Today I learned about Charmstone, an alternate task-switcher for macOS. I’m still using the free version, so I’m missing a few features, but even the free version is interesting. You press your two trigger keys (ctrl opt by default) and move the mouse/cursor a little and it pops open a floating app switcher with four options next to the cursor. Besides plain old apps, you can include folders or scripts.

It looks like this:

I didn’t get “why do you need to do the little bit of mouse motion” for a couple of seconds, then realized “oh, it’d be bad if you were just trying to do a keyboard operation with those two keys.” Then after a few repetitions I realized that you also begin to remember where in the floating selector your apps are, so the actual gesture is “press your two trigger keys as you begin to move in the direction of the target app.” It, uh, sort of suggests this in the part of the marketing copy that reads “Use spatial memory to put your desired app in focus more quickly.”

Anyhow, I am going to keep using it for a while longer. I haven’t been using it for long and it isn’t baked in yet. It’s interesting just because it makes me rethink an operation I don’t think twice about anymore (task switching) in that way where you begin to realize you’ve perhaps internalized needless motion, or at least motion that could be more economical.

The same developer is one of those kinda cool cottage software houses who’s got a Thing They’re Focused On, and their thing is UI. In addition to Charmstone, they’ve got a window management tool, an interesting keyboard-oriented UI search tool that takes cues from things like Vimari without being devoted to vim everywhere, and a bunch of other stuff similar to the little enhancements and “haxies” that have long existed in the Mac ecosystem, but with a really pleasant visual design and consistent sensibility that isn’t always there.

I haven’t tried his window manager, but I’ve been using Sizeup for years and I think they are pretty similar. His is open source and free.

Small update

I’ve been using Charmstone all day and really appreciate its Launch active apps setting, which gives you a new window if you switch to an app with none (e.g. you cmd w’d the last active window and get nothing when you use the normal cmd tab switcher to get to it.) Mailmate is one of those apps that just sits there and does nothing, so checking mail every so often has been a great reinforcer as I remember I’d have saved a step if I’d just used Charmstone.

In the neighborhood

The hideous, sewage-leaking, decrepit duplexes described in this article are in my neighborhood. I live in Lents, and these houses are right off the Springwater near the Foster Floodplain Natural Area, on the other side of 205 from us.

Al and I walk by them every several days on our way to the Floodplain, and they’ve inspired a lot of speculation. We remember when they were just this odd little colony that appeared to have folks renting them. We remember when they started being boarded up. We remember when the gate collapsed, and was then replaced with something makeshift, and then when it was obvious squatters were living there.

Al has worked in homeless policy and services for several years now, so she’s got a working library of landlord patterns that found an easy match in this case: Absentee landlord who doesn’t particularly care; tenants holding on despite backed up sewage because there’s not a next rung down, just an abyss.

The area around those duplexes was home to a small colony of RVs and vans, but about the time the police cleared out the squatters in the duplexes, there was also a general sweep of that area. Several of the RV owners found their way to our neighborhood. That’s part of the pattern every year once the weather starts warming up. We’ll have three or four waves before it gets cold again at the end of the year. Each wave is four or five RVs parked along the block and around the corner. In between the big waves individuals we’ve come to recognize over the years leave to avoid the new crowd, then orbit back through when it quiets down again. Each time we make an effort to get to know them. Al’s better at it than I am, and is more helpful anyhow: She helps them understand where to start in the social services labyrinth, saving them a few steps. But she also takes them food and water, and sometimes just listens to what they have to say.

I have a little distance from the problem. It’s not my work or career. I’ve thought (and felt) through how I feel when a new wave comes through. It took a few years of experiencing a particular cognitive dissonance about the matter to finally put my finger on what I was feeling and realize this is one topic where I’m not any of “liberal,” “progressive,” “moderate,” or “conservative.” Few combinations of political program and cultural leaning I’ve come across work for me here, which makes it easy to listen to pretty much nobody on the matter.

