original /blog path url to its new location
Ass I had to do, was add an aliases: ["/blog/old-post-slug"] line to each post’s metadata.¹
Line 4 in the snippet below shows, what I added to fix the post above.

1
2
3
4
5
6
7
8

---
title: "Using Hugo Variables to Help With Mailto Links in Hugo"
date: 2024-05-30T18:17:35+05:30
aliases: ["/blog/using-hugo-variables-to-help-with-mailto-links-in-hugo"]
categories: ["work"]
tags: [100WordHabit, Dgplug, Hugo]
summary: Shortcodes! Hugo Variables in Shortcodes!
---

I did not want to do this by hand for 800+ posts.
One stroke of luck for me, was that I let Hugo use its default behaviour of generating url slugs from the file names. So even if the category slugs had changed (from /blog to /work or from /blog to /personal), the url slugs would stay the same. Which meant, I could whip up a script to run through all my markdown posts and add the alias line.
So I did.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12

from pathlib import Path

INPUT_FOLDER = Path("old-posts-folder")
OUTPUT_FOLDER = Path("modified-posts-folder")

for each_file in INPUT_FOLDER.iterdir():
    with open(each_file, 'r') as file_to_read:
        alias_derived_from_file = each_file.stem
        contents_as_list = file_to_read.readlines()
        contents_as_list.insert(3, f"aliases: [\"/blog/{alias_derived_from_file}\"]\n")
        with open(Path(OUTPUT_FOLDER, each_file.name), 'w+') as file_to_write:
            file_to_write.writelines(contents_as_list)

It takes all the posts from my old folder, inserts the alias line and puts them into a new folder.² In essence, take each file, figure out the url slug from the file name, read in the contents as a list, insert my alias at position 3 (fourth actually. zero based indexing) of the list (below the title and date) and then write it all out to a new file.

I ran it, published the site and then went to check on the old urls with bated breath.
Hurrah, it all worked :)

1. Launch Terminal
2. Launch Tmux
3. Split it into three windows err … panes.
4. Go to the top left pane and launch Hugo server
5. Switch to the right pane and then launch Emacs with whatever new post I want to write today.

So of course, a Rube Goldberg-esque, tiny bash-pipey monster took form. It now resides, chained to an alias, hssx¹ in my .bash_aliases file

1
2
3

alias hssx='cd /path/to/my/hugo/folder && \
xdotool windowsize $(xdotool getactivewindow) 100% 100% && \
tmux new-session \; split-window -h \; select-pane -l \; split-window -v \; select-pane -U \; send-keys "hugo serve" C-m \; select-pane -R'

1. The first line switches to my hugo folder
2. The second calls xdotool and maximises the terminal window
3. And the last line is a series of instructions to the tmux command. I’ve split it below for readability.

tmux new-session \; #Tmux start a new session
split-window -h \; # Split the window into two vertical panes
select-pane -l \; # Switch to the left pane
split-window -v \; # Split that into two horizontal panes
select-pane -U \; # Select the upper pane
send-keys "hugo serve" C-m \; # Type in `hugo serve` followed by Enter
select-pane -R # Select the right pane

Like Dale Carnegie says,

Shut off the past! Let the dead past bury its dead. Shut out the yesterdays which have lighted fools the way to dusty death.

The load of tomorrow, added to that of yesterday, carried today, makes the strongest falter. Shut off the future as tightly as the past.

The future is today. There is no tomorrow. The day of man’s salvation is now.

Waste of energy, mental distress, nervous worries dog the steps of a man who is anxious about the future. Shut close, then the great fore and aft bulkheads, and prepare to cultivate the habit of life of day-tight compartments.

So I want then, my Org Mode to do show me, only the things, I have on my plate today.

And Org is nothing if not adaptable. org-agenda-span is the variable that controls what the Agends shows me. According to the docs,

This variable can be set to any number of days you want to see by default in the agenda, or to a span name, such a day, week, month or year.

So that’s what I did. This little snippet went into my init file

;; Set Agenda to show a day as the default timespan, instead of a week
(setq org-agenda-span 'day)

Et voilà!

Now that I use my local machine to draft all my posts, desktop Emacs makes everything easier … with one exception. It swallows up a lot of my Espanso expansions.

My current bugbear is that it will not render a = when expanding.
I know this is entirely something that I brought on myself, and neither Emacs or Espanso in isolation will have issues.

My workaround after rummaging through Espanso’s documentation was to set the backend to use xdotool, in my config. (~/.config/espanso/config/default.yml)

x11_use_xdotool_backend: true

One downside now is that it no longer jumps back to a predefined location in the snippets where I’ve defined such a thing. No matter. I use them rarely and I’ll figure something out later. (Alongside why desktop emacs won’t render markdown expansions sometimes as well)

For now, most of my expansions work. Hurrah!

I never had the time to deal with my Espanso¹ hijinks until today.
While it worked perfectly, when I installed it all those years ago, when I migrated over from the Mac, Espanso itself has changed and evolved over the years.
It took over my old configuration like a champ and mostly worked, with the exception of a few shortcuts; ones that I frequently used 😂

Emacs was one application of mine that never quite worked right with Espanso.
I’d frequently get timed out waiting for reply from selection owner whenever I tried expansions in there. Typing :joy to get 😂 would work in every other program, but no joy with Emacs, in additon to plenty of other expansions err … not expanding.

All my browser url expansions would not expand properly either, with mangled expansions most of the time.

So today I dove in to the docs, and realised two things.
1. My emacs needed a longer time out
2. Espanso now tries to identify the kind of text, and maybe those were causing my issues?

Emacs

I realised I needed a longer clipboard threshold, only for Emacs.
So I created an App specific configuration, just for Emacs to use and gave it said option. Here’s what the contents of my espanso/config/emacs.yml look like

filter_class: 'Emacs'
clipboard_threshold: 10000

Rich Text Expansions

That helped with a lot of expansions in Emacs, but not with my joy expansion.
And not with stuff that were links and oh … links! and html! and markdown! Could those be the culprits?²
The docs mention that Espanso now has rich text support
What that means, is that the trigger nows supports two new keywords html and markdown in addition ye ole replace So I changed most of my affected shortcuts to either of those two keywords; markdown for most everything and html for linky stuff. Here’s what my beloved joy looks like now …

matches:
  - trigger: ":joy"
    markdown: 😂

And those two things did it! Every shortcut expands everwhere! What joy 😂

I want to slowly increase my French vocabulary, so I got this comprehensive frequency word deck from Shared Decks section of the Anki website.

I keep forgetting to look it up during the day, after my morning session and the only way I can get those stubborn words¹ to stick in my mind, is if I keep doing the deck 4–5 times a day.

I know! With all my newfound devops/python skills, could I figure out a way to remind myself to do it? Turns out I can! :)

I already use Pushover, to get notified of darn near anything.²
And I read enough of Anki’s add-on documentation to know I could whip something up.
So a bit of searching on the net, a bit of jiggery pokery with Claude, and some spelunking through Anki’s source code and forums later, I present to you … anki-push-u!³

As long as your Anki’s running, this little add-on will find cards due, at the interval you tell it to, and then notify you wherever you have Pushover running!
No more forgetting due cards!

Find the add-on and instructions over at Github.

Update, 2024-08-20:
Added a missing step, to install the qBittorrent files to their relevant places
Also, it had to happen. No releases for close to three months and then just as I learn to compile, a new version has to release the very next day 😂
It gave me a chance to test my notes. It hardly took me a few minutes to get the new release in place. So all is good!

I wanted to always have the lastest stable version of qBittorent on my Pi.
So today, I sat down and followed the project’s excellent instructions and got it running.
I did however stray off the beaten path a teensy bit¹.
I didn’t want a gui and I wanted to be able to control it with systemd.

So here’s my checklist for next release. I’m currently running Bookworm on my Pi 5

1. In case you run into problems, compile instructions are here: https://github.com/qbittorrent/qBittorrent/wiki/Compilation-Debian,-Ubuntu,-and-derivatives

2. Do all off this after switching to root. (or sudo)

3. Install build tools
   sudo apt install build-essential cmake git ninja-build pkg-config libboost-dev libssl-dev zlib1g-dev libgl1-mesa-dev

4. Install Qt dependencies (qtbase5-private-dev is not part of the instructions, but compile will fail if you don’t include it)²
   sudo apt install --no-install-recommends qtbase5-dev qtbase5-private-dev qttools5-dev libqt5svg5-dev

5. Qt requires Python at runtime, so: sudo apt install python3

6. Install the libtorrent library. I compiled from source. Check page for fresh instructions, if stuff here doesn’t work

   git clone --recurse-submodules https://github.com/arvidn/libtorrent.git
   cd libtorrent
   git checkout RC_2_0 # or a 2.0.x tag
   cmake -B build -DCMAKE_BUILD_TYPE=RelWithDebInfo -DCMAKE_INSTALL_PREFIX=/usr/local
   cmake --build build -j4 # the -j4 is to do on four cores at once. my addition.
   sudo cmake --install build
   

   If you ever want to uninstall libtorrent, then don’t delete this main folder. The build process would have created a install_manifest.txt in the build folder. So you can then uninstall with: sudo xargs rm < install_manifest.txt

7. Get qb source: https://www.qbittorrent.org/download.

8. Unzip: tar xzvf qbittorrent-4.6.5.tar.gz and then cd qbittorrent-4.6.5/

9. Create Build configuration (disabled GUI and enabled systemd. Those aren’t part of the defaults)
   cmake -G "Ninja" -B build -DCMAKE_BUILD_TYPE=RelWithDebInfo -DCMAKE_INSTALL_PREFIX=/usr/local -DGUI=OFF -DSYSTEMD=ON

10. Build cmake --build build

11. Install all the files to their relevant locations cmake --install build

12. Install service to run for a specific user.

    systemctl daemon-reload
    systemctl enable qbittorrent-nox@<some-user>.service
    systemctl start qbittorrent-nox@<some-user>.service
    

13. We’re done.

I love to write.
Correction, I need to write.
Writing helps me think, to be deliberate about things, to slow down.

At the same time, writing is also a slow, plodding process for me. Every post seems to take forever. So many years of doing it and it does not get any easier.
I agonise over what I want to say, the picture or quotes I want to include, the overall arc of my thoughts, the structure, everything has to be just so.¹
I’ve lowered the bar, to build consistency, with my #100WordHabit and now I have a new problem. First, I could not get started, and now I cannot stop once I get started.
I just have to finish my thoughts, my post, and that sometimes takes hours.
Which brings me back to me not wanting to start, for the fear that nothing else (that are in fact, more important) will get done.