Al, on the other hand, doesn’t ever get distance. It is her job. And when she comes home at night, it’s parked across the street. It wakes her up at night with 3 a.m. screaming matches. When families roll through, it’s a thing to think about because she’s a mandatory reporter. She has friends in county and metro policy circles, so happy hour with them is … that. When she comes downstairs and says “I need to take some food boxes over there” it is not because her selflessness knows no bounds, but because her capacity to live with what’s going on around her is close to exhausted.

We’ve become a political party of two on the matter, one of us who just lives with it 24/7, and the other with just enough distance to say “so you’re telling me you were briefly considering whether or not to take yet another stupid opinion on the matter personally?”

vim keybindings and the shell

I’ve gotten so used to Evil mode in Emacs, and so comfortable with stuff like Vimari elsewhere, that I figured I’d give oh-my-zsh’s vim mode a try. That didn’t go well. jeffreytse/zsh-vi-mode promises to make it all better, so I’m giving it a try.

I can’t even explain what wasn’t working about the original plugin I was trying. I just know it was confusing, but that it feels even worse to go back to Emacs/readline keybindings, so I’m just going to give this a try.

YNAB plug

For years I was a Microsoft Money person, even if it meant running Windows in a VM to use it. During our paycheck-to-paycheck years Money’s cash flow forecast tool was like some sort of oracle I could consult about the eventual downstream impact of emergency expenses. I knew it wasn’t the right answer, but it was one of those “bad answer is better than worse alternatives” things that sort of reinforced a bad status quo.

I don’t know how long its been since Money was even a thing – Wikipedia says they killed it in 2009, so okay – but I jumped to Quicken for Mac, which had a similar tool that wasn’t nearly as good, but that was okay, too: Money’s version seemed to take your actuals into account, whereas Quicken’s was a straight “here are your budgeted inflows, here are your budgeted outflows, here’s the difference over time,” with the added ability to sort of amortize a pre-determined monthly variance for unbudgeted expenses.

It was worse because it was less accurate than Money, but it was better because it nudged me in the direction of “oh, maybe I ought to be budgeting for these things instead of guessing them.”

When Mac Quicken 2007 was finally retired and replaced by Quicken for Mac, the tool got even less helpful – they took away the ability to add a variance, which meant that if you didn’t budget it the cash flow forecast tool wasn’t going to consider it. My little hack around that was to go to “allowance” accounts, but all that did was isolate personal and less predictable expenses from the more predictable monthly bills and payments.

Basically, I refused to learn the real lesson of that feature, which was that it wasn’t a proper substitute for real budgeting.

You Need A Budget uses envelope budgeting:

Typically, the person will write the name and average cost per month of a bill on the front of an envelope. Then, either once a month or when the person gets paid, they will put the amount for that bill in cash in the envelope. When the bill is due, the money is taken out to pay for that bill.

This prevents the person from spending the money out of their pocket or bank account, because it is already allocated to the bill.

… it just does it with software: You tell it your budget and what you have in the bank, and then record the draw on each “envelope” as you move through the month.

It has been around for a while, first as a standalone app, then in its newest incarnation as a web app. I tried it in its earlier days, when the developers refused to support downloading transactions to force you to do the work of tracking where your money was going, but there was no way I was going to do that data entry, and I was also pretty hooked on some kind of cash flow forecasting tool.

Last year, though, I decided to give it another spin. They’ve softened up their position on automated transactions because, I imagine, in most households the “envelopes” are now completely metaphorical. My physical currency on hand goes up during camping season, because that’s how you buy firewood at the state parks, but otherwise?

Like any tool built around a particular mindset, the YNAB social experience – its subreddit, support forums, assorted online enclaves – can feel more like a spiritual movement or ideological tendency than a way to budget. I tune all that out. I have enough needless rigor in my life reading Metafilter comments.

The “just enough rigor” part to me comes down to the fact that it makes the envelope metaphor work. The iOS app comes with a little widget that tells you how much money is left in key envelopes. “Can we do this thing?” Well, check the widget: Says there’s $x in the envelope for that kind of thing, which would ordinarily mean “no,” but I see there’s $y in this envelope over here – so is that tradeoff okay? And from that follows an easier time sticking to your goals.

The closest it comes to a forecasting tool is the ability to take anything you already have in the bank, or that is left over in the budget at the end of the month, and pre-budget it in the months ahead.