This in turn, brings me to the brainwave I had a couple of days ago, in the bath.²
The pressure comes from wanting to publish what I write, the day I write.
What if, I didn’t?

Nothing has to change. With the exception of a hard stop.
One hour.
One hour to write and publish.
Or one hour to write and then put my pencil down.
And then I can come back to it the next day.
Yesterday’s post was the result of this approach.
I wrote it over two days, in two one hour sessions.
Very like Abby’s rain lillies up up above. They are prepping through the year, to bloom at this time, every time. Just because I cannot see it, does not mean nothing’s happening

One downside, that I potentially see, is that I will use this as an excuse to slack.
I hope I don’t.
But then on the other hand, without a pressure control valve like this, I can very easily see myself burning out again.

So let’s see how this goes.
I’m betting on me being a rain lily.

I finally went ahead and read about what Emacs Diary S-expressions were.

Going back, a bit.
I make heavy use of Org Mode’s deadline and scheduling features.
I put schedules and deadlines in lots of tasks across my ten or so org files, so that I can look at them every week, to decide what I’ll be doing this week as well as to review at the beginning of every month.
To wit, there are plenty of tasks that repeat on a weekly and monthly cadence, and Org’s repeaters.

But¹, there’s one kind of event I have, that Org doesn’t support quite well.
If I want something on, let’s say the second Friday of every month, there’s no easy way to describe / do that with Org.
Org defers to Emacs’ native Diary Mode to do something like this.. Ok, not defers, but it supports Diary expression entries with a few caveats.
Now those were scary when I first saw them, which is why I kicked the can down the road, and decided to just live with manually making entries for the second Friday of every month.
A year of doing this, and I’ve finally had enough. So I finally went ahead and read about what Emacs Diary S-expressions were.

Only to realise, that they were not as scary as I feared, after close to a year of messing around with my Emacs config files. They look² just like Emacs functions with arguments.

I scribbled one out the way, Org needs them written, by referring to the Emacs help.

<%%(diary-float t 5 2)>

I marked the sections out in the scrawly figure above.

• diary-float seems to be the function that does all the magic and then I have three arguments
• The first one, refers to the month by number. 8 will mean August. t will make things repeat every month.
• The second argument is day of week. 0 being Sunday, and 6, Saturday.
• The third one refers to which week of the month. 2 in my case, means the second week of the month.

So …

* My super important task
  DEADLINE:  <%%(diary-float t 3 2)>

shows up in my agenda as a task for the second Wednesday of every month!

There are a couple of caveats to this approach.

• This is not as task, even though I write it as one. It behaves more like a calendar entry. It more or less, just shows up in the agenda and I have to look at it.
• I can’t mark it as done and expect it to repeat, like a normal Org repeater. If I do that, the task is, uh done and then that’s that.

Neither of these things bother me, because of the idiosyncratic way I drive my day.
I normally review my daily agenda, and create a day plan for myself.
Which means that I copy the repeater task into my daily agenda as a single task and mark the repeater as done. So with the weird way I work, writing it out as a diary sexp, actually saves me a step ❤️
And my cranky brain can now heave a sigh of relief.³

After which I go to the file which needs to contain my link, and select my text that I’ll want converted into a link (“inserting a link” in org speak. In my case, the top file in the screenshot below).
I select the Misc Brain Dump part of the “Look at the Misc Brain Dump” task and hit
C-c C-l
It then switches me to a buffer that shows the links that Org currently has stored, with the minibuffer prompting me to choose one

I use the arrow keys to choose the one I want, hit return and it creates the link for me.

Here, I’ve switched away from my brain dump. (I’ve closed the file in fact)

I click the link in my tasks file and …

It opens right up, at the exact location I want. Woohoo!

And just like everything else with Org, it’s plain, simple, elegant, text functionality.
I examined the link and it was a borg standard Org link, with the link having the header tacked on the end, like so …
file:brain-dump.org::*Misc Brain Dump¹

I want to someday, automate all my domain certificate renewals.
Namecheap doesn’t support it, ergo my move to Porkbun.
And certbot doesn’t support Porkbun, so I’ve moved to Lego.

I still need to someday figure out how to do this automatically, but for now, since all is in place, this is the new manual process.

1. You’ll need to generate a Porkbun API key along with its associated secret, and store it somewhere safe and secure. (This is a one time thing, the Generating API keys section here describes how)
2. Another one time step is to get Lego installed and have it generate your certs, and then have your applications, point to where they are.
3. Command to renew: PORKBUN_SECRET_API_KEY="your-api-*secret*" PORKBUN_API_KEY="your-porkbun-api-key" ./lego --domains your-domain.tld --domains *.domain.tld --email your-email@domain --dns porkbun renew
4. And the renewal’s done!
5. Restart yer engines … err your applications¹ et voilà!
6. If you have multiple machines, figure out a way to securely transfer the certs there too.

I have a folder (called syncbox) full of files, that Org Mode looks at to generate the agenda for me.
Now these files are also synced via Syncthing over to my phone, where another program called Beorg looks at them too and shows them as entries in my calendar, so that I don’t miss the really important things I need to do.

Functionally, all is well with this setup, except for one teensy thing.
Every once in a while, when certain files are in use on both, my desktop and my phone, Syncthing will create conflicting copies of said files in use.
This happens most often, with my tasks.org file.
Once it conflicts, my syncbox gets full with files like,
tasks.sync-conflict-20240408-041556-2TQWENY.org and
tasks.sync-conflict-20240516-101335-2TQWENY.org and
tasks.sync-conflict-20240520-053451-2TQWENY.org

What happens then, is that Org Agenda happily reads those files too, and I end up with lots of duplicated tasks.¹

As usual, tackling it by writing lisp, was an idea fraught with terror for me. I tried quite a bit and failed.
My next plan of attack was that Org Agenda had a regex², that it would process. My regex is better than my lisp, but even that was to no avail.
Emacs regex does not support lookaheads / lookbehinds and I did not want to have some complicated pattern, that even I would not recognise, a few months down the line.
In the meanwhile, Ihor told me to stop faffing about and just write a function³

So I asked Claude to go do the work for me and I ended up with …

(defun mjb/set-org-agenda-files-from-syncbox-directory ()
  "Set `org-agenda-files' to all Org files in the 'syncbox' directory, excluding files with '.sync-conflict' in the name."
  (let ((syncbox-dir (expand-file-name "syncbox" my-org-directory)))
    (when (file-directory-p syncbox-dir)
      (setq org-agenda-files
            (mapcar
             (lambda (file) (expand-file-name file syncbox-dir))
             (seq-filter
              (lambda (file) (not (string-match-p "\\.sync-conflict" file)))
              (directory-files syncbox-dir t "\.org$")))))))

(mjb/set-org-agenda-files-from-syncbox-directory)

which I dutifully dumped into my init file.
And that did the trick!

What did strike me, is that now that I am using this approach, my chicken and egg struggle is finally at an end. I don’t have to struggle to learn lisp to hack emacs which needs me to learn lisp.
I realise it takes only a short time of dedicated effort. But right now, it is time I don’t have, and don’t know when I’ll get.
So the new fangled LLM AIs can help me sort my Emacs pain points and I can learn Lisp at leisure.
And I’ll use, friend on the fediverse, Vincent Dardel’s excellent course to do just that.

Sheena: Queen of the Jungle, Vol. 2
Dynamite Entertainment
CBZ
185.4 MB
EPUB
105.4 MB
PDF
439.9 MB
Queen Sonja - Issue. #2
Dynamite Entertainment
CBZ
17.7 MB
EPUB
16.6 MB
PDF
137.2 MB
Vampirella: Throne of Skulls Vol 03
Dynamite Entertainment
CBZ
229.3 MB
EPUB
135.9 MB
PDF
438.6 MB
Red Sonja and the Ballad of the Red Goddess
Dynamite Entertainment
CBZ
132.9 MB
EPUB
50.2 MB
PDF
376.8 MB
Vampirella Strikes Vol 01
Dynamite Entertainment
CBZ
106.4 MB
EPUB
118.7 MB
PDF
351.7 MB
… really long list elided

And I go, Ugh! It’s Saturday! I’m too lazy to go selecting and deleting that cruft between my titles. Isn’t there something in Emacs that can help?
Spoiler Alert: Of course there is! Meet flush-lines

M-x flush-lines asks me for a regex pattern.
I gave it \ MB¹ et voilà! Every line that had a space followed by MB disappeared!

Sheena: Queen of the Jungle, Vol. 2
Dynamite Entertainment
CBZ
EPUB
PDF
Queen Sonja - Issue. #2
Dynamite Entertainment
CBZ
EPUB
PDF
Vampirella: Throne of Skulls Vol 03
Dynamite Entertainment
CBZ
EPUB
PDF
Red Sonja and the Ballad of the Red Goddess
Dynamite Entertainment
CBZ
EPUB
PDF
Vampirella Strikes Vol 01
Dynamite Entertainment
CBZ
EPUB
PDF
… long list elided

And so a few, rapid commands to excise lines with Dynamite Entertainment, CBZ, EPUB and PDF later …

Sheena: Queen of the Jungle, Vol. 2
Queen Sonja - Issue. #2
Vampirella: Throne of Skulls Vol 03
Red Sonja and the Ballad of the Red Goddess
Vampirella Strikes Vol 01

… still a long list, but exactly what I wanted :)

I love you, Emacs!

As I struggle to write, sometimes I wonder why I even do it.
And when I don’t, I then get this strange uncontrollable urge to write something, anything.
Damned if you do, damned if you don’t.

And now in addition to that crazy struggle, I’ve caught myself getting upset of late. All because nobody “shares” the posts I put out.
And I worry if anyone even reads my stuff?
And do they like it?
And if they do, why would they not share it?
Not good enough?
Should I be doing more?
Write better?

All of which came to a head today.
And after a bit of laughing at my silliness, I decided to write down what I tell myself every year.

Because I do write.
Whether anyone reads it. Or not.
Which is why I wonder why I get trapped in this weird neurotic spider web every year.
I’m done comparing, in the words of Lamott, my insides to their outsides.