That’s not to say cash flow forecasting doesn’t have its place.

When I knew a layoff was coming, it was simply not possible to reconcile the YNAB mindset with something that meant a lot to me, which was the ability to go still for a few months and not worry about anything once I finished up. It was important to me as I dealt with the emotional stuff you deal with when you both know you are done somewhere but have committed to going out on a professional note, and then it became critical once I realized I had a health thing to address.

On the other hand, YNAB does such a good job of recording budget information and making it exportable that it was trivial to take my carefully considered envelopes, export them to a spreadsheet, and make a very simple forecasting tool I could build scenarios around. It meant I could calm down a little, come up with a plan, and do the whole “mind like water” thing once I had that plan. At the same time, it wasn’t something I could imagine maintaining because it was a spreadsheet and I am not one of those people.

I’m glad I adopted more rigorous budgeting when I didn’t need to, removing my dependency on a forecasting tool. And I’m glad that the careful work I did making that change made it easier to build something I would not want to maintain daily, but was able to use to look 3, 6, 9 or 12 months ahead given my situation.

This is a weird thing to write about.

Years ago a friend of mine shared some personal stuff around money. I don’t think they meant to share as much as they did, and then after doing so felt circumspect about it, trying to sort out whether they’d done a TMI thing, or maybe revealed a defect they shouldn’t have, or had opened themselves up to judgment. I mean, not from me. All due respect to my fellow veterans, I was not one of the ones who woke up in an army barracks one morning because of his great life choices.

And because even since then – even after a crash course in making the best of a bad decision – I had the good/bad fortune of not having to figure this stuff out for a long while: A life privileged and easy enough that I got to concentrate on other life skills. I think since getting laid off that I have said of the experience “I guess my luck finally ran out” exactly once, and it was a self-evidently foolish thing to say the second it escaped my lips. My political commitments preclude much belief in luck, and my spiritual commitments do not include it in their reckoning.

But still … it’s a weird thing to write about. An uncomfortable thing. US culture is messed up about money and deeply infused with an ideological commitment to the moral virtues of precarity. Even the kindest, least judgmental people vibrate around this topic, because most of us live an existence of gauging whether there are yet rungs below us – a duplex with overflowing sewage owned by an indifferent slum lord from some other state – or just the abyss.

You would hope, in a society that has so thoroughly ravaged its own safety net, that we’d recognize that deep unease and transcend our alienation and atomization, even if not to try to put some of it back right now after due consideration of the horrors we see just stepping out onto a downtown sidewalk. Even if only to say to each other, “we are collectively worthy of more kindness and care than we are choosing to extend to ourselves,” whether that’s on a civilizational scale – where our actual priorities include taking away tents and tarps on the coldest week of the year, or simply being the bluest state with the worst mental health services – or a personal scale, where friends apologize to each other for bringing up money because everybody’s so anxious about it.

So, one of the aspirations I have for my writing is to be helpful. YNAB helped me understand how to budget and plan, and while I would not say it is for everybody, or that everybody needs it, it is definitely for me and I definitely did need it. If you’re uneasy about money, or not sure you’re doing it right, their content marketing is pretty top notch: Even if you don’t buy their product, they do a great job of articulating a particular approach to money and budgeting that might be helpful.

It has been supplanted by org-mode.

Final verdict: It’s pretty good. If I had to put all my to-do lists in a proprietary, closed-format tool, it would probably be the one, or at least in the running with Things. I just appear to be due for an org-mode kick, and TickTick happened to be standing around when it happened.

In terms of getting out of org-mode what I was getting out of TickTick beyond a simple todo list, it came down to mobile, habits, and pomodoros.

[beorg][] handles Reminders integration. Doom Emacs lets you pull in org-habits out of the box, so I just did that and set up a habits file. They’re pulled into my agenda, which shows streaks information for them.