So a few thoughts to myself on why I do it. A non exhaustive list …

1. I love to read.
2. I love to write.
3. I write primarily for myself. First and foremost.
4. I write because I cannot help it.
5. I write because I cannot not write.
6. I write to document things and thingamajigs and doodads and small things and big things. Because I forget. And writing helps me remember. And then it helps other people learn. And then they teach me new things. And I learn more. And I write more. And the whole virtuous cycle repeats.
7. I write because everything in me, has got to come out on the page, so that there is space within me, to then function.
8. I write because then everything slows down to the pace of my fingers rather than the crazy light speed dizziness within my brain.
9. I write because I can then think.
10. I write because it sharpens my focus.
11. I write because it makes it easy to forget.
12. I write because I read what I write. Even if no one else does. Even if no one else wants to. Even if no one else can. And I love my past selves, who’ve scribbled their thoughts and feelings to protect me, shield me, guide me and give me comfort, solace and love.
13. I write because then, my heart then aches a little bit less.
14. I write because it’s brought me back from the edge of the void. Twice.
15. I write because it helps me be resilient. Persistent. To quote Taleb, Antifragile. And to quote Angelou, “You may trod me in the very dirt / But still, like dust, I rise.”
16. I write because I have scars. And writing lessens the hurt. To quote Bareilles, it helps me hold on, till there’s something else to hold on to.
17. I write because …
18. I write because today I realise, that’s who I am.
    I am a writer, because that’s what I do. I write.

Lisp is on the list of things to learn someday.¹
And in the meanwhile, I need to be able to hack Emacs to do the stuff I want.
But also, there are a ton of other things, that are currently much, much higher on my list of priorities.²
And while the community is extremely kind and generous, the expectation is that you need to put in the time and the work.
Which I currently don’t have and can’t do.

And so, it finally happened.
I signed up for a Claude account and told it to write me some lisp.

I am heavily Org Mode task driven, these days and I have a manual workflow that I need a teensy bit of help with.

1. I start work on a task
2. I mark it as WORKING (and clock in manually)
3. I want to switch to another task
4. I mark my original task as WAITING (and clock out manually)
5. I then mark my new task as WORKING (and clock in manually)
6. After a while I switch to the original task, and clock in manually (after marking the current task as WAITING and clocking out there manually)

What bugs me, is the clocking in and out, because …

1. I tend to forget it, messing up all my clock times and …
2. because it is now becoming painful on my middle aged fingers³,

So, like I said, I prompted Claude to give me code that would clock me in automatically, whenever I changed state to WORKING and clock me out when I switched to WAITING⁴
It spat out some code.
I tried to reasonably guess and verify what it did.
It looked ok.
I tried it.
It worked!
And that’s that.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17

(defun mjb/org-clock-in-when-task-state-changes-to-working ()
  "Clock in an Org mode task when the status is changed to 'WORKING'."
  (when (and (eq major-mode 'org-mode)
             (member (downcase (or (org-get-todo-state) "")) '("working")))
    (org-clock-in)))


(defun mjb/org-clock-out-when-task-state-changes-to-waiting ()
  "Clock out an Org mode task when the status is changed to 'WAITING'."
  (when (and (eq major-mode 'org-mode)
             (member (downcase (or (org-get-todo-state) "")) '("waiting")))
    (org-clock-out)))

;; These lines make the functions above, hook to Org’s state change mechanism.
;; So that everytime a task’s state changes, these functions are called
(add-hook 'org-after-todo-state-change-hook #'mjb/org-clock-in-when-task-state-changes-to-working)
(add-hook 'org-after-todo-state-change-hook #'mjb/org-clock-out-when-task-state-changes-to-waiting)

Amanda says,

“Make bad things, instead of making nothing”

Something similar that’s been drummed into my head innumerable times over the years, by many mentors:

“If you cannot do it well, do it badly.” :)

If you love doing something, if you want to do something … do it.
Does not matter how poorly or badly its done.
You’re going to love it nonetheless.

I’m almost 50, and here is the best thing I have learned so far:
every strange thing you’ve ever been into,
every failed hobby or forgotten instrument,
everything you have ever learned will come back to you,
will serve you when you need it.

No love, however brief, is wasted.

(I’d add every act of kindness,random or otherwise, to that list.)

I recently read Scott H. Young’s article on How to Learn Vocabulary in Another Language and realised it would be a handy guide for me, as I go about increasing my French & Urdu vocabulary.

This is what I took away:

• The first 2000 words will cover 80-90% of a language’s vocabulary. 3000 will cover 95%. That should be your goal.
• Use a frequency dictionary, then according to frequency do 1000/2000/3000 words
• 10000 will lead to almost being fluent. Hard to do that quickly with French, sitting in India. But, aspirational goals :)
• Do:
  • Flash cards, knowing that this will not teach you the nuance and flavour and texture of French
  • Graded reading, Like the article states

  To successfully guess the meaning of a new word from its context, we need to understand between 95% and 98% of the surrounding words. This is difficult to achieve with native-level materials. Graded readers, which deliberately limit vocabulary, can be beneficial here.