I also added org-pomodoro, which works about like you’d expect: Pick an item from the agenda, trigger the timer, and it adds a time entry to that item in its home file for each Pomodoro completed:

** omg.lol Docs

*** Git workflow

**** DONE Beginner section
DEADLINE: <2023-03-24 Fri>
:LOGBOOK:
CLOCK: [2023-03-24 Fri 14:19]--[2023-03-24 Fri 14:44] =>  0:25
CLOCK: [2023-03-24 Fri 13:49]--[2023-03-24 Fri 14:14] =>  0:25
CLOCK: [2023-03-24 Fri 11:25]--[2023-03-24 Fri 11:50] =>  0:25
CLOCK: [2023-03-24 Fri 10:14]--[2023-03-24 Fri 10:39] =>  0:25
CLOCK: [2023-03-24 Fri 09:38]--[2023-03-24 Fri 10:03] =>  0:25
:END:

I found this presentation helpful for figuring out how to add reporting with the time data you can gather.

Always coming home

At some point we have to acknowledge to ourselves that maybe the chaos is the actual pattern, or that the things we think are chaotic are not after all, even if we can’t quite feel the rhythm.

When it comes to tools – especially productivity ones – I am fickle. I’ve been through four or five major “all in on org” moments in my life, and then I’ve fallen out of them.

org-mode can be tough to stick with: Emacs can be crabby, the ecosystem feels fragile sometimes, and you occasionally go through these periods where everyone’s carrying on about some new Emacs build or hotness and you chase after it ten minutes before a day full of meetings where the segfaults start in the middle of a note.

No amount of fluency with the tool or joy in the format can get you around the days where the tool just doesn’t feel steady.

On the other hand, one missing piece from previous Big Org periods has been a habit I picked up during the pandemic of being more circumspect about the tools I pick up and what I want to do with them. I write about what I hope to get out of something. I think about what I want to use it for. Sometimes I learn about a new feature or approach and decide to roll it in, but I’ve gotten a lot better at knowing why I use the things I do, and what I want out of them. I’m also more suspicious of changing them up much.

In assorted parts of my technical life, that new habit has been a real boon, because “tech” in the broadest sense is something I like to play with, and I used to constantly break the rule “Don’t try anything new on race day,” with all the ensuing chaos that comes with it. Like, years ago past Mike got really disgusted at work and swore he was done with his job and applied for a bunch of jobs, and decided that was the right time to change email providers because he heard that the one he was switching to had some cool features. Yes, some mails went missing.

Where my Emacs life is concerned, it just comes down to “one thing at a time, only one thing a day, if at all.”

I self-interpret that to mean “you can turn on one module and snarf up one quick tweak or imported config to go with it.” Adding org-super-agenda? Great – turn it on and add one config change. Doing pomodoros? Okay, don’t change your org agendas to show the timing, just make sure you can record the times consistently, and that it isn’t making things feel wobbly. I usually just do a pomodoro of fiddling at the beginning of the day, which is about enough time to find something, turn it on, and see if it passes the initial use test before it’s time to do other stuff.

I’ve been harping a lot elsewhere on how stable Doom Emacs feels compared to my own init.el ecosystem, but I suspect part of it is just that I’ve been very careful about what I add and how much I add at a time. I think Emacs is in a unique tier of sensitivity to a lot of thrashy change, but just about anything built on a lot of user-generated content (e.g. modules, add-ons, plugins) can be made to misbehave and become hard to troubleshoot.

Another thing that has changed about me over the past few years has been a growing awareness that I can get into a headspace where I prize automation and less overall motion to the point of paralysis. It took seeing other people falling into that pit – believing that they came into work that day to automate something and not to achieve an outcome – to get me to see it in myself and snap out of it a little.

Vim Adventures

I mentioned yesterday that I’ve been enjoying Vim Adventures:

“Enjoying” is sort of relative. It wants you to learn stuff that is a slog, and it has a few puzzle challenges that can be challenging since they not only demand you use the keys you’ve been learning to accomplish the task, they sometimes require you to do a little lateral thinking. You can’t get away with just memorizing a keystroke for a given lesson, use the keystroke a few times, then get on with it. You end up using the keystroke repeatedly trying to solve a puzzle in under n keystrokes, sometimes not realizing there’s another lesson buried in there about cursor motion behavior or what have you.