  • If you’re doing the slow and steady way, because you lack the time, then growth will be slow. Assume 400 new word families a year
  • Use the four pillars of learning. Paraphrasing & Quoting the article:
    1. Meaning-focused input.
       Input provides the raw data for learning new words and enriches the contextual associations of words studied deliberately elsewhere. This includes conversations, books, films, television and other media that you attend to primarily for their meaning.
    2. Meaning-focused output. (Note: I am focussed on reading and writing)
       Speaking and writing are more difficult than simply understanding particular words from input, as using words correctly requires a more precise knowledge of each word and its meaning.
    3. Language-focused learning.
       This is the deliberate act of memorizing words, studying flashcards, or receiving explanations about word meanings. Webb and Nation argue that this should account for ~25% of the time spent learning a language.
    4. Fluency development. (I don’t know how I’ll do this. Re read stuff? Watch TV again.)
       Finally, attention should be paid to activities that speed up the understanding and production of words already known. Familiar materials that enable quick reading or conversations on familiar topics may not be needed to build new vocabulary, but they reinforce what was learned previously.

Cal Newport recently did a deep dive on his podcast, on a minimalist note taking system for various areas of your life.
Video’s on Youtube, if you want to watch. It’s called A Productivity System To Remember Everything You Learn.

It matches, what I’ve organically been doing all these years.
And I was really happy to see it structured and rendered so well.
So I wrote and paraphrased and jotted it all down below.
Note: These are my mangling of Cal’s words, not what he actually said. The audio/video above should give you exact specifics.

Why?

Minimal Friction

• too much friction with other systems
• friction in work that matters is ok, is good, is perfect. McPhee & Caro! But for most other things, not needed.
  • Cal goes on to verbally list what McPhee does, but my notes are better.
  • In a nutshell, friction slows him (McPhee) down in a way that’s beneficial, that lets him do his best work
• Notetaking does not need such friction. When information is coming at you, your time to act + energy to think are limited
• Friction stops you from taking action. That information could then be lost
  • For e.g. a complicated note taking process when reading a book, might stop you from getting into the book or ever keep you from reading the book altogether.
• Your system should therefore reduce friction, so that you can capture as much information as possible quickly, efficiently, painlessly

Outsource your brain? No!

• When it comes to things that matter, stuff should live in your brain. So that stuff percolates into your value systems and mental models and you evolve!
• Your brain needs to be part of this curation process. Your brain needs to build hooks. To remember the big things. For accuracy and precision, details can always be looked up.

The System

For Books (the Corner Marking Method)

• If there’s somethng interesting on the page, mark the corner fo the page (dog ear with pen) and then …
  • Mark up the thing that interest you! Simple marks in the margins. Put a box around text. Checkmarks next to a line. Curly braces next to a paragraph to remember
  • Occasionally write a short note, to help you remember the context, or to tell yourself what it reminds you of, or is there any other place where you’d find this useful?
• Barely slows you down. Does not get in your way. Does not prevent you from reading
• Then when you go back to your book, and look at your dog marks and then look at stuff you marked on those pages? You will in a few minutes, reconstitute all the ideas from that book.
• Bonus: One “shortcoming”? You need to remember! Oh that book? That was the one that had the interesting ideas about such and such
  • That is not a problem. This is the one bit of friction that is useful. It allows you to use the gist of what you learned from the book in the schemas of knowledge that I am constructing and modifying and growing in my head. So that’s not really a bug, rather a feature. In English? It helps me build better Mental Models!
  • You become a better reader

Projects (Professional and Personal)

• Where do notes relevant to a project go?
• Store notes relevant to a project in the location where you will one day work the project.
  • I do this on a personal level too. Everything is where my mind says, I ought to look for them, not where they “ought to be.” Which is why my keys are next to my Minnie mouse plushie and my flash drives live in an old, soup bowl.
• So when it comes time to do the work, everything that you’ve gather is all there for you to use.
• Duplication is ok. (If there’s too much duplication, which happens rarely, I organically figure something out)
• Bonus: Whenever you add something new to the pile, you always encounter your old stuff and it keeps refreshing a mental picture of the project in your mind and helps create new mental grooves

Ideas about Life? Your values, your inspiration, you want to do something with and in your life

• Keep a fancy, awesome, cool, aspirational notebook!
• A fancy pen you like
• Basically whatever you used when you started your journey? You wrote notes, you made headlines, you had marginalia in your pages, you did everything by hand. Just use the classiest things you like and can afford, to give it heft and meaning.
• Your life has few ideas, important as they are. Easy to keep track of them in a notebook
• You want the form to matter.
• Digital notebooks? They’re cool too, if you like them.
• All you need is semi regular review and process
  • A good lazy way to do it is when the notebook fills up
  • Review and copy over the summaries of everything good and lasting and important
  • See what sticks

Now everything was fine and dandy.
There are folks who do leave well enough alone. Their blogs stay up and running for years on end, held together with nothing more than spit and baling wire.
Those folks aren’t me, sadly. Aditya keeps improving the theme and I really wanted the new shiny. So I went and got it.
And everything broke!
I got weird artifacts. RSS broke. Fonts broke. Footnotes broke.³

I’ve managed to put everything back together, over the past week, and I just got footnotes working again yesterday.
So a few notes to myself, on what to do, when I break it again.⁴
Footnotes today, and the rest later …

• The litlefoot.js repo has excellent documentation. That’s where you pulled everything from (for when you wonder, just how you figured all the CSS out)
• You download a static copy with a npm install --save littlefoot and then stick it into Hugo’s static folder
• Copy the extend_head.html from the theme’s layout/partials into your own. You should already have an extend_footer.html since you put in PaperMod’s hamburger stuff in there.
• Adapt the Usage from a CDN step to directly access it from /littlefoot/dist/… instead of UNPKG. Put the CSS link into your extend_head.html and the JS into extend_footer.html
• We want the footnotes to be numbers, rather than floating ellipses. So modify the default (blank) script call in extend_footer.html, to the stuff below. This is verbatim from the docs.

<script type="application/javascript">
    littlefoot.littlefoot(
        {
            buttonTemplate: `
            <button
            aria-label="Footnote <% number %>"
            class="littlefoot__button"
            id="<% reference %>"
            title="See Footnote <% number %>"
            />
            <% number %>
            </button>
            `
        }
    )
</script>

• We want to style it a teensy bit, so those styles go into extend_head.html. For some reason, it does not pick the footnote styling correctly when you put stuff into your normal assets/css/extended/ folder.

<!--Custom littlefoot styles-->
<style>
    .littlefoot {
		/*the --attributes come from the littlefoot.js docs*/
		/*Change button background to something that works in both modes*/
        --button-background-color: #666; 
		/*Change padding to make the button rounder*/
        --button-padding: 0 0.35rem;
		/*The footnote ought to be italicised*/
        --popover-font-style: italic;
		/*Make the popover work with my theme’s colour scheme*/
        --popover-background-color: var(--theme);
        --popover-text-color: var(--primary);
        /*the oldstyle numerals used everywhere else, don’t look good here
		Neither do they align properly in the box.
		So switch them to normal*/
		font-variant-numeric: normal;
		/*Italic numerals for the footnotes*/
        font-style: italic;
    }

    .littlefoot__button {
		/*very subjective. 
		try to get the numerals centred as much as I can.*/
        line-height: 1;
        vertical-align: middle!important;
    }
</style>

Et voilà! Footnotes galore!

With all the hullabaloo yesterday, on how you’d go about organising your day, and things to let go of, and marking your place, I forgot the overarching Note To Self.
Please remember. Never lose sight of what’s important.
Never forget why you’re doing all of this.

This started with a note to self on leaving breadcrumbs, and ended with a stream of consciousness journey on how I reached this conclusion. So what I’m trying to say, in a slightly rambling manner, is that this is a slightly incoherent, rambling post.

I’m running my days now to slow, daily, consistent tiny amounts of progress.
And I am seeing results. When there are so many demands on my time, prioritising them and then doing them slowly yet consistently is yielding fruit.
While I certainly enjoy the deep dive, full of focus, lost in work sessions a lot more, I have too much life happening to allocate that kind of time and attention to everything.
The first thing I did, when I kinda-sorta revamped the way I went about my days¹, was to drop as many hot potatoes as I could, and then d-d-d-ed all the small yet important things in my life.
And then came to the sobering realisation that the rest of life just had to be done and dealt with. Even things I don’t like.

A couple of pieces then had to fall into place, before I gained some semblance of sanity and progress.

1. I can’t do it all. Some things, important things even, will fall through the cracks. Somethings that I really want to do, will get pushed to someday/maybe and some will just never get done. I have to make my peace with it.
2. Knowing what I am giving up, the things that I am doing? They better be worth it.
3. Making the time to think this through, and having mental muscle to do the thinking² helped.
4. Everything does not have to be good and perfect. Good enough, is good enough.
5. Yes, stuff is hard and unpleasant, but then you have a blessed life. There are so many people with hard lives. Suck it up. And go do something about it. Deal with it. Leaving stuff unattended in the recesses of your mind, is a sure way to rust and ruin.³. At the same time, while you are hard and focussed on the work, be kind to yourself. Find that balance.
6. Alright, mentally set. Now how do I go about the doing?
7. By doing as much or as little as you can, depending on how you prioritise it.
8. By doing it consistently. Set a schedule. Hit it.
9. If you think you don’t want to do it, do 10m regardless.⁴ And then move on.⁵
10. If interrupted, deal with the interruption. Come back. And continue.
11. Dealing with interruptions does not mean giving in to it. It means figuring out just what it is, and figuring out what to do about it. Most stuff can be jotted down and scheduled for later. Make sure the significant folk in your life know what you are up to, so that they know when they can expect you to be present and when you are off doing your own thing.⁶
12. Doing stuff this way is slow. And that’s ok. If your life permitted focussed deep work, you’d be doing it. But that’s not how it is. So find islands of focus. And do the best you can.
13. Do the best you can, does not mean going all out. Slow, sure, steady. That is your drumbeat. Slow, sure, steady. Unflustered. Calm. Measured.

So what happens once you do this? And what hiccoughs did you run into? And what do you do about it?

1. Stuff gets done! It doesn’t look like this ought to work, but it does.
2. Leaving stuff undone, putting things down for the day, causes lots of mental turmoil.
3. Also whenever you go back to stuff, there’s a lot of going back before your minds grasps where it was at, and only then does it go forth.
4. And that brings me to the crux of why I began writing this dastardly note in the first place.
5. Log. Summarise. Review!
   1. Log how long you take for your tasks. Don’t judge yourself. Just dispassionately do it.
   2. Every time you think you’re done with something for the time being and are putting it down? Spend the time to write down what you did and what future you needs to be doing. Create bookmarks. Give directions. Do this no matter how tired you are. No matter how shitty the note.
   3. Review. This is the most important bit. Review at regular intervals and at various depths.⁷ And only then make your plans accordingly. Daily time logging & summarising leads to daily review leads to a better sense of what is done, how far you’ve come, what to change, where to swerve, what to do next, and where to go. Review, gosh darn it. Review!