In the end, though, the puzzles are well constructed. New stuff comes in at a steady pace, and each new set of challenges provide a mix of what you’ve already learned with the new stuff, so you have to constantly adapt. I tried another, less gamified vim tutorial to see if it would work better for me and it was disorienting after a few hours with Vim Adventures: I couldn’t see how any of it could possibly stick in any meaningful timeframe.

I’m typing this in the space between writing pomodoros this morning. This is the first day I’ve had all week to be deep in a writing/revision cycle using what I’ve been learning. It’s been interesting to realize how much stuff Vim Adventures has taught me to “just do” with minimal delay, but I can still sense the occasional spike in cognitive load when a few decades of muscle memory collide with newly learned things that are still up there in the thinking layers.

It’s also interesting, as someone whose first deep Unix experience dates back to an Ultrix box and VT100 in 1991, how much these assorted UI ecosystems have come to acknowledge each other. Doom understands the basic Mac text keybindings, so when I’ve exhausted my vi keybinding knowledge and just need to do a damn thing to a block of text, cmd-x and cmd-v are right there.

I have asked myself a few times “why are you doing this at all,” and as near as I can figure it comes down to:

• Finding I prefer hand positioning that involves fewer control key chords.
• Enjoying the learning experience. It’s stimulating.
• A recent encounter with the mythical “vi only” system in the form of my EdgeRouter X. Dinking around with files in that thing was a pain, just as the vi bigots have always warned. Completely doable with my limited repertoire of vi skills, but clumsy.

I suppose I should also sing Doom Emacs’ praises once more: It does a lot to make Emacs feel more stable by taking care of housekeeping and providing some reasonable defaults. It’s a good learning platform because there are fewer random things going on out of the box. I don’t know if I’d be able to tolerate the learning curve with my own init.el in place.

Probably time to wind down on micro.blog

A few months ago I took this site and moved it from Jekyll to Hugo. Jekyll was fine and all, but the more I dug into Hugo’s features the more it seemed like an interesting direction to go. I also wanted to move the site from looking personal marketing heavy to just being a blog that also has stuff in it about work/business sometimes.

That was a liberating move: I was able to get myself into a pretty easy writing/publishing workflow that was perhaps a little more computer-bound than what I could get away with on micro.blog, but not so bad.

I kept sticking to my micro.blog presence partially because it has robust cross-posting features I hadn’t bothered to suss out using something like IFTTT or Zapier. At the same time, I have a little too much web producer left in me to be comfortable with two domains that are roughly doing the same thing. I once automated a consolidation, content migration, and re-canonicalization of four dozen websites. People sneer about SEO and all the evils of the SEO industry, but there are search engines in the world and there are reasons to care about what they make of your web presence.

If all I wanted to do was have a blog in the most traditional “reverse-chronological-ordered posts” sense of the word, I’d probably stick with micro.blog. Its Hugo foundations allow just enough flexibility to do the basics, and your content can be exported with relatively few idiosyncracies – no worse than any other Hugo theme. You can write shortcodes and customize a lot of the way the site works.

It’s kind of a frustrating platform for iterative development, though. There’s enough variance between a basic Hugo theme and what micro.blog needs that you have to do some trial-and-error. Using vanilla Hugo on a laptop, iterative learning is lightning fast and the feedback loops border on instantaneous, especially with hugo server --navigateToChanged. With micro.blog, there’s a build time delay to figure out if your thing worked or not, or to get feedback on how it is failing.

I have also found I prefer the free-wheeling nature of Mastodon a little more appealing than the social layer of micro.blog. I’ve written about that elsewhere, so we’ll leave it alone here.

Anyhow, it’s time to go through the past couple of years of micro.blogging and work out how to manage that migration. One idea I want to explore is the creation of a “microblog” content type that would allow me to just move all that stuff into its own silo largely untouched, with some design work to cope with the titleless posts over the years, and with some logic in my Atom feed to allow for ongoing microposts that stay out of Atom but still get syndicated over Mastodon. Similarly, that could be a quick path to getting all my old dot unplanned content moved over under its own “vintage” content type.

At some point I am not going to have the kind of time on my hands that I have now, so my thinking is beginning to shift to “what’s expedient” vs. “what could enhance your career as a Hugo consultant with yourself as your only client.”