It’s been working well the past three months. It’s reduced a lot of friction, and quite a bit of time. It makes it easy to get into the swing of things.
I’ll come back in six months and write about how it went.

The only reason we know that both, an emperor as well as a slave, both practiced the same philosophy that inspires people even now in this day and age, is because the former’s personal notes and the latter’s words compiled by a student, have come down to us through the ages.

The only reason that Leonardo Da Vinci has such a legacy, is that the few pieces of his work that were created and displayed, have somehow, miraculously lasted the ravages of the past half century. But so much of his work was unpublished, stowed away in private journals, never to be seen for centuries. How much more would we have known, or learnt if only? If only.

Decades of stunning art, never released. No one knew the genius of Vivian Maier, when she was alive.

Like David Kadavy writes,¹

Finishing what you start is overrated.
[…]
Contrast this with many other things Leonardo never finished. The observations that lie hidden in his notes would have advanced humanity by centuries in the fields of medicine, flight, geology, and more. These sketches and notes weren’t finished – but, most tragically, they weren’t published.
You don’t have to finish to have an impact. You do have to publish.

Honoré De Balzac had it right,

“For artists, the great problem to solve is how to get oneself noticed”

On which Austin Kleon expounds²

If you just focus on getting really good, Martin says, people will come to you. I happen to agree: You don’t really find an audience for your work; they find you. But it’s not enough to be good. In order to be found, you have to be findable. I think there’s an easy way of putting your work out there and making it discoverable while you’re focused on getting really good at what you do.

Imagine if your next boss didn’t have to read your résumé because he already reads your blog. Imagine being a student and getting your first gig based on a school project you posted online. Imagine losing your job but having a social network of people familiar with your work and ready to help you find a new one. Imagine turning a side project or a hobby into your profession because you had a following that could support you.

Or imagine something simpler and just as satisfying: spending the majority of your time, energy, and attention practicing a craft, learning a trade, or running a business, while also allowing for the possibility that your work might attract a group of people who share your interests.

All you have to do is, Show Your Work.

So the process then becomes …

1. Open the post on the server in Emacs (after logging in, via ssh)
2. Make an edit.
3. Build and publish.
4. Reload the page and re-read the post.
5. Find a typo. Fix it.
6. Repeat steps 3 & 4.
7. Keep the Emacs pane to keep editing and open another terminal pane, just to build and deploy.
8. Twenty three edits later … Be ok with what I have. (with a build and deploy, and reload and reread every couple of edits)
9. Do a final build and deploy

Having done this, for God knows how long now, I’m used to this workflow and decided to just lean into it.
But while I loved my edit, build, deploy, reload, reread, workflow, I felt like I was tempting the fates, everytime I did it. Besides I did not like messing with the live website like that.

So I wondered, if the way I did it locally, could be kinda sorta done remotely.
When I write posts locally, all I need to do is run hugo server in a terminal and it spins up a local server, with the all the posts, built and rendered in memory for speedy access.
Could I have that? But remote? Could I run hugo server on my server and view the results in my browser from anywhere?
About an hour of exploration later, I did have it up and running!

Here’s what I did, after delving into Hugo’s server command options

1. Created a subdomain (drpr.janusworx.com) and pointed it to my server. (this is where I’ll see my temporary, work-in-progress, results whenever I run hugo server on my server)
2. Used Nginx as a proxy to point to it. (Also serves as my SSL termination point, freeing Hugo from the burdens of HTTPS/SSL drama)
3. Spun up Hugo with hugo serve -D -b https://drpr.janusworx.com --appendPort=false --liveReloadPort=443 where …
   a. -D builds posts marked with drafts
   b. -b tells Hugo the hostname and path to the root of the blog (in my case, https://drpr.janusworx.com)
   c. --appendPort is set to false, which stops appending the port number to the url (happens by default, http://localhost:1313)
   d. --liveReloadPort is set to my SSL/TLS port, so that I can watch changes live with every edit.

And everything worked!
Except for the live reloading. That bit just refused to work. I could make changes and then go manually reload the web page and it would update. Grr! So near, yet so far!
Some more exploration led to Pothi Kalimuthu’s extremely helpful comment on Github …

Where I realised that Hugo’s LiveReload script (that reflects changes) needs a location in my Nginx config.
So a bit of copying and pasting and adapting and tada! 🎉

This blog post was written entirely on the server, with me watching the changes live as they happened …

And once this was done, I could just do a single, final, build and deploy to publish.

What I had to do, was to switch Manjaro’s software branch to unstable (as opposed to the default stable. Read more here.)
The only reason I use Manjaro, instead of Arch⁴, is that the Arch kernel does not boot on this Pi and I want the latest Arch stuff, which is what the unstable Manjaro software branch offers.

So a …

sudo pacman-mirrors --api --set-branch unstable

switches me to the unstable branch and …

sudo pacman-mirrors --fasttrack 5 && sudo pacman -Syu 

will resync all the software with what’s available on that branch.

Tada, I have syncthing v1.27.2 "Gold Grasshopper" on all my machines :